The problem this guide is about
A working definition of in-store marketing, the structural reason most of it stops paying at the door, and the loop that closes the gap.
1. What is in-store marketing?
In-store marketing is every deliberate action a retailer takes inside a physical store to influence what a shopper notices, chooses, buys and remembers. It covers merchandising, signage, in-store promotions, product sampling, staff interaction, checkout marketing, and enrolment into a loyalty program — and, in its modern form, it also covers what happens after the shopper leaves.
In-store marketing
The set of marketing activities executed inside a physical retail environment to drive conversion, basket size and return visits. It is distinguished from advertising by its location — the shopper is already present and already in a buying frame of mind — and from ecommerce merchandising by its physicality: shelf position, lighting, adjacency, staff, queue design and packaging are all levers.
Why it matters
The short version is that this is where the money is. According to the US Census Bureau's Quarterly Retail E-Commerce Sales release for the second quarter of 2026 — published on 18 August 2026 — e-commerce accounted for 17.1% of total US retail sales on a seasonally adjusted basis ($340.2 billion of $1,986.5 billion). By arithmetic, 82.9% of retail sales were not ordered online.
The Census figure measures where the order was placed, not where the customer stood. Buy-online-pickup-in-store and curbside orders are counted inside the 17.1% even though the shopper physically visits the store. That makes 82.9% a floor on store-involved sales, not a ceiling. It also excludes food services and drinking places entirely, because the series covers retail trade only. Both caveats point the same way: the physical store is involved in more retail than the headline e-commerce share suggests.
Source: US Census Bureau, Quarterly Retail E-Commerce Sales, Q2 2026.
The second reason it matters is attention. A shopper inside a store has already spent the scarcest resource in marketing — they have travelled, parked, walked in, and set aside time. No digital channel starts from that position. The cost of getting a message in front of someone who is standing in front of your shelf is effectively zero, and it reaches a person already in a buying frame of mind — something no paid placement can guarantee.
The third reason is the one this guide is really about, and it is a warning rather than a boast: that attention is not stored anywhere. Unless the visit ends with the customer identified and reachable, all of it evaporates at the door.
The disciplines inside "in-store marketing"
The phrase covers at least seven distinct disciplines that are often collapsed into one budget line. Separating them is useful, because they fail differently and are measured differently.
| Discipline | What it is | Primary job | Measured by | Persists after the visit? |
|---|---|---|---|---|
| Merchandising | Product placement, adjacency, planograms, shelf height, cross-merchandising, hero displays | Increase units per transaction and margin mix | Sales per linear metre, attachment rate | No |
| Signage & wayfinding | Window graphics, aisle markers, shelf-edge tickets, price communication, category navigation | Reduce friction and communicate value | Category conversion, dwell time | No |
| In-store promotions | Multibuys, markdowns, bundles, sampling, demonstrations, seasonal events | Drive short-term volume and clearance | Promotional uplift, sell-through, margin impact | No |
| Staff-led selling | Greeting, needs discovery, recommendation, fitting, add-on suggestion | Convert browsers and raise basket value | Conversion rate, units per transaction, mystery shop | Only through memory |
| Checkout marketing | Queue-line merchandising, impulse fixtures, receipt messaging, add-on prompts, enrolment ask | Capture the last decision — and the customer's identity | Impulse attachment, enrolment rate | Yes, if it enrols |
| Loyalty & membership | Cards, points, stamps, tiers, member pricing, VIP access | Convert a transaction into an identified, repeatable relationship | Enrolment, repeat visit rate, member revenue share | Yes |
| Post-purchase engagement | Wallet updates, reward notifications, replenishment reminders, win-back offers | Bring the customer back before a competitor does | Return rate, redemption rate, reactivation | Yes — that is the point |
Read the final column. Five of the seven disciplines produce no residue. They work, they work well, and they work only while the shopper is inside the building. The two that persist — loyalty and post-purchase engagement — are the ones that most independent retailers under-invest in, usually because the tooling has historically been expensive, app-dependent, or tied to a point-of-sale system they did not want to replace. That is no longer true, and the rest of this guide is about what changed.
Examples of in-store marketing
- A window display built around a single hero product, with the same product replicated at the till.
- Shelf-edge tickets that show a member price alongside the regular price.
- A sampling table where the sample is paired with a same-day-only offer.
- A queue-line fixture stocked with items under a fixed price point.
- A staff greeting script that ends with a specific product question rather than "can I help?"
- A QR code at the counter that adds a loyalty card to the customer's phone before they have put their wallet away.
- A receipt line that says how many stamps the customer now has and what the next one earns.
- An in-store event — a launch, a stylist evening, a repair clinic — that gives regulars a reason to come in when they were not going to buy.
Every item in that list except the last three is spend that ends when the shopper leaves. The last three are spend that creates an asset. Both are legitimate. But a retailer with a fixed budget who never shifts any of it from the first category to the second is buying the same customer again, forever, at full price.
What in-store marketing is not
Two boundaries are worth drawing, because blurring them is how retail budgets get misallocated.
It is not advertising. Advertising's job is to get someone to the door. In-store marketing's job starts after they arrive. They are measured differently, they are bought differently, and a store with excellent advertising and poor in-store execution converts expensive traffic into browsing.
It is not the same as a loyalty program — but a loyalty program is the part of it that survives the visit. Treating loyalty as a separate marketing programme, run by a different person, on a different budget line, from the rest of in-store marketing is one of the most common structural mistakes in independent retail. The enrolment happens at the counter. It is in-store marketing.
- In-store marketing is everything a retailer does inside the store to influence notice, choice, purchase and memory.
- 82.9% of US retail sales in Q2 2026 were not ordered online (US Census Bureau) — and that is a floor, not a ceiling, on store involvement.
- Seven disciplines sit inside the phrase; five of them stop working at the door.
- Loyalty and post-purchase engagement are the only two that create a reusable asset.
Budgeting in-store marketing entirely as promotional spend. Promotions buy volume this week; identification buys the ability to influence next month. A store that never spends on the second is permanently renting its own customers.
2. Why retailers lose customers after the purchase
Because nothing about a normal retail transaction records who the customer was. Cash and card payments identify an account, not a person you are allowed to market to. Without identification at the counter, a store cannot tell a first visit from a fiftieth, cannot notice when a regular stops coming, and cannot reach anyone deliberately. The relationship does not break — it was never created.
Walk the journey as it actually happens in an unidentified store:
Figure 1 — Where the retail customer relationship leaks. The default path ends in a competitor's till. The alternative path branches at one single point: whether the customer was identified at checkout. (Original PushNotice diagram.)
The five reasons the leak persists
1. Payment is not identification. A card transaction tells your acquirer who paid. It does not give you a marketing relationship, a name you may use, or consent to contact anyone. Retailers routinely assume that because money moved, a customer record exists. It does not.
2. The ask is uncomfortable and badly designed. "Can I take your email?" at a busy counter is a poor experience for both parties. It slows the queue, the customer suspects spam, and the staff member — reasonably — stops asking after the third refusal. Enrolment friction is almost always a design failure, not a customer-attitude problem.
3. Plastic and paper cards get lost. A punch card lives in a drawer, a coat pocket or a bin. When it is not present at the moment of purchase, the loyalty step silently does not happen, and neither the customer nor the retailer notices the program quietly dying.
4. Apps ask for too much, too early. Asking someone to download an app, create an account and verify an email in exchange for a future 10% discount inverts the value exchange. The cost is paid now, in friction; the benefit arrives later, in theory.
5. Nobody is watching for absence. This is the quiet one. Most retailers can tell you what they sold last month. Very few can tell you which regulars stopped coming. Churn in physical retail is invisible by default — customers do not cancel, they simply stop appearing — and you cannot notice an absence in a population you never identified.
The usual objection to digital loyalty — "my customers aren't like that" — is worth testing against the data. Pew Research Center's Mobile Fact Sheet reports that 91% of US adults own a smartphone and 98% own a cellphone of some kind, based on its National Public Opinion Reference Survey of 5,022 US adults fielded February–June 2025. On payments, the Federal Reserve Bank of Atlanta's Survey and Diary of Consumer Payment Choice — 2025 edition, published May 2026, based on a three-day diary completed by 6,079 respondents — reports that more than 72% of consumers used a mobile phone to pay at least once in the twelve months to October 2025. Note the definitional caveat: "paid with a phone" there includes person-to-person apps and in-app purchases, not only tap-to-pay at a terminal. The same research programme reports that purchases "remain predominantly an in-person activity, with more than three-quarters of purchases made at the in-person point of sale." Both figures are quoted from the report PDF linked in Sources; confirm the current edition before reusing them.
Sources: Pew Research Center; Federal Reserve Bank of Atlanta, SDCPC.
What the leak actually costs — and how to state it honestly
It is tempting at this point to quote the familiar line that retaining a customer is five times cheaper than acquiring one. We are not going to, because we could not find a primary source for it, and neither can anyone else. The most rigorous treatment we found is in Loyalty Myths (Keiningham, Vavra, Aksoy and Wallard, published by Ipsos Loyalty in 2005), which traces the claim to unpublished late-1980s work by the Technical Assistance Research Project, notes it spread through secondary citation, and concludes that "there is currently enough contrary information to bury or significantly qualify this truism."
What is traceable is narrower and older: Frederick Reichheld and W. Earl Sasser Jr.'s 1990 Harvard Business Review article "Zero Defections: Quality Comes to Services," summarised by Bain as finding that companies "can boost profits by almost 100% by retaining just 5% more of their customers." That is a 36-year-old, services-focused finding built around a single case study, and the methodology sits behind a paywall we could not read. We cite it as an idea with a name attached, not as a benchmark for your store.
The retention and loyalty category is unusually polluted with statistics that cannot be traced to any study. If a number in a marketing article does not name a publisher, a sample and a period, treat it as decoration. We have applied that test to ourselves throughout this guide: every figure here names its source and the date we checked it, and where we could not verify something we say so instead of substituting an estimate.
The honest framing does not need a statistic anyway. It is arithmetic. If a customer who visits you twelve times a year stops after four, you have not lost one sale — you have lost eight, plus every sale in the years that followed, plus the referrals. And because you never identified them, you will not find out. That is the cost, and it is invisible on every report you currently look at.
- Payment is not identification. A card transaction creates no marketing relationship.
- The leak has one hinge: whether the customer is identified at the counter. Everything else is downstream.
- Physical-retail churn is invisible — customers stop appearing rather than cancelling.
- The "5× cheaper to retain" claim has no traceable primary source. Do not build a business case on it.
3. The PushNotice Retail Retention Loop™
The Retail Retention Loop™ describes the seven steps that turn a store visit into a repeatable relationship: Attract, Identify, Reward, Save, Remind, Return, Repeat. Traditional in-store marketing is very good at step one and usually stops at step two. The loop only closes if the customer leaves carrying something you can update.
1. Attract — get the shopper into the store. Advertising, location, window, reputation, local search, events. Cost per unit is high and rises. This is the step everyone already funds.
2. Identify — link the transaction to a person, with consent. A scanned pass, a phone-number lookup, an online order match. This is the hinge of the entire loop. Skip it and steps 3–7 are unavailable at any budget.
3. Reward — give the identification a reason to happen. Stamps, points, member pricing, a welcome offer, status. The reward is what you exchange for identity; make sure the exchange is obviously fair from the customer's side.
4. Save — put something durable on the customer's device. A wallet pass that carries balance, tier and a scannable code, stored on the phone, working offline, updatable by you afterwards.
5. Remind — reach them at a moment that is useful to them: reward unlocked, offer expiring, replenishment due, new arrival in a category they buy. Rationed, not broadcast.
6. Return — the visit that would not otherwise have happened. This is the only output the loop is judged on.
7. Repeat — each loop should cost less than the last, because you now know more and the customer already carries the card. Attraction is a purchase; retention is compounding.
Figure 2 — The PushNotice Retail Retention Loop™. Seven stages; one hinge. Most retail marketing budgets fund stage 1 generously and stage 2 not at all. (Original PushNotice framework diagram.)
How the loop connects physical retail to digital retention
The loop is not a metaphor — each arrow is a concrete handoff between a physical action and a digital one, and each handoff is where programs break in practice.
| Stage | Who does it | Physical or digital | Typical failure | The fix |
|---|---|---|---|---|
| 1 · Attract | Marketing, location, reputation | Physical | Traffic is bought but never converted into identity | Attach an enrolment ask to the highest-traffic moment you already have |
| 2 · Identify | Counter staff + the customer's phone | Both | The ask is slow, awkward, or quietly abandoned by staff | One scan the customer initiates; nothing typed by staff |
| 3 · Reward | The loyalty rules | Digital | The reward is too distant to motivate the exchange | Divide the threshold by real visit frequency; if the wait exceeds a few months, redesign |
| 4 · Save | The customer, in one tap | Digital | An app install is required and most people decline | Wallet pass — no download, no account, works offline |
| 5 · Remind | The retailer | Digital | Broadcasting everything to everyone until the pass is deleted | Segment by recency and category; send only what would be missed |
| 6 · Return | The customer | Physical | Nobody measures it, so nobody can improve it | Track repeat visit rate for members against a matched non-member group |
| 7 · Repeat | The system | Both | Every cycle is run as a fresh campaign from zero | Let the card persist; the second loop should cost far less than the first |
In-store marketing should not stop when the customer leaves the store. Wallet-based loyalty gives retailers a persistent customer touchpoint that can carry rewards, offers, membership status and timely messages beyond the shop floor.
A homewares shop puts a QR code on the card machine. A customer buying a lamp scans it while their card is still in the reader, saves a wallet card in one tap, and the counter screen shows they are a member. Three weeks later, the shop sends one message to customers who bought lighting and have not returned: a note that the new season's shades have landed and members see them first. The customer walks past on Saturday, the card is already on their phone, and they come in. Nothing here required an app, a data science team, or a new till system. This is an illustrative example of the mechanic, not a description of a specific customer or a claimed result.
- Seven stages: Attract, Identify, Reward, Save, Remind, Return, Repeat.
- Identify is the hinge — every later stage is unavailable without it.
- Each stage has one characteristic failure mode; most of them are operational, not technical.
- The measure of the loop is stage 6, the return visit. Not passes issued.
Before you evaluate any software, draw this loop for your own store on one page and write a real sentence under each stage. In our experience most retailers finish with four stages written in detail and stage 2 left blank. That blank is the project.
Channels, cards and mechanics
An honest comparison of every in-store marketing channel against wallet loyalty, what actually belongs on a retail pass, and how to choose a reward mechanic that fits how your customers really shop.
4. Traditional in-store marketing vs wallet-based loyalty
Signage, printed coupons, email, SMS, apps, point-of-sale loyalty and wallet loyalty each win on different dimensions. Signage wins on immediate, unpersonalised reach inside the store. Email wins on length and cost per message. SMS wins on raw urgency. Apps win on depth. Wallet loyalty wins on persistence — it is the only channel that is simultaneously always present on the device, updatable by the retailer, usable offline, and free of an install. It is not a universal replacement for any of the others.
The comparison below is deliberately unflattering to wallet in several rows, because the honest case for it is narrower and stronger than the marketing case. Read the "best at" column rather than counting ticks.
| Dimension | In-store signage | Printed coupons | SMS | Retail mobile app | POS-native loyalty | Wallet loyalty | |
|---|---|---|---|---|---|---|---|
| Visibility to the customer | ✓ Unmissable — in store | ⚠ Only if kept | ⚠ In a crowded inbox | ✓ High | ✕ Only if opened | ✕ Invisible between visits | ✓ In the wallet they open to pay |
| Friction to join / receive | ✓ None | ✓ None | ⚠ Type an address | ⚠ Give a number | ✕ Download, register, verify | ⚠ Give details at the till | ✓ One tap, no account |
| Persistence after the visit | ✕ Zero | ⚠ Until lost | ⚠ Until buried | ⚠ Until scrolled past | ⚠ Until deleted | ✕ Lives in the till | ✓ Stays on the device |
| Updatable after delivery | ✕ Reprint | ✕ Reprint | ✕ Sent is sent | ✕ Sent is sent | ✓ Yes | ⚠ In the till only | ✓ Yes — the card itself changes |
| Personalisation | ✕ None | ✕ Rarely | ✓ Strong | ⚠ Basic | ✓ Strong | ⚠ Depends on vendor | ✓ Per-customer card and balance |
| Identifies the customer | ✕ No | ⚠ Only by code | ✓ Yes | ✓ Yes | ✓ Yes | ✓ Yes | ✓ Yes — unique code per member |
| Works offline / no signal | ✓ Yes | ✓ Yes | ✕ No | ⚠ Once received | ⚠ Partially | ⚠ Depends | ✓ Yes — pass is on the device |
| Notification capability | ✕ None | ✕ None | ⚠ Inbox only | ✓ Direct | ✓ Full push | ✕ Usually none | ⚠ Lock-screen, platform-capped |
| Message length / richness | ⚠ A few words | ⚠ One offer | ✓ Unlimited | ✕ Very short | ✓ Unlimited | ✕ N/A | ✕ Short by design |
| Marginal cost per message | ⚠ Print + labour | ✓ Near zero | ✕ Per-message fee | ✓ Near zero | ✓ N/A | ⚠ Near zero, but rationed | |
| Setup complexity | ✓ Low | ✓ Low | ⚠ Medium | ⚠ Medium + compliance | ✕ High and recurring | ⚠ Tied to your till | ✓ Low — configuration |
| Consent burden | ✓ None | ✓ None | ⚠ Marketing consent | ✕ Strictest rules | ⚠ Push consent | ⚠ Data consent | ⚠ Consent to enrol and to message |
| Retention potential | ✕ None | ⚠ One cycle | ✓ High | ⚠ High but abrasive | ✓ Very high, small audience | ⚠ Medium | ✓ High across a large audience |
| Best at | Converting shoppers already in front of the product | Driving one specific, immediate action | Depth, storytelling, catalogue, receipts | Genuine urgency and time-critical alerts | Ordering, payment, browsing, accounts | Automatic earning at the till | Being present, current and identified between visits |
Which channel for which situation
The useful question is not "which channel is best" but "which job is this?" Four jobs cover most retail marketing, and each has a clear answer.
| The job | Best channel | Second choice | Why |
|---|---|---|---|
| Sell more to someone standing in the store | Signage and merchandising | Staff recommendation | They are already present; the message costs nothing to deliver |
| Capture who they are | Wallet enrolment at checkout | Email capture at the till | One tap beats typing an address into a queue |
| Tell them a reward is ready | Wallet pass update | The card is the message; the balance change is the notification | |
| Announce something time-critical today | SMS | Wallet message | SMS is intrusive, which is the point — but budget and consent limit it |
| Tell a longer story — lookbook, guide, brand | App or web | Wallet passes cannot carry length and should not try | |
| Give a member a reason to feel recognised | Wallet tier and member pricing | Staff at the counter | Status has to be visible to work, and the card is where it lives |
| Win back someone who has stopped coming | Wallet message plus email | SMS if consented | Use two channels only for the customers who are actually gone |
If you need to send a 900-word seasonal lookbook, email wins and it is not close. If you need to tell 4,000 people that the store is closed today because of a burst pipe, SMS wins. If your customers order and pay in an app every week, the app is the product and loyalty is a feature of it — building a separate wallet program alongside it can dilute rather than add. Wallet's advantage is persistence, not breadth, and a channel strategy that pretends otherwise will disappoint.
You will see email open rates quoted confidently in retail marketing articles. Treat them with care. Mailchimp's own published benchmarks — drawn from campaigns sent to at least 1,000 subscribers across its customer base, last updated December 2023 — report an all-industry average open rate of 35.63% and an e-commerce open rate of 29.81% with a 1.74% click rate. Mailchimp itself warns that open-rate accuracy "may be impacted by Apple's privacy changes and their Mail Privacy Protection feature," which auto-opens messages and inflates the measure. Click rate is the defensible metric; open rate is directional at best. We could not find any SMS engagement benchmark from a named publisher that discloses both sample period and message volume, so no SMS figure is quoted anywhere in this guide.
Source: Mailchimp Email Marketing Benchmarks.
- Each channel wins on a different dimension; wallet's dimension is persistence.
- Wallet is the only channel that is present, current, identified and install-free at the same time.
- It is weak at length, and it is rationed on notifications. Design around both facts.
- Pick by job, not by channel preference — most retailers need three of these running together.
5. What should a retail loyalty pass contain?
A retail loyalty pass should contain the member's identifier and a scannable barcode, the one number they care about (stamps, points or tier), the reward that number is heading towards, and the minimum store information needed to use it. Everything else belongs on the back of the pass or nowhere. The exact fields available depend on the platform and on your loyalty program.
The most common design mistake is treating the pass as a brochure. It is not: it is a card that will be glanced at for two seconds in a queue, and then scanned. Its job is to answer one question instantly — where am I up to? — and to present a code.
Figure 3 — Retail wallet pass anatomy. Seven front-of-card elements, everything else on the back. Field names and availability differ between Apple Wallet and Google Wallet — see Section 8. (Original PushNotice diagram.)
Component by component
| Component | Purpose | Needed? | Practical guidance |
|---|---|---|---|
| Member identifier | Links every visit to one record | Always | Use a pseudonymous ID. It is often the only personal identifier you need to hold. |
| Barcode or QR code | The scan that identifies the customer at the till | Always | QR is the safest default for most retail counters — but confirm your own scanner is 2D-capable, because 1D-only laser scanners cannot read it. |
| Customer name | Recognition and a warmer counter interaction | Optional | Nice, not necessary. Make it an optional field at enrolment, never a required one. |
| Points balance | The core progress signal for spend-based programs | If you run points | Make it the largest element. One number, no ambiguity about units. |
| Stamps earned / remaining | Progress signal for visit-based programs | If you run stamps | Show both earned and remaining; "7 of 10" outperforms "7". |
| Tier / status | Recognition, and the benefit of belonging | If tiers exist | Do not launch tiers with too few members — an empty top tier is not aspirational. |
| Next reward / threshold | Converts a balance into motivation | Strongly recommended | The single highest-value field on most retail cards. Show the gap, not just the total. |
| Active coupon or offer | Carries a specific, redeemable promotion | Campaign-dependent | Consider a separate coupon pass so redeeming it does not disturb the loyalty card. |
| Expiry date | Urgency, and control of open-ended liability | If offers expire | Disclose at enrolment. Warn before it bites, or it converts loyalty into irritation. |
| Store information | Address, hours, phone — usefulness beyond the program | Recommended | Back of the pass. It quietly makes the card worth keeping. |
| Customer service contact | A route to a human when something is wrong | Recommended | Back of pass. Costs nothing; prevents the worst support experiences. |
| Terms and privacy notice | Consent, transparency and compliance | Always | Back of the pass, in plain language, with a stated way to leave the program. |
| Artwork / brand imagery | Recognition in a list of a dozen cards | Recommended | Contrast matters more than beauty. Test the card in dark mode and in sunlight. |
Design the card for the queue, not for the pitch deck. Hold your phone at arm's length in a bright shop and ask three questions: can I tell whose card this is, can I read my balance, and does the code scan first time? If any answer is no, the design is wrong regardless of how it looks in a mock-up.
Putting the terms and conditions on the front of the pass and the balance on the back. It happens more often than you would think, usually because a legal review lands after the design is finished. The front is for the two-second glance; nothing else belongs there.
- Front of card: brand, the one number, the gap to the next reward, and a code that scans.
- Back of card: terms, expiry, store details, service contact, privacy, how to leave.
- A pseudonymous member ID is usually the only identifier you need to hold.
- Show progress as a gap ("3 to go"), not just as a total.
6. Retail loyalty mechanics
Choose the simplest mechanic that matches how your customers actually buy. Stamps suit frequent purchases of similar value. Points suit variable baskets. Tiers suit high-value, low-frequency relationships. Birthday, referral, product-specific and limited-time offers are layers you add to a working base mechanic — never substitutes for one.
The ten mechanics
Points
Customers earn a fixed number of points per unit of currency spent and redeem them against a reward or a discount.
- Best use case: retail with wide basket variation, where a £5 purchase and a £150 purchase should not earn the same.
- Advantages: proportional and intuitively fair; flexible reward catalogue; produces a spend signal you can segment on.
- Disadvantages: requires access to transaction amounts; customers do arithmetic and notice when the exchange rate is poor; accrued points are an open liability on your margin.
- Ideal retailer: fashion, homewares, beauty, electronics, general merchandise.
- Example: 1 point per £1; 500 points redeems £15. (Illustrative parameters.)
Stamps
A fixed number of qualifying purchases earns one reward — the digital descendant of the punch card.
- Best use case: frequent, similarly priced purchases where the reward can arrive within weeks.
- Advantages: immediately understandable; no transaction data required — a visit is enough; visually satisfying progress.
- Disadvantages: over-rewards small baskets if your ticket varies; needs an anti-abuse cap such as one stamp per customer per day.
- Ideal retailer: bakery, coffee counter, convenience, pet supplies, florist, refill shops.
- Example: buy 9, get the 10th free, one stamp per day maximum. (Illustrative.)
Tiers
Cumulative spend or visits unlock status levels that carry standing benefits.
- Best use case: when the top decile of customers is worth materially more than the median and is worth protecting explicitly.
- Advantages: recognition motivates high-value customers more than discounts do; benefits can be non-discount (early access, free alterations, personal appointments), which protects margin.
- Disadvantages: meaningless below a few hundred active members; demotion is emotionally costly and must be handled carefully; complexity at the counter.
- Ideal retailer: apparel, luxury, jewellery, specialist sports, beauty.
- Example: Silver at £250 annual spend, Gold at £750, with Gold receiving first access to sale. (Illustrative.)
Spend-based rewards
A reward triggers at a cumulative spend threshold, without a points currency in between.
- Best use case: retailers who want spend proportionality without explaining an exchange rate.
- Advantages: simpler than points; no currency to devalue; easy to communicate ("£10 back at every £200").
- Disadvantages: less flexible than a points catalogue; large thresholds can feel unreachable to occasional shoppers.
- Ideal retailer: homewares, garden centres, hardware, general merchandise.
Visit-based rewards
Rewards accrue per visit rather than per pound.
- Best use case: when frequency is the behaviour you actually want to change, and basket size is roughly stable.
- Advantages: requires no transaction data at all; drives the habit rather than the spike.
- Disadvantages: a customer can game it with minimal purchases unless you set a qualifying threshold.
- Ideal retailer: convenience, health food, refill, pet, hobby stores.
Product-specific rewards
Only nominated products or categories earn, or the reward is a specific product.
- Best use case: shifting a category, protecting margin on another, or launching a range.
- Advantages: precise merchandising control; ties loyalty to buying decisions you care about.
- Disadvantages: needs item-level data, which carries real privacy weight; harder for staff to explain; easy to confuse customers.
- Ideal retailer: beauty, grocery, specialist retail with strong own-brand.
VIP / paid membership
Customers pay a fee for ongoing benefits — free delivery, member pricing, exclusive access, services.
- Best use case: where the benefit is genuinely valuable, repeatable and cheap for you to deliver.
- Advantages: revenue up front; paid members behave measurably differently because they are committed; strong retention.
- Disadvantages: you must deliver the benefit every single time; refund and cancellation obligations; a bad month damages more than a free program would.
- Ideal retailer: specialist retail with services, premium grocery, sports and fitness retail.
Birthday rewards
A time-limited offer around the customer's birthday.
- Best use case: almost any retailer with a gifting or treat dimension.
- Advantages: high relevance for a very low send volume; feels personal without requiring personalisation infrastructure.
- Disadvantages: needs a date at enrolment, which adds a field; useless if you cannot deliver it in a timely way.
- Guidance: collect day and month only. You almost never need the year.
Referral rewards
An existing member earns when they bring a new customer who enrols or buys.
- Best use case: socially shareable products with margin that can support two rewards.
- Advantages: acquisition at a known cost, from a trusted source.
- Disadvantages: attribution is genuinely difficult in a physical store and invites gaming; you must define what counts as a successful referral before launch.
Limited-time offers
A promotion with an explicit deadline, delivered to identified customers.
- Best use case: clearance, quiet trading periods, seasonal pushes, and reactivating lapsed members.
- Advantages: urgency works; the deadline gives a clean measurement window; you can target only the customers who need the nudge.
- Disadvantages: trains customers to wait for offers if overused; erodes full-price sales; the single fastest way to exhaust a notification channel.
Plot your store on two axes and read off the base mechanic. Horizontal: purchase frequency — how often a typical customer buys from you. Vertical: basket variability — how much the value of each purchase differs.
High frequency · low variability → Stamps. Bakery, convenience, coffee counter, refill. The reward arrives often enough to feel real, and no transaction data is required.
High frequency · high variability → Points. Grocery, pharmacy, pet, garden. Stamps would over-reward the small basket and insult the large one.
Low frequency · low variability → Visit rewards on a long horizon, or a membership. Specialist services retail. The reward must be worth the wait, or the card is a filing cabinet.
Low frequency · high variability → Tiers and recognition. Furniture, jewellery, electronics, luxury. Accumulation is meaningless at one purchase a year; status and service are not.
Figure 4 — The Retail Loyalty Mechanic Matrix™. Frequency and basket variability determine the base mechanic; birthday, referral, product-specific and limited-time offers are layers on top of it. (Original PushNotice framework diagram.)
| Mechanic | Data required | Ideal retailer | Watch out for |
|---|---|---|---|
| Points | Transaction amount | Fashion, homewares, beauty, general merchandise | Devalued exchange rates; unredeemed liability |
| Stamps | A visit only | Bakery, convenience, coffee, florist | Small baskets earning the same as large ones |
| Tiers | Cumulative spend or visits | Apparel, luxury, sports, beauty | Launching with too few members to fill the top tier |
| Spend thresholds | Cumulative amount | Homewares, hardware, garden | Thresholds that occasional shoppers can never reach |
| Visit rewards | Visit date | Convenience, refill, hobby | Token purchases gaming the count |
| Product-specific | Item or category | Beauty, grocery, own-brand retail | Privacy weight of item-level data; staff confusion |
| VIP / paid membership | Subscription status | Premium grocery, sports, specialist | A benefit you cannot deliver every time |
| Birthday | Day and month | Almost any retailer | Collecting the birth year without needing it |
| Referral | Attribution link | Socially shareable categories | Attribution gaming; double reward cost |
| Limited-time offer | Segment + deadline | Any retailer, sparingly | Training customers never to pay full price |
Three design rules that survive contact with a real shop floor
1. Divide the reward threshold by real visit frequency. A ten-stamp card in a shop visited fortnightly pays out in five months. Ask honestly whether anyone would wait that long for what you are offering. If not, lower the threshold or change the mechanic.
2. A member of staff must be able to explain it in one sentence, while serving. Complexity in the rules is paid for in staff time, at every transaction, forever — and it is paid first by your newest and least confident employee.
3. Decide the expiry policy before launch, and disclose it at enrolment. Unexpired balances accumulate as an open-ended liability against your margin. Expiry is legitimate and common; introducing it retroactively is what turns a loyal customer into a complaint.
- Frequency and basket variability determine the base mechanic. Everything else is a layer.
- Stamps and visit rewards need no transaction data — a scan at the counter is sufficient.
- Tiers need scale to work; launching them too early makes the top tier invisible.
- Set expiry at launch and disclose it. Retrofitting expiry is a trust event.
Launching points, tiers, a birthday reward, a referral scheme and a paid VIP club simultaneously. Every additional mechanic multiplies the explanation burden at the counter and divides staff attention. Launch one mechanic, get it working, then layer.
7. Twenty in-store marketing ideas for retail stores
The ideas that repay the effort are the ones that either capture a customer's identity or give an identified customer a specific reason to return. Below are twenty, each with its objective, the retailer it suits, the mechanic, a wallet implementation, the notification it justifies, and the one KPI that tells you whether it worked.
Every example figure below is a hypothetical parameter chosen to make the mechanic concrete. None of it describes a specific customer, and none of it is a claimed result. Substitute your own margins, thresholds and visit cycles before using any of it. Where an idea depends on a platform capability, Section 8 sets out what is actually documented and Section 10 sets out what is not.
1New customer welcome reward
- Objective
- Convert an anonymous first-time buyer into an identified member at the moment of highest goodwill.
- Best retailer
- Any retailer with plausible repeat purchase — strongest in fashion, beauty and specialist retail.
- How it works
- At checkout, the customer is offered an immediate small benefit in exchange for saving the loyalty card. The exchange is explicit: identity for value, right now, not later.
- Example
- "Save our card and get 10% off today" — applied to the transaction in progress, so the customer sees the benefit before they leave.
- Wallet implementation
- Counter QR code links to the pass; the welcome benefit is pre-loaded on the card so staff apply it by scanning rather than by remembering a rule.
- Notification opportunity
- None immediately. Let the first message be genuinely useful, several days later.
- Primary KPI
- Wallet save rate — passes saved as a share of transactions where the ask was made.
2Checkout reward moment
- Objective
- Use the highest-attention seconds in the store — while the customer waits for a receipt — to advance the relationship.
- Best retailer
- High-footfall retail with a queue: convenience, bakery, pharmacy, gift.
- How it works
- Staff scan the customer's card, the balance visibly changes, and the customer is told where they now stand. The recognition is the marketing.
- Example
- "That's your seventh — three more and your next one's on us."
- Wallet implementation
- The pass updates in seconds and the new balance is visible on the customer's own screen before they leave the counter.
- Notification opportunity
- The balance change itself. On both platforms this is the lowest-cost, highest-relevance message you have.
- Primary KPI
- Scan rate — the share of member transactions where the card was actually presented.
3Digital coupon
- Objective
- Drive a specific, measurable action without printing anything.
- Best retailer
- Any retailer running promotions; particularly strong in grocery, beauty and homewares.
- How it works
- A coupon is issued to a defined segment as its own pass, with a stated deadline and clear terms. Redemption is recorded by scanning it.
- Example
- "£5 off a £30 spend, this week only" sent to members who have not visited in 30 days.
- Wallet implementation
- Both platforms provide a dedicated coupon or offer pass type, separate from the loyalty card, so redeeming it does not disturb the member's balance.
- Notification opportunity
- Issue, then one reminder before expiry. Two messages maximum.
- Primary KPI
- Redemption rate against the segment size.
4Double-points day
- Objective
- Move volume into a specific, usually quiet trading day.
- Best retailer
- Points-based programs in fashion, homewares, garden and general merchandise.
- How it works
- Earning is multiplied for a single day, announced only to members, which makes membership visibly worth having.
- Example
- "Double points this Tuesday" on the slowest day of your trading week.
- Wallet implementation
- A campaign message to segmented members; the multiplier is a rule in the loyalty platform, not something staff calculate.
- Notification opportunity
- One message the day before, one on the morning itself if the audience is small and well targeted.
- Primary KPI
- Member transactions on the target day versus the same weekday, four-week average.
5VIP early access
- Objective
- Reward your best customers with something that costs margin only once — access, not discount.
- Best retailer
- Fashion, footwear, luxury, specialist sports, limited-run product.
- How it works
- Top-tier members shop the sale, the drop or the new season a day or an evening before everyone else.
- Example
- "Gold members shop the sale from 6pm Thursday — everyone else from Friday."
- Wallet implementation
- Tier shown on the card so the entitlement is self-evident at the door; the card is the ticket.
- Notification opportunity
- One message to the tier, timed to the access window, not the day before.
- Primary KPI
- Share of tier members transacting in the access window.
6Birthday reward
- Objective
- Generate a visit on a date that has nothing to do with your trading calendar.
- Best retailer
- Beauty, gifting, fashion, food retail, anything with a treat dimension.
- How it works
- A time-boxed benefit arrives near the customer's birthday, redeemable in store for a defined window.
- Example
- "A gift from us this month — 20% off anything, any time in your birthday month."
- Wallet implementation
- A coupon pass issued automatically; the loyalty card stays untouched.
- Notification opportunity
- One at issue, one three days before it expires. Nothing else.
- Primary KPI
- Birthday redemption rate, and incremental spend on the redeeming visit.
7Product-category reward
- Objective
- Shift a specific category — a launch, an own-brand line, an overstock — without a blanket discount.
- Best retailer
- Beauty, grocery, pet, hardware, retailers with a meaningful own-brand.
- How it works
- Nominated categories earn a multiplier, or a reward is unlocked by buying from the category.
- Example
- "Triple points on skincare this month."
- Wallet implementation
- Requires category-level data from the till, so confirm that field is available before designing the campaign around it.
- Notification opportunity
- One message, targeted only to members who have bought that category before or an adjacent one.
- Primary KPI
- Category units sold to members versus the prior period.
8Seasonal promotion
- Objective
- Capture demand that already exists in the calendar rather than trying to create it.
- Best retailer
- Every retailer, with the season varying by category.
- How it works
- A themed offer, event or bundle runs for a defined period and is announced to members first.
- Example
- A back-to-school bundle in a stationery and gift shop, with members getting first pick.
- Wallet implementation
- Update the pass artwork for the season so the card itself signals the campaign; issue a coupon pass for the offer.
- Notification opportunity
- One at launch, one mid-run only if there is genuinely new information.
- Primary KPI
- Member share of seasonal revenue.
9Limited-time offer
- Objective
- Compress demand into a short window when you need volume now.
- Best retailer
- Any retailer, used sparingly.
- How it works
- A clearly deadlined offer goes to a defined segment, with the deadline stated on the pass itself.
- Example
- "48 hours only — £10 off £50."
- Wallet implementation
- Coupon pass carrying the expiry date; the card visibly stops being valid when the window closes.
- Notification opportunity
- One at launch, one in the final hours. Resist a third.
- Primary KPI
- Redemption rate and, critically, the effect on the following week's full-price sales.
10Bounce-back offer
- Objective
- Turn one visit into two — the single most efficient retail mechanic there is.
- Best retailer
- Fashion, homewares, beauty, gift, garden.
- How it works
- At the point of purchase the customer receives an offer that can only be redeemed on a separate, later visit within a defined window.
- Example
- "£10 off your next visit, valid from tomorrow for 21 days."
- Wallet implementation
- Issued as a coupon pass at checkout, dated so it cannot be used on the current transaction.
- Notification opportunity
- One reminder around three days before expiry — this is one of the highest-relevance messages in retail.
- Primary KPI
- Second-visit rate within the offer window.
11Spend threshold reward
- Objective
- Raise average transaction value at the decision point.
- Best retailer
- Homewares, fashion, hardware, garden, general merchandise.
- How it works
- A benefit unlocks at a basket value slightly above your current average — close enough to be reachable, far enough to matter.
- Example
- "Spend £50 today, get a £10 credit on your card." Set the threshold from your actual average basket, not from a round number.
- Wallet implementation
- Credit applied to the loyalty balance so the pass updates immediately and the customer sees it.
- Notification opportunity
- The balance update. No campaign message needed.
- Primary KPI
- Average transaction value for members versus non-members over the period.
12Visit-frequency reward
- Objective
- Change the habit rather than the basket.
- Best retailer
- Convenience, refill, health food, pet, hobby.
- How it works
- A reward triggers at a number of visits inside a defined period, which rewards rhythm rather than accumulation.
- Example
- "Four visits in a month unlocks a fifth-visit reward."
- Wallet implementation
- Visit counter on the card; requires only a scan, no transaction data at all.
- Notification opportunity
- A nudge when a member is one visit short with a few days of the window left.
- Primary KPI
- Visits per active member per month.
13Referral reward
- Objective
- Acquire new customers from the people who already like you.
- Best retailer
- Specialist, boutique, community-facing retail with a shareable proposition.
- How it works
- An existing member shares a link or code; both parties receive a benefit when the new customer enrols and buys.
- Example
- "Give a friend £10, get £10 when they spend £30."
- Wallet implementation
- A share link on the back of the pass; the referred customer's enrolment carries the referrer's ID.
- Notification opportunity
- One message when the referral completes — this one is genuinely welcome.
- Primary KPI
- Referred enrolments that go on to make a second purchase.
14Tier upgrade campaign
- Objective
- Move mid-value customers into a higher tier, where behaviour tends to be stickier.
- Best retailer
- Apparel, beauty, sports, specialist retail with an established tiered program.
- How it works
- Members close to a threshold are told precisely how close, and given a limited window in which the gap is easier to close.
- Example
- "You're £40 from Gold — spend it before the end of the month and you'll keep Gold for a full year."
- Wallet implementation
- Tier and distance-to-next-tier both visible on the card; the update itself carries the message.
- Notification opportunity
- One message, only to members inside a defined proximity band. Never broadcast this.
- Primary KPI
- Upgrade rate among the targeted band, and 90-day retention of the upgraded cohort.
15Event-based offer
- Objective
- Give people a reason to come in when they were not planning to buy anything.
- Best retailer
- Boutique, beauty, sports, hobby, bookshops, garden centres.
- How it works
- An in-store event — a launch, a clinic, a demonstration, an evening — is offered to members first, with capacity as the scarcity.
- Example
- "Members-only styling evening, 20 places, Thursday 7pm."
- Wallet implementation
- An event pass carrying the date, time and a scannable code for the door.
- Notification opportunity
- Invitation, then one reminder on the day. Attendance reminders are among the most tolerated messages you can send.
- Primary KPI
- Attendance rate, and spend from attendees within seven days.
16Store-opening or relocation promotion
- Objective
- Build an identified customer base from day one instead of counting anonymous opening-week traffic.
- Best retailer
- Any new site; especially valuable for multi-location retailers adding a location.
- How it works
- The opening offer is delivered through enrolment rather than through a printed flyer, so the traffic spike leaves a residue.
- Example
- "Founding member: save the card this week and keep double points for three months."
- Wallet implementation
- A distinct enrolment code for the new store so you can measure it separately and message it locally later.
- Notification opportunity
- A welcome message, then nothing for two weeks.
- Primary KPI
- Enrolments in the opening fortnight, and how many of them return within 60 days.
17Clearance incentive
- Objective
- Clear terminal stock without publicly discounting the brand.
- Best retailer
- Fashion, footwear, seasonal goods, homewares.
- How it works
- Clearance is offered to members privately, before or instead of a public markdown.
- Example
- "Members' clearance: 40% off end-of-line, in store only, until Sunday."
- Wallet implementation
- A coupon pass to a targeted segment; presenting the pass is what unlocks the price.
- Notification opportunity
- One message. Clearance repeated too often becomes your pricing strategy.
- Primary KPI
- Sell-through of the clearance lines and realised margin.
18Local customer offer
- Objective
- Concentrate spend among the people who can realistically come back often.
- Best retailer
- Independents, neighbourhood retail, and any store in a mixed tourist/resident area.
- How it works
- Members whose behaviour indicates they are local — frequency, weekday visits, home store — receive benefits that are worthless to a one-time visitor.
- Example
- "Neighbour rate: 10% off every Wednesday for members with a home store here."
- Wallet implementation
- Segment by home store and visit frequency. Prefer behavioural signals over device location — they are more reliable, and they carry far less privacy weight (see Section 10).
- Notification opportunity
- Monthly at most, and only to the segment.
- Primary KPI
- Weekday visit frequency among the local segment.
19Win-back campaign
- Objective
- Reach customers who have stopped coming, while they can still be reached.
- Best retailer
- Every retailer with an identified customer base — this is the mechanic that most justifies having one.
- How it works
- Members who have not transacted in a period meaningfully longer than their own normal cycle receive a single, generous, deadlined offer.
- Example
- "We've missed you — £15 off anything over £40, for the next 14 days."
- Wallet implementation
- Coupon pass with a clear expiry; the pass is already on their phone, which is precisely the advantage over email here.
- Notification opportunity
- One message, one reminder. If neither works, reduce contact frequency for that member rather than escalating.
- Primary KPI
- Reactivation rate — lapsed members who transact within the window.
20Post-purchase reward and care
- Objective
- Make the days after a purchase part of the marketing, not a silence.
- Best retailer
- Considered-purchase retail: furniture, electronics, beauty, footwear, sports equipment.
- How it works
- A useful follow-up — care instructions, a fitting appointment, a complementary accessory, a warranty registration — arrives at a sensible interval, with a small reward attached.
- Example
- Ten days after a leather-goods purchase: care guidance and a member price on the care kit.
- Wallet implementation
- Store the purchase category against the member record and time the follow-up to the product, not to your campaign calendar.
- Notification opportunity
- One, timed to when the customer is likely to be using the product.
- Primary KPI
- Attachment rate on the follow-up, and 90-day repeat purchase rate.
Read them back and a structure appears. Ideas 1, 2 and 16 buy identity. Ideas 3–9, 11 and 17 buy a transaction. Ideas 10, 12, 14, 19 and 20 buy a return visit. Ideas 5, 13, 15 and 18 buy belonging. A balanced retail calendar contains all four types; most stores run almost exclusively from the second group, which is why their marketing feels like a treadmill.
- Ideas that capture identity or trigger a return visit compound; pure promotions do not.
- The bounce-back offer is the highest-leverage single mechanic in this list.
- Nearly every idea justifies one or two messages — never a stream.
- Each idea needs one named KPI before it launches, or it cannot be judged afterwards.
The platform layer, verified
What Apple Wallet and Google Wallet actually do for retail loyalty, what the notification channel really allows, and an honest account of geofencing — including the one place where the two platforms genuinely differ.
8. Apple Wallet and Google Wallet for retail loyalty
Both Apple Wallet and Google Wallet can hold a retail loyalty card that carries a member identifier, a scannable barcode, a balance and a tier, and both can be updated remotely so the customer's card changes without them doing anything. Neither wallet knows your till exists. Something must identify the customer at checkout, and something must push the updated balance back — that connective work is yours or your platform's, not the wallet's.
Apple Wallet and Google Wallet are containers for passes. They do not integrate with your point-of-sale system, they do not know what you sold, and they do not award anything automatically. Every working wallet loyalty program supplies two things itself: (a) identification at the counter — a barcode scan, a lookup, or an NFC read on a certified terminal — and (b) synchronisation, updating the pass after the balance changes. Any claim that "Apple Wallet integrates with your POS" is either shorthand for that architecture or simply wrong.
What a wallet pass is
A wallet pass is a small, structured file that a customer saves to the wallet app on their phone. It contains fields of text, artwork, and a barcode. Once saved, it is stored on the device, which is why it works with no signal — and, crucially, it remains addressable by the issuer, so the balance printed on it today can be different tomorrow without the customer lifting a finger.
For a retailer, three properties matter more than any feature list. It requires no app install: Apple Wallet ships on every iPhone, and Google Wallet is available on most Android phones. It persists: unlike an email, it does not scroll away. And it is updatable: the card is a live object, not a snapshot.
How customers save one
- Scan a QR code at the counter — the highest-converting route, because the customer is present, engaged, and has just bought something.
- Tap a link in an email, an order confirmation, a text or a social post.
- A button on your website — most useful when it sits on the order confirmation page, not buried in a footer.
- From a receipt — a printed QR on the receipt captures the customers who did not want to stop at the counter.
On Google's side, the mechanism for a web save link is a signed JSON Web Token behind a pay.google.com/gp/v/save/ URL. Google's documentation gives a hard practical limit worth knowing at design time: "The safe length of an encoded JWT is 1800 characters… If the length is over 1800 characters, the save may not work due to truncation by web browsers."
Capability comparison — from the platforms' own documentation
| Capability | Apple Wallet | Google Wallet |
|---|---|---|
| Loyalty card type | The storeCard pass style — Apple's developer guide describes it as appropriate for "store loyalty cards, discount cards, points cards, and gift cards" | The Loyalty API's loyaltyClass (shared template) and loyaltyObject (one per customer) |
| Coupon / offer type | The coupon pass style — "appropriate for coupons, special offers, and other discounts" | offerClass / offerObject; the class carries a redemptionChannel of INSTORE, ONLINE, BOTH or TEMPORARY_PRICE_REDUCTION |
| Member identifier | Pass fields plus the value encoded in the barcode | accountId and accountName on the object, plus the barcode value |
| Points balance | Custom fields on the pass | Native loyaltyPoints, plus secondaryLoyaltyPoints "shown in addition to the primary loyalty points" |
| Tier / status | Custom fields | rewardsTier and secondaryRewardsTier — note these sit on the class, not the object |
| Barcode formats | QR, PDF417, Aztec and Code 128, with Code 39, Codabar, EAN-13 and ITF also listed in the current reference. The 1D additions are recent — check Apple's reference for the OS versions they require before designing around them | Barcode on the object; NFC read via Smart Tap where certified |
| Remote updates | Yes — a pass web service plus a push to registered devices | Yes — the object is updated server-side through the REST API |
| Update mechanism | Push carries "an empty JSON dictionary for the payload"; the device then fetches the updated pass from your web service | Direct UPDATE or PATCH on the object; class changes "propagate immediately across all Passes Object instances that reference it" |
| Lock-screen change alert | Only on fields carrying a changeMessage: "You need to provide a value for the system to show a change notification" | Via addmessage with TEXT_AND_NOTIFY, or an update with notifyPreference set where the changed field is allow-listed |
| Fields that can trigger an update notification | Any field with a change message | An explicit allow-list of five: rewardsTier, secondaryRewardsTier and programName on the class; loyaltyPoints.balance and secondaryLoyaltyPoints.balance on the object |
| Published notification cap | None published | "You may send a maximum of 3 messages that trigger a push notification in a 24 hour period"; maximum of 10 messages stored on an object |
| Location relevance | Up to 10 locations and up to 10 beacon UUIDs. Apple: relevance "is passive… It doesn't present alerts or post notifications" | merchantLocations — "up to 10 locations per class and 10 per object". Google: "When a user is within a set radius of this lat/long, and dwells there, Google will trigger a notification" |
| NFC identification | Value Added Services: an NFC certificate from Apple, a VAS-certified terminal, and POS software supporting "VAS Only" and "Payment and VAS" modes | Smart Tap: "You must be certified in order to use this protocol," an 8-digit collector ID, a key exchange, enableSmartTap on the class and a redemption value on the object |
| Expiry | Pass-level expiry and relevance settings | Object state plus validity settings on the class |
| App required? | No — Wallet is pre-installed on iPhone | No — Google Wallet is available on most Android phones |
| Documented platform coverage | iOS, watchOS (with some field exceptions) | Google's documentation describes passes on Android-powered devices and does not document iOS support |
First, we do not state that Google Wallet does not work on iPhone. We checked Google's loyalty overview, its FAQs and its web-issuing documentation, and none of them makes an explicit statement either way; what they consistently describe is Android. The accurate sentence is the one in the table: Google's documentation describes Android and does not document iOS support. Second, we do not quote a numeric Apple notification limit, because Apple publishes none. The correct phrasing is "Apple publishes no cap," not "there is no cap." Apple's own guidance is qualitative: send pushes "only when the pass has changed. Don't send unnecessary pushes."
How the two update models differ — and why it changes your campaign design
Apple hosts nothing for you. Your platform runs a pass web service; devices register against it; when a balance changes you send a content-free push, and the device comes and collects the new pass. The visible notification is therefore generated from the pass itself — from a field carrying a change message. There is a well-documented trap for anyone building this: Apple states that "a push notification for a pass update works only in the production environment," which catches a large number of teams during testing.
Google holds the object on its own servers. You call the REST API and the change is live for that holder with no per-device handshake. Because class changes propagate to every object referencing them, a single edit to a class touches every card you have ever issued from it — powerful, and worth respecting. Notifications are explicitly rationed: three push-triggering messages per 24 hours, and only five loyalty fields are allow-listed to trigger an update notification at all.
On both platforms, the balance update is the message. You do not need to write a campaign to tell somebody they earned points — the card changing on their phone already told them, in the most relevant possible way, with no send budget consumed. Reserve deliberate broadcasts for the things a balance cannot say: a reward about to expire, a new collection, an event, a store closure. This reframing removes a large part of most planned send calendars, because the balance change has already delivered the news. We publish no figure for the reduction — run the exercise against your own plan and see what survives.
What is new, and what to check yourself
Apple's "What's new in Wallet" page currently lists several items relevant to retail — including a "Poster Generic" template Apple describes as being for "loyalty, rewards, membership, or gift cards," interactive tiles called Featured Actions, a Pass Designer application for macOS, and a server-side Pass Builder package — with OS availability noted per item. We are not restating version numbers or per-feature limits here, because platform release detail dates quickly and the only reliable version is Apple's own page on the day you read it. The same applies to Google's release notes. Check both before you commit a design to a capability.
- Both wallets carry a member ID, barcode, balance, tier and remote updates, with no app install.
- Neither wallet talks to your till. Identification and synchronisation are your responsibility.
- Apple pushes an empty payload and the device fetches; Google updates the object server-side.
- Google publishes a hard cap of three push-triggering messages per 24 hours and allow-lists just five loyalty fields; Apple publishes no numeric cap.
- Barcodes work everywhere. NFC identification requires certification on both platforms.
Designing a tap-to-earn experience before checking terminal certification. Apple's VAS requires an NFC certificate, a VAS-certified reader and POS software supporting VAS modes; Google's Smart Tap requires certification, a collector ID and a key exchange. A QR code requires a scanner you almost certainly already own, and it works on every phone in the queue.
9. Push notifications for retail
Wallet notifications work best for a narrow set of retail moments: a reward earned, a reward about to expire, a tier change, a genuinely new collection, an event, and a win-back. They work badly for anything a customer would describe as "another promotion." Both platforms ration the channel deliberately, and Google publishes a hard limit of three push-triggering messages in any 24-hour period.
What notifications can legitimately support
| Message type | Why it is welcome (or not) | Trigger | Sensible ceiling |
|---|---|---|---|
| Reward earned | Confirms something the customer just did; pure good news | Balance change at checkout | Every qualifying visit — it is the update, not a campaign |
| Reward about to expire | Prevents a loss the customer would resent | Days before expiry | Once, occasionally twice |
| Tier achieved | Recognition; the reason tiers exist | Threshold crossed | Once per change |
| Loyalty milestone | "One more to go" is a service message as much as a marketing one | Proximity to a threshold | Once per cycle |
| Event or appointment reminder | Requested and time-critical | Booking, then day-of | Twice per event |
| New collection or restock | Welcome only to customers who buy that category | Product launch, segmented | Monthly at most, segmented |
| Seasonal campaign | Tolerated when infrequent and genuinely seasonal | Calendar | A handful per year |
| VIP / member-only access | Signals the value of membership | Tier + access window | Per access window |
| Win-back | Welcome once; corrosive if repeated at someone who has left | Lapse threshold | Once, one reminder, then stop |
| General promotion to everyone | The message most likely to cause pass deletion | — | Avoid |
Notification fatigue is a real cost with a visible symptom
Fatigue is not a vague concern; it has a specific, measurable consequence in wallet marketing. Email fatigue produces silence — the customer simply stops opening, and your list quietly rots while the metrics look survivable. Wallet fatigue produces deletion. The customer removes the pass, and you lose the channel permanently and irreversibly. There is no re-engagement campaign for a deleted pass; they have to re-enrol from scratch.
That asymmetry should govern the whole strategy. In email, an unnecessary send costs you a fraction of an open rate. In wallet, an unnecessary send risks the asset itself. Both platforms appear to have designed with this in mind — Google by publishing an explicit cap, Apple by tying notifications to changed fields with an explicit change message rather than to arbitrary broadcasts.
Right Customer + Right Message + Right Time + Right Context = Better Customer Experience
Right Customer — the person for whom this is actually true. Not "everyone with a pass." A message about a men's footwear restock sent to your whole base is wrong for most of the people who receive it, no matter how well written it is.
Right Message — one idea, one action, in a sentence a person can absorb on a lock screen. If it needs a paragraph, it is an email.
Right Time — aligned to the customer's cycle, not to your calendar. A replenishment reminder at week ten of a twelve-week cycle is useful; the same message on the first Monday of the month is spam that happens to be well timed for you.
Right Context — what else is happening for them. Do not send a win-back offer to somebody who bought yesterday, or an "expiring soon" alert to somebody who has already redeemed.
The test: if this message did not arrive, would the customer be worse off? If the honest answer is no, do not send it. Applied strictly, this single question eliminates most retail sends — and the ones that survive it are the ones a customer would notice not receiving.
A practical send policy
| Rule | Default | Reasoning |
|---|---|---|
| Transactional updates (balance, tier) | Unlimited — they follow real events | The customer caused them; they are confirmations, not marketing |
| Deliberate broadcasts | 2–4 per member per month, maximum | Beyond this, deletion risk rises faster than incremental revenue |
| Same-day repeats | Never | Google's documented cap makes this a technical limit as well as a taste one |
| Segmentation floor | Never send to 100% of the base | If a message is right for everyone, it is probably too generic to act on |
| Quiet hours | No sends outside trading hours | A retail message at 11pm cannot be acted on and is remembered badly |
| Recency suppression | Suppress anyone who transacted in the last 48 hours | Prevents the "we miss you" message landing on yesterday's customer |
| Post-send review | Check pass deletions after every campaign | Deletion is the true cost metric of this channel and almost nobody watches it |
Weak: "🎉 HUGE SALE NOW ON! Up to 50% off everything in store this weekend — don't miss out!!" Sent to every member. It is about you, it is untargeted, it contains no fact specific to the recipient, and it exhausts the channel for the week.
Better: "Your £8 reward expires Sunday." Sent only to members holding an unredeemed reward inside seven days of expiry. It is about them, it is true only for them, and it is the kind of message a customer would be annoyed not to receive. (Illustrative wording.)
- Balance and tier updates are the channel's best content and cost you nothing.
- Wallet fatigue produces deletion, not silence — an irreversible loss of the channel.
- Google publishes three push-triggering messages per 24 hours; treat it as an architectural constraint.
- Apply the test: if the message did not arrive, would the customer be worse off?
- Watch pass deletions after every campaign. It is the metric that tells the truth.
Treating the wallet as a cheaper SMS list. Its unit cost is indeed near zero — but unit cost is the wrong variable. It is a small, permission-shaped surface where the currency is relevance and the penalty for abuse is permanent removal.
10. Geofencing and location-based retail marketing
Geofencing means triggering an action when a device enters, dwells in or leaves a defined area. In wallet marketing specifically, the two platforms behave differently and this is widely misreported: Apple documents pass relevance as passive lock-screen surfacing that "doesn't present alerts or post notifications," while Google's MerchantLocation reference states that "when a user is within a set radius of this lat/long, and dwells there, Google will trigger a notification" — with Google controlling the radius, the dwell time and the wording, and the user having granted precise, always-on location to the Google Wallet app.
This is the section of the article where the most confident claims in the category tend to fall apart, so it is worth being unusually precise about who does what.
The four things people mean by "geofencing"
| Capability | What it really is | Who provides it | What it requires |
|---|---|---|---|
| Apple Wallet pass relevance | The pass surfaces on the lock screen when the customer is near a listed location. Apple: relevance "is passive… It doesn't present alerts or post notifications." | Apple, from the pass's locations array | Up to 10 locations per pass; Apple interprets store cards, coupons and generic passes with a small radius, "on the order of a hundred meters or closer" |
| Google Wallet geo notification | A real notification when the user enters and dwells within a radius of a merchant location | Google, from merchantLocations | Up to 10 per class and 10 per object; the user must have "enabled notifications and granted precise, always on location access to the Google Wallet app"; Google sets radius, dwell and the notification text |
| App-based geofencing | Your own app registers geofences with the OS and reacts when they are crossed | Your app, using the platform's location APIs | An installed app, the user's location permission — and on Android, background location, which since Android 11 cannot be granted from the app's own dialog |
| Third-party location advertising | Ad targeting based on location data collected from other apps and exchanges | Ad platforms and data brokers | Media budget — and a category with a significant and growing enforcement record (see below) |
This is the single most commonly misstated fact in wallet marketing, in both directions. Apple's Wallet Developer Guide is explicit: "Relevance information is passive—it helps users find passes when they need them by putting relevant passes right on the lock screen. It doesn't present alerts or post notifications." Google's documentation is equally explicit in the other direction: its MerchantLocation reference states that "when a user is within a set radius of this lat/long, and dwells there, Google will trigger a notification. When a user exits this radius, the notification will be hidden," and its loyalty push-notification guide adds that "Google decides how close a user needs to be and how long they need to stay in the area before the notification is sent. Google also controls the text of the notification." One further trap: the older locations field on the loyalty object is deprecated, and carries the note "This field is currently not supported to trigger geo notifications" — the working field is merchantLocations, which Google's reference describes as replacing "the deprecated LatLongPoints." Anyone citing the deprecated field will conclude, wrongly, that Google cannot do this at all.
Sources: Apple Wallet Passes documentation; Google Wallet MerchantLocation reference.
How geofencing works, conceptually
Three things have to be true for a location trigger to fire. The device must know where it is, to sufficient accuracy. Something must be watching a defined boundary. And the user must have granted a permission that allows that watching to happen — often, in the background.
Each of those is a real constraint, and the published numbers are sobering. GPS.gov states that "GPS-enabled smartphones are typically accurate to within a 4.9 m (16 ft.) radius under open sky" and that "their accuracy worsens near buildings, bridges, and trees" — that is, precisely where shops are. Google's own geofencing documentation recommends that "the minimum radius of the geofence should be set between 100 - 150 meters," notes that "when Wi-Fi is available location accuracy is usually between 20 - 50 meters," and warns that "if Wi-Fi is turned off, your application might never get geofence alerts." It also documents an app limit of 100 geofences per app per device user, typical latency "less than 2 minutes," and the fact that geofences must be re-registered after a device reboot, an app reinstall, or a data clear.
A 100–150 metre minimum radius in a town centre or a shopping mall does not describe your shop. It describes your shop, the two either side, the street outside and quite possibly a competitor. Location triggering is a proximity signal, not a doorway sensor, and any campaign that depends on knowing the customer is inside your specific unit is depending on something the technology does not reliably deliver.
When location-based marketing makes sense
- Destination retail with a large catchment — a garden centre, a retail park, an outlet, a flagship — where "you are near" is genuinely informative.
- Multi-location retailers, to surface the right store's card rather than the wrong one.
- Travel-adjacent retail — airports, stations, tourist districts — where the customer's presence really is the event.
- Passive relevance generally. Surfacing the card on the lock screen as someone approaches is genuinely useful and costs the customer nothing. On Apple, this is exactly what relevance is for, and it is the safest form of location use in the whole category.
When it does not
- Dense urban retail, where the radius covers your competitors as reliably as it covers you.
- Small independents where the customer base is local and passes the shop constantly — you will trigger repeatedly and pointlessly.
- Any campaign whose value depends on the message arriving inside the store, given the documented latency and radius floors.
- Anywhere the customer has not clearly understood that location is being used. This is not only an ethical point; it is now an enforcement point.
Privacy, consent and the regulatory record
Location is treated as sensitive by both regulators and platforms, and the enforcement record is not hypothetical.
In January 2024 the FTC issued an order prohibiting data broker X-Mode Social / Outlogic from selling sensitive location data — the agency's first such ban — alleging the company sold precise location data that "could be used to track people's visits to sensitive locations such as medical and reproductive health clinics, places of religious worship and domestic abuse shelters," and that it "failed to ensure these third-party apps obtained informed consumer consent." Days later the FTC announced an order banning InMarket Media from selling precise location data; FTC Chair Lina M. Khan stated that "firms do not have free license to monetize data tracking people's precise location." In December 2024 the FTC took action against Gravy Analytics and Venntel for "unfairly selling sensitive consumer location data" and against Mobilewalla for collecting location data from real-time bidding exchanges "even when it didn't have a winning bid." In May 2026 the FTC settled with Kochava and a subsidiary, prohibiting the sale of sensitive location data absent "a consumer's affirmative express consent." The FTC's own business guidance puts the principle plainly: "'consent' to one use without an explanation of other uses is no consent at all."
Source: FTC press releases and business guidance, 2024–2026.
The legal position in the UK and EU is equally clear on the fundamentals. Location data is named in Article 4(1) of the GDPR among the identifiers that make information personal data. Consent, under Article 4(11), must be "freely given, specific, informed and unambiguous… by a statement or by a clear affirmative action." Separately, the message itself is regulated: the ICO states that you "must have consent to send electronic mail marketing to individual subscribers (unless one of the 'soft opt-ins' apply)," and its definition of electronic mail expressly includes "messages sent using a short message service."
The ICO notes that the PECR rules on location data apply to data processed by a network or service provider, and that this "does not generally include GPS-based location information from smartphones, tablets, sat-navs or other devices, as this data is created and collected independently of the network or service provider." In practice, for a retailer using app or platform geofencing, the governing framework is UK GDPR — a lawful basis under Article 6 — plus PECR for the marketing message itself. Note also that as of the ICO's April 2026 update the UK lists seven lawful bases, not the six most marketing guidance still cites, following the addition of "recognised legitimate interest." Nothing here is legal advice; confirm your position with a qualified adviser.
Frequency control and false positives
Two operational realities finish the case for restraint. False positives are endemic: a customer walking past on the way to the station, a neighbour whose flat is inside your radius, a bus route that runs along your street. Repetition compounds it: your most loyal local customer is also the person who passes your door most often, so a naive geofence punishes exactly the people you most want to keep. Any location campaign needs a hard frequency cap per customer per period — and the honest test is whether the campaign still works with one trigger per customer per month.
- Apple pass relevance is passive lock-screen surfacing and posts no notification; Google's merchant locations do trigger notifications, on Google's terms.
- Google's own guidance recommends a 100–150 m minimum geofence radius — larger than most high-street units.
- Location triggering requires user permission, and on Android background location cannot be granted from an app's own dialog since Android 11.
- Precise location has a substantial and continuing FTC enforcement record; treat it as sensitive by default.
- Behavioural signals — home store, visit frequency, category history — are usually more accurate and far less fraught than device location.
Start with passive relevance, which surfaces the card when it is useful and asks nothing of the customer. Add active location triggers only when you have a genuine catchment reason, a frequency cap, and a plain-English explanation of what you are doing on the enrolment screen. If you cannot write that sentence comfortably, do not run the campaign.
Applying it to a real business
Recommendations by retail category, the omnichannel case, the ladder customers climb, and a measurement framework that survives scrutiny.
11. In-store marketing by retail type
The right program follows the visit pattern, not the sector name. Frequent, low-ticket retail wants stamps and speed at the counter. Considered, high-ticket retail wants recognition, service and post-purchase care. Multi-site retail wants one identity across every store. The table below gives a starting position for eleven retail types — start there, then adjust from your own visit data.
Everything in this section is an illustrative strategy showing how the frameworks apply to a category. None of it describes a named retailer, and none of it contains a measured result. Any figure is a plainly hypothetical parameter for a worked example.
| Retail type | Customer behaviour | Best loyalty mechanic | Best in-store promotion | Wallet strategy | Notification strategy | Primary KPI |
|---|---|---|---|---|---|---|
| Fashion retail | Seasonal peaks, wide basket range, browses before buying | Points with two tiers | Early access to sale and new season | Tier and points on the card; refresh artwork each season | New-season and access messages only; segment by category bought | Member share of full-price revenue |
| Boutiques | Relationship-led, staff know regulars by name | Spend thresholds or a simple tier | Private viewings and stylist evenings | Membership card that makes the existing relationship explicit | Rare, personal, event-led | Repeat visit rate among members |
| Beauty retail | Predictable replenishment cycles, high attachment | Points plus product-category rewards | Sampling paired with a bounce-back offer | Points balance plus an active offer; store the category bought | Replenishment timing beats calendar timing | Repeat purchase within the replenishment window |
| Electronics | Infrequent, high ticket, heavy research beforehand | Membership with service benefits | Warranty registration and setup services at the counter | Membership card carrying warranty and service entitlements | Post-purchase care and accessory timing | Attachment rate and 12-month return rate |
| Home goods | Project-driven bursts then long gaps | Spend thresholds | Threshold rewards set just above average basket | Credit balance visible on the card between projects | Seasonal and project-cycle timed | Average transaction value |
| Grocery | Very frequent, highly variable baskets | Points, with category multipliers | Member pricing on shelf-edge | Points balance updated every visit; speed at the till is everything | Almost entirely balance updates; broadcasts rare | Visits per member per month |
| Specialty retail | Enthusiast customers, deep category knowledge | Tiers plus event access | Clinics, demonstrations, expert evenings | Tier and event passes; the card signals belonging | Event-led and genuinely informative | Event attendance converting to purchase |
| Sports & fitness retail | Goal-driven, seasonal, community-adjacent | Visit rewards or paid membership | Fitting and gait analysis services for members | Membership card usable at events and club nights | Season-start and event timing | Member retention across a season |
| Luxury retail | Very infrequent, very high ticket, service-sensitive | Invitation-based membership; no visible points | Private appointments and preview access | A discreet membership card — status, not a balance | Extremely rare; a promotional tone damages the brand | Annual spend per client |
| Multi-location retail | Customers use two or more sites interchangeably | One balance across all stores | Consistent offers, locally executed | Home store on the card; one identity everywhere | Segment by home store; never send one store's news to all | Share of members transacting at more than one site |
| Franchises | Same brand, independent operators, mixed systems | Central rules, local redemption | Brand-wide campaigns with local opt-in | One brand card honoured at every franchisee | Central governance, local windows | Cross-location redemption rate |
Three of these in more depth
The behaviour: customers visit four to six times a year, spend variably, and the relationship is genuinely personal — staff know names.
Mechanic: two tiers based on twelve-month spend, with the benefit at the top being access and service rather than discount: first pick of new season, a stylist appointment, free alterations.
Wallet implementation: the card carries the tier and the customer's name. There is no points balance, because a number would cheapen a relationship that is already qualitative.
Notification strategy: four to six sends a year — new season, private evening, sale access, and a personal birthday note.
KPI: repeat visit rate among members compared with a matched group of non-members over the same twelve months.
The behaviour: project-driven. A customer buys heavily for three weeks and then not at all for eight months.
Mechanic: spend thresholds rather than points — "£10 credit at every £200" — because it is explainable in one sentence and needs no exchange rate.
Wallet implementation: one card, one balance, valid at all three sites, with the home store shown so the customer knows which one is "theirs."
Notification strategy: segmented by home store. A message about a Northgate event never reaches customers who only shop at Southside.
KPI: the share of members who transact at more than one site — the specific value that a central customer record creates and that a per-store system cannot.
The behaviour: highly predictable. A given product lasts roughly eight to twelve weeks and the customer either reorders or drifts to a competitor.
Mechanic: points, with a category multiplier used tactically to launch ranges.
Wallet implementation: points balance on the card, purchase category stored against the member record so the follow-up can be timed to the product rather than the month.
Notification strategy: one message at roughly week ten of the cycle. This is the single highest-relevance message in beauty retail, and it does not read as marketing.
KPI: repeat purchase rate inside the replenishment window.
- Visit pattern, not sector name, determines the program.
- High-ticket categories should reward with recognition and service, not with a balance.
- Multi-site and franchise retail need one identity across locations — that is the whole point.
- In luxury, a promotional tone costs more than the campaign earns.
12. Omnichannel retail: connecting store and ecommerce
Treat store and online as one customer relationship or you will pay twice to acquire the same person and count them as two. The connective tissue is a single customer identity that works in both places — and a wallet card is unusually good at being that identity, because it is issued online, carried physically, and scannable at a counter.
The strongest argument for omnichannel is not a strategic one, it is an accounting one. A retailer with disconnected systems runs paid acquisition to bring a customer to the website, and separately runs local marketing to bring the same customer into the shop, and books them as two customers with two acquisition costs and two half-histories. Neither record is accurate; both are marketed to as though the other did not exist. The customer, who experiences one brand, receives two conversations.
Figure 5 — The In-Store-to-Home Marketing Framework™. The identity layer is what makes the two entry points one relationship; the wallet card is the object that physically bridges them. (Original PushNotice framework diagram.)
The four crossings, and how a wallet card handles each
| Crossing | What the customer does | What usually breaks | Role of the wallet card |
|---|---|---|---|
| Online → store | Buys online, collects or returns in store | Staff cannot see the online history; the customer explains themselves twice | The card is issued with the order and identifies them at the counter |
| Store → online | Buys in store, later shops the website | The online account has no record of the store purchase | Enrolment at the till creates the identity the website can recognise |
| Store → store | Uses more than one of your locations | Two records, two balances, no view of the real customer | One member ID, one balance, home store as an attribute not a boundary |
| Channel-agnostic browsing | Researches online, buys in store, or the reverse | Attribution is impossible and the store gets blamed for online spend | A scanned card at the counter closes the loop between the two |
Buy online, pick up in store
Click-and-collect is the crossing where identity matters most and is most often wasted. The customer is standing at your counter, having already paid, with a confirmed identity and an active order — and in most implementations the interaction ends with a bag being handed over and nothing else happening. It is the single best enrolment moment a retailer has, and it is almost universally unused.
We looked for a citable figure for buy-online-pickup-in-store volume and could not find one that meets a reasonable standard. The nearest usable data point is Adobe Analytics' 2025 holiday report, which found that 17.1% of online orders used curbside pickup among retailers that offer the service, based on "over 1 trillion visits to U.S. retail sites, 100 million SKUs and 18 product categories." Note carefully what that is and is not: it measures curbside specifically, not in-store collection; its denominator is retailers offering the service, not all online orders; and it covers the holiday period, which overstates the annual rate. Every other BOPIS "statistic" we encountered came from paywalled market-sizing vendors or unattributed listicles, and none is quoted here.
Source: Adobe Analytics, 2025 holiday shopping season.
First-party data is the underlying asset
A unified customer identity is not merely operationally tidier. As third-party tracking continues to erode, a consented list of customers who chose to carry your card is one of the few marketing assets that does not depend on another company's permission. That raises the stakes on a question most retailers never ask: can you export it? An asset you cannot take with you is an asset you are renting.
Put your enrolment link in three places you already own and pay nothing for: the order confirmation page, the order confirmation email, and the collection counter. Between them they cover every customer who has just given you money and is therefore at maximum willingness to accept a card.
- Disconnected channels mean paying twice for one customer and marketing to two half-records.
- The wallet card is a genuine bridge object: issued digitally, carried physically, scanned in store.
- Click-and-collect is the most under-used enrolment moment in retail.
- First-party data is only an asset if you can export it. Test that before you rely on it.
13. The Retail Engagement Ladder™
Customers climb five rungs: Anonymous, Identified, Enrolled, Active, Advocate. Each rung unlocks marketing actions that are impossible on the rung below, and each has one specific move that promotes a customer upwards. Knowing what share of your base sits on each rung tells you more about your marketing than any campaign report.
Rung 1 · Anonymous. They bought. You do not know who. You can do nothing except hope. Promotion move: an enrolment ask at checkout worth accepting.
Rung 2 · Identified. You know a transaction belongs to a person and hold a pseudonymous ID. You can count, but not contact. Promotion move: a reason to accept a card and a consent to message.
Rung 3 · Enrolled. They carry the card and have consented. You can now reach them, segment them, and — importantly — notice their absence. Promotion move: a first reward reachable inside their normal cycle.
Rung 4 · Active. They earn, redeem and return. The program is working. Promotion move: recognition — tier, access, or a benefit they cannot buy.
Rung 5 · Advocate. They refer, they attend, they defend you. Rare and disproportionately valuable. Promotion move: ask them for something — a referral, an opinion, a place at an event.
| Rung | What you know | What you can do | What you cannot do | Measure |
|---|---|---|---|---|
| 1 · Anonymous | A sale happened | Merchandise, price, staff well | Contact, segment, notice churn, measure retention | Transactions with no member attached |
| 2 · Identified | Which person bought, pseudonymously | Count visits, measure frequency, spot lapses | Send anything without marketing consent | Identified transaction share |
| 3 · Enrolled | Identity, consent, a card on their phone | Reward, remind, segment, win back | Assume engagement — a saved card is not an active one | Enrolment rate; passes still installed |
| 4 · Active | A real behavioural history | Personalise timing, offer status, predict cycles | Take them for granted; actives churn quietly too | Active member share; visits per member |
| 5 · Advocate | They act on your behalf | Referral, events, product feedback, community | Scale it — advocacy is earned, not campaigned | Referrals; event attendance |
Count your customers by rung once a quarter. Most independent retailers doing this for the first time find that the large majority of their transactions sit on rung 1 — we publish no figure for that share, because we have no data we could cite for it. Whatever your own number turns out to be, it reframes the marketing question completely: the constraint is not creative, budget or channel. It is that almost nobody has climbed to a rung where marketing is even possible. Fixing the rung-1-to-rung-3 conversion is nearly always worth more than optimising anything happening above it.
- Five rungs: Anonymous, Identified, Enrolled, Active, Advocate.
- Each rung unlocks actions impossible below it; the jump from 1 to 3 is the valuable one.
- A saved pass is rung 3, not rung 4. Do not confuse enrolment with engagement.
- Count the distribution quarterly — it is a better diagnostic than any campaign report.
14. Measuring in-store marketing
Measure three things above all others: repeat visit rate among members, redemption rate on what you issue, and member share of revenue. Everything else is supporting evidence. And capture a baseline before you launch, because without one, every number you produce afterwards is decoration.
The twelve metrics that matter, and when
| Metric | What it answers | Right for | Watch out for |
|---|---|---|---|
| Foot traffic | Are people coming in at all? | Advertising, window, location, events | It measures the top of the funnel only; high traffic hides poor conversion |
| Conversion rate | Do visitors become buyers? | Merchandising, staffing, layout, pricing | Needs a reliable footfall count, which many stores do not have |
| Average order value | How much does a buying visit produce? | Threshold rewards, bundles, attachment | Can be lifted by discounting in ways that reduce margin |
| Repeat visit rate | Do they come back? | The loyalty program itself | Self-selection — members were already your better customers |
| Purchase frequency | How often, per member, per period? | Stamps, visit rewards, replenishment | Seasonality distorts short windows badly |
| Redemption rate | Was the offer actually wanted? | Coupons, bounce-backs, birthday, win-back | Low redemption is usually a targeting failure, not an offer failure |
| Loyalty enrolment rate | Is the counter ask working? | The enrolment moment specifically | Measure per transaction where the ask was made, not per footfall |
| Wallet save rate | Do people who see the prompt save the card? | Enrolment design, QR placement, staff script | A high rate with low scan rate means the card is saved and forgotten |
| Offer engagement | Did the message land? | Individual campaigns | Vanity-adjacent; pair it with redemption or ignore it |
| Customer reactivation | Can you bring lapsed customers back? | Win-back campaigns | Define "lapsed" from your own cycle, not a generic 90 days |
| Customer lifetime value | What is a customer worth over time? | Strategy, acquisition budgets, tier design | Fragile at small sample sizes; use trends, not absolute values |
| Incremental revenue | What would not have happened anyway? | Justifying the program's existence | The hardest of all to measure honestly — see the framework below |
Which metric for which campaign
| Campaign type | Primary metric | Supporting |
|---|---|---|
| Enrolment drive | Wallet save rate | Saves per staff member; saves per till |
| Bounce-back offer | Second-visit rate in the window | Redemption rate; spend on the returning visit |
| Double points / multiplier day | Member transactions on the day vs. baseline weekday | Average transaction value |
| Win-back | Reactivation rate | Second purchase within 60 days of reactivation |
| Tier upgrade push | Upgrade rate in the targeted band | 90-day retention of upgraded members |
| Seasonal campaign | Member share of seasonal revenue | Redemption; new enrolments during the season |
| Event | Attendance rate | Spend by attendees within 7 days |
| Clearance | Sell-through of targeted lines | Realised margin; effect on next-week full price |
| The program overall | Repeat visit rate, members vs. matched non-members | Member revenue share; active member percentage |
Four layers, each answering a different question. Skipping a layer is how programs end up unjustifiable.
Layer 1 · Baseline (before launch). Record three numbers: average visits per customer per month, average transaction value, and the share of revenue from customers you can identify. Without these, nothing below is interpretable.
Layer 2 · Adoption (weeks 1–8). Wallet save rate, scan rate at the till, and passes still installed. This layer answers "is the mechanism working?" — not "is the program working?"
Layer 3 · Behaviour (months 2–6). Repeat visit rate, purchase frequency, redemption rate, average transaction value for members. Compare members with a matched group of non-members over the same period, not with members' own past.
Layer 4 · Value (months 6+). Member share of revenue, reactivation rate, and estimated incremental contribution:
Estimated incremental contribution =
(Δ visits × contribution per visit)
+ (Δ spend per visit × member visits)
+ Reactivation value
− Reward cost (at marginal cost)
− Software cost
− Implementation and staff time
This is a framework for structuring an estimate, not a promise of a result.
Self-selection. The customers who join a loyalty program are, on average, the ones who were already visiting most often. Comparing members' behaviour after joining with the same members' behaviour before joining will therefore show a lift that is partly real and partly an artefact of who volunteered. The rigorous fix is a matched comparison: compare enrolled members against non-enrolled customers with similar prior behaviour over the same period. If you cannot do that, present your figure as an estimate with a wide error bar and say so — a program defended with an honest range survives scrutiny; one defended with an inflated point estimate does not.
A homewares retailer with 900 enrolled members. Baseline before launch: members averaged 0.8 visits per month; average transaction value £46; contribution margin 45%, so £20.70 per visit.
After six months, suppose member visits average 0.95 per month. Δ visits = 0.15 × 900 × 6 = 810 additional visits × £20.70 = £16,767. Suppose average transaction value is unchanged: £0. Reactivation not separately measured: counted as £0 to stay conservative.
Costs: 340 rewards redeemed at £6.50 marginal cost = £2,210. Software 6 × £62 = £372. Implementation and staff time = £900.
Estimated incremental contribution ≈ £13,285.
The caveats that must travel with that number: self-selection is uncontrolled; seasonality is uncontrolled; the six-month window may not cover a full purchase cycle for this category. Treat it as a structured estimate, not a result. Every figure here is invented to demonstrate the method.
- Capture the baseline before launch. It is the cheapest step and the most-skipped.
- Adoption metrics and behaviour metrics answer different questions — do not mix them.
- Compare members against matched non-members, never against members' own past.
- Value rewards at marginal cost, not retail price, or you will kill a healthy program.
- One primary metric per campaign. More than four metrics and nobody reviews any of them.
Reporting passes issued as the headline number. Passes issued measures your counter script. It says nothing about whether anyone came back — which is the only thing the program exists to change.
Building, running and paying for it
A twelve-step implementation guide, the app question settled, the mistakes that sink programs, a calendar, the architecture, and an honest treatment of cost and consent.
15. How to build a retail wallet loyalty program
Twelve steps, in this order: define the objective, choose the mechanic, define customer identity, design the pass, configure the wallet platforms, configure the loyalty rules, plan the messaging, train the staff, test, launch, measure, optimise. The two most-skipped steps — defining identity and training staff — are the two that determine whether any of the others matter.
Step 1 · Define the objective
Write one sentence naming the behaviour you want to change: more visits, larger baskets, fewer lapses, or a shift in category mix. "Increase loyalty" is not an objective; it is a category. The objective determines the mechanic, and the mechanic determines everything downstream, so a vague first step produces a vague program.
Step 2 · Choose the reward mechanic
Use the Retail Loyalty Mechanic Matrix™ in Section 6. Then apply the reachability test: divide your reward threshold by the real visit frequency of a typical customer. If the answer exceeds a few months, either lower the threshold or change the mechanic. A reward nobody reaches is not a reward; it is an accounting liability with a marketing budget attached.
Step 3 · Define customer identity
Decide exactly how a customer will be recognised at your counter, and walk through it physically before you buy anything. Options, in ascending order of complexity: the customer shows a QR code and staff scan it; staff look up a phone number; an NFC tap on a certified terminal; a match against an online order. Whichever you choose, the process must survive your busiest hour with your newest employee. Time it. If the loyalty step adds noticeable delay at peak, it will quietly stop happening within weeks.
Step 4 · Design the pass
Follow Section 5. Front of card: brand, the one number, the gap to the next reward, a code that scans. Back of card: terms, expiry, store details, service contact, privacy notice, how to leave. Test the design in bright light, in dark mode, and at arm's length.
Step 5 · Configure Apple Wallet and Google Wallet
The specifics differ by platform and by the tool you use, but the shape is the same: on Apple you need a pass definition and a web service that can update it; on Google you need a class and per-customer objects created through the API. Two details worth planning for at this stage rather than discovering later: Apple's push for pass updates works only in the production environment, and Google's save link is a signed JWT with a documented safe length of 1,800 characters. Confirm both platforms render your design on real devices before you print a single QR code.
Step 6 · Configure the loyalty rules
Earning rate, reward threshold, caps, exclusions, expiry, tier boundaries if you have them. Add at least one anti-abuse rule from the start — most commonly one earning event per customer per day — because retrofitting one after a customer has exploited a gap is a conversation nobody enjoys.
Step 7 · Plan the messaging
Write the three messages you will actually send before you launch: the welcome, the reward-earned, and the reward-expiring. Then write your send policy — frequency ceiling, segmentation floor, quiet hours, recency suppression — and put it somewhere the whole team can see it. Programs do not over-message by decision; they over-message by drift.
Step 8 · Train the staff
This is the step that most often decides the outcome, and it is usually allocated ten minutes. Every team member should be able to say the program in one sentence, complete the enrolment in one motion, and handle the three questions customers actually ask: what do I get, what are you doing with my data, and what if I lose my phone. Have every staff member enrol themselves, earn once and redeem once before launch. Then plan to repeat the training, because retail staff turnover means your program is only ever one rota away from being forgotten.
Step 9 · Test
- Complete a full transaction including the loyalty step, timed at simulated peak.
- Save the pass on both an iPhone and an Android handset and confirm it renders correctly on each.
- Change a balance and confirm the pass updates on both devices. Run the Apple half of this against production — Apple documents that pass-update pushes work only in the production environment.
- Confirm the lock-screen message appears and reads correctly on both platforms.
- Earn a reward end to end and redeem it at the till, using the real workflow.
- Export your customer list and open the file — check identifiers and balances are complete and correct.
- Deliberately break the connection and confirm staff can still award manually.
- Read the back of the pass as a customer would, and check the privacy wording is true.
Step 10 · Launch
Soft-launch for two or three days with staff enrolling regulars they already know by name. Put the QR code where customers already wait — the card machine, the counter edge, the receipt. Convert existing paper or plastic card holders in person, honouring their existing progress; someone carrying a half-full punch card has already demonstrated exactly the behaviour you are trying to buy.
Step 11 · Measure
Compare against the baseline you captured in Layer 1 of the measurement framework. Review one primary metric monthly. Watch pass deletions after every campaign.
Step 12 · Optimise
Change one variable at a time and give each change a full purchase cycle before judging it. In our editorial judgement, in this order: the enrolment ask and its placement; the reward threshold; the message timing; the reward itself. Most retailers start with the reward, which is the hardest of the four to change and the one we would try last.
Week 1 — Strategy and baseline
- Write the objective in one sentence.
- Choose the mechanic using the Mechanic Matrix™.
- Record the baseline: visits per customer per month, average transaction value, identified-revenue share.
- Decide and physically walk through the identification method.
- Set expiry policy and earning caps.
- Decide what data you will collect at enrolment — and what you will not.
- Choose the one KPI you will review monthly.
Week 2 — Build
- Design the pass front and back.
- Configure earning rules, thresholds, caps and expiry.
- Set up the identification workflow at the counter.
- Import and de-duplicate any existing customer list.
- Create enrolment assets: counter QR, receipt line, window sticker, email footer, order confirmation link.
- Write the welcome, reward-earned and reward-expiring messages.
- Write and circulate the send policy.
Week 3 — Test and train
- Run every item in Step 9 above.
- Train all staff; each one enrols, earns and redeems personally.
- Agree the one-sentence pitch and write it where the tills are.
- Confirm the privacy notice is accurate and reachable from the pass.
Week 4 — Launch and tune
- Soft-launch with known regulars for two to three days.
- Convert paper and plastic card holders, honouring existing progress.
- Watch enrolment rate daily; low enrolment is nearly always placement or the ask, not the reward.
- Review at day 30 against the week-1 baseline and change one variable only.
- Identity and staff training are the two steps that decide the outcome, and both are usually rushed.
- Test the failure path, not just the happy path.
- Convert existing paper-card holders in person — they are your best starting cohort.
- Optimise the enrolment ask before you optimise the reward.
16. Retail wallet loyalty without a mobile app
You do not need a mobile app to run a retail loyalty program. There are three delivery routes — a native app, a mobile website, and a wallet pass — and the wallet route removes the install barrier while keeping a durable, updatable card and a notification channel. An app still wins where loyalty is inseparable from ordering, payment or rich browsing.
| Dimension | Native mobile app | Mobile website | Wallet pass |
|---|---|---|---|
| Installation | App store download plus account creation | None — a URL | One tap; the wallet is pre-installed |
| Development | Two codebases, ongoing | One web app | Configuration, not code |
| Maintenance | OS updates, store review cycles, regressions | Moderate | Handled by the pass platform |
| Notifications | Full push, rich formats, your rules | Web push only, patchily supported | Lock-screen updates, platform-capped |
| Friction to join | Highest in retail marketing | Low | Low |
| Friction to use at the till | Unlock, find app, open, sign in | Unlock, find browser, find the page, load it | Card is in the wallet, often surfaced automatically |
| Discoverability a month later | Buried on a home screen | A lost bookmark | In the wallet they open to pay |
| Loyalty visibility between visits | Only if opened | Only if revisited | Balance visible and current on the card |
| Works offline | Partially | No | Yes — the pass is stored on the device |
| Cost shape | Build plus maintain plus platform | Build plus hosting | Subscription, usually flat |
| What only it can do | Ordering, payment, browsing, account management, rich content | Any web experience, deep-linkable from anywhere | Sit permanently in the payment surface with no install |
Where an app genuinely wins: if customers order and pay through your app weekly, the app is the product and loyalty is a feature of it. Building an app so that customers can display a barcode is an expensive solution to a cheap problem — and it fails on adoption long before it fails on capability.
If you already have an app with genuine weekly usage, run loyalty in both places — the pass reaches the customers who installed the app once and never opened it again. If you do not have an app, do not build one for loyalty. Start with the wallet pass; install friction is the single most common cause of retail loyalty programs never reaching critical mass.
- Three delivery routes; only the app requires an install, and the install is where programs die.
- Wallet passes work offline, update remotely, and live where customers already look to pay.
- Apps win when loyalty is attached to ordering and payment.
- If you have an app already, run both — they reach different people.
17. Common in-store marketing mistakes
The most expensive mistake is running promotions without ever capturing identity. The most common is over-complicating the reward. The most damaging long-term is over-messaging, because in wallet marketing the penalty is deletion rather than silence.
| Mistake | What it looks like | Fix |
|---|---|---|
| Too many promotions | A permanent sale; customers wait rather than buy | Fewer, deeper, targeted at segments rather than at everyone |
| Unclear offers | Terms staff cannot explain and customers dispute at the till | One sentence, one exclusion at most, one deadline |
| Weak loyalty incentive | A reward so small it does not justify the enrolment ask | Make the first reward reachable and visibly worth having |
| Complicated rewards | Points, tiers, multipliers and blackout dates on day one | Launch one mechanic; layer only after adoption is proven |
| Poor staff training | Half the team never mentions the program | One-sentence pitch, one-motion enrolment, refreshed with turnover |
| No customer identification | Thousands of sales, no idea who bought | Fix this before anything else; it gates every other action |
| Excessive notifications | Three sends a week; pass deletions climb | Let balance changes carry the load; broadcast only new information |
| No segmentation | Every message to everyone, including yesterday's customer | Segment by recency and category as an absolute minimum |
| No measurement | "It feels like it's working" | Baseline before launch; one primary metric reviewed monthly |
| Weak post-purchase strategy | Silence between the sale and the next campaign | Bounce-back offers and product-timed follow-ups |
| Ignoring online/offline integration | The same person counted as two customers | One identity across both; the wallet card is the bridge |
| Collecting unnecessary data | A six-field signup form abandoned in the queue | Ask for the minimum the reward rule needs; enrich later |
| Poor privacy practice | Consent buried, no clear way to leave, vague purpose | Plain-English purpose statement; one-tap opt-out; a tested deletion path |
Treating in-store marketing as a creative exercise rather than an operational one. The offers, the artwork and the calendar are the easy half. The hard half is a workflow that a tired member of staff performs correctly a hundred times a day, in a queue, on a Saturday. Design that first and the creative work becomes much easier.
18. The retail marketing campaign calendar
A workable retail calendar has roughly one major moment a month, a small number of always-on triggered campaigns, and deliberate quiet periods. The always-on campaigns — welcome, birthday, bounce-back, replenishment, win-back — are individually timed to each customer and run without recurring attention, which is why they are worth building before the calendar. We publish no comparative revenue figure for the two.
The calendar below is a Northern-Hemisphere retail example. Your categories, your peaks and your local events will differ, and some of these moments will be irrelevant to you. Use it as a shape to argue with, not a plan to copy.
| Period | Retail context | Loyalty role | Wallet campaign | Primary metric |
|---|---|---|---|---|
| January | Post-holiday quiet; clearance; new-year intent | Reactivate December's one-time gift buyers | Win-back coupon to anyone who bought once in Q4 and not since | Reactivation rate |
| February · Valentine's | Gifting spike in specific categories | Target only the categories that apply | Category-segmented offer; nothing to the rest of the base | Category revenue from members |
| Spring | New season; refresh intent | Tier access to new-season stock | Early access window for the top tier; new pass artwork | Tier member transactions in the window |
| Summer | Mid-year sale; holiday and travel demand | Clear stock privately before public markdown | Members' clearance coupon | Sell-through and realised margin |
| Back to school | Concentrated, deadline-driven demand | Bundle and threshold rewards | Spend-threshold reward for the period | Average transaction value |
| Autumn | New season; gifting build-up begins | Enrolment push before the peak | Enrolment drive so peak traffic leaves a residue | Wallet save rate |
| Black Friday period | Highest traffic, lowest margin, most noise | Enrol, do not just discount | Every transaction gets an enrolment ask; bounce-back into December | New enrolments; second-visit rate |
| Christmas | Peak trade, many one-time gift buyers | Capture identity from people buying for others | Enrolment at the counter; a January bounce-back issued in December | Enrolment rate; January redemption |
| Customer birthdays | Always-on, individually timed | A visit on a date unrelated to your calendar | Automated birthday coupon | Birthday redemption rate |
| Store anniversary | A moment that is genuinely yours | Reward existing members rather than discount to strangers | Members-only day; double earning | Member transactions on the day |
| Local events | Markets, festivals, sports fixtures, school terms | Relevance no national retailer can match | Local segment message tied to the event | Footfall and member transactions that day |
| Always-on triggers | Continuous, individually timed | The engine running underneath the calendar | Welcome · bounce-back · replenishment · lapse win-back | Second-visit rate; reactivation rate |
Peak trading periods are usually treated as pure margin sacrifice in exchange for volume. For a retailer with an identification mechanism, they are something better: the cheapest customer-acquisition window of the year. You already have the traffic and you are already paying for it in margin. Adding an enrolment ask to every transaction converts a one-day margin event into a customer base you still have in March. If you do one thing from this entire guide before the next peak, do that.
- Always-on triggered campaigns are individually timed and need no recurring attention — build them before the calendar.
- Peak trading periods are enrolment opportunities, not just discount events.
- Deliberate quiet periods are part of the calendar, not a gap in it.
- Segment every seasonal campaign; not every season applies to every customer.
19. POS + loyalty + wallet architecture
Four layers: the point of sale records the transaction; the loyalty platform holds customer identity, rules and balances; the wallet holds the customer's card; the notification carries the message back. Only the first is owned by your till, which is why you almost never need to replace it to run a modern loyalty program. The exact architecture depends on your own stack.
Figure 6 — POS, loyalty and wallet architecture. Four layers, four possible connection methods, and the two points where connective work is genuinely required. (Original PushNotice diagram.)
The important property of this architecture is that the layers are separable. Change your till and the customer keeps the same wallet card with the same balance. Change your loyalty vendor and the counter workflow does not change. That is not a product claim — it is simply what layering gives you, and it is why "you need a new POS to run loyalty" is usually a product boundary rather than an architectural fact. Our companion guide on adding loyalty to an existing POS works through the integration options in far more depth.
We are not going to tell you which of the four connection methods you need, because it depends on your till, your reward mechanic and your appetite for an integration project. What we will say plainly: start at the loosest coupling that supports your mechanic and tighten only when a measured problem justifies it. It is far easier to add an integration to a program that is already working than to rescue a program that never launched because the integration was not ready.
20. The Wallet Loyalty Maturity Model™
Five levels: Anonymous, Issued, Identified, Engaged, Orchestrated. In our experience of the category, retailers commonly sit at level 0 or 1 and try to buy their way straight to level 4. The value is in moving one level at a time, because each level's problems are only visible from the level below it.
Level 0 · Anonymous. Transactions happen; nobody is identified. Marketing is entirely in-store and entirely disposable. Next move: choose an identification method.
Level 1 · Issued. Cards exist and some customers have saved them, but scanning is inconsistent and the data is patchy. Next move: fix the counter workflow and the staff script.
Level 2 · Identified. Most member transactions are captured. You can now measure frequency, spot lapses and segment. Next move: start using the data — timing and segmentation before offers.
Level 3 · Engaged. Messages are segmented and timed to customer cycles rather than your calendar; redemption is healthy; deletions are low. Next move: connect the online channel to the same identity.
Level 4 · Orchestrated. One identity across store and web; always-on triggered campaigns doing most of the work; measurement against matched cohorts; the calendar is the smaller half of the program. Next move: maintain it — this level decays without attention.
| Level | Diagnostic question | Typical symptom | Highest-value action |
|---|---|---|---|
| 0 · Anonymous | Can you name a customer who came in twice? | Marketing feels like starting over every week | Add an enrolment ask to the checkout |
| 1 · Issued | What share of member transactions include a scan? | Passes issued far exceed passes used | Retrain staff; move the QR to where customers wait |
| 2 · Identified | Can you list who has not visited in 90 days? | Good data, no campaigns using it | Launch one triggered campaign — bounce-back or win-back |
| 3 · Engaged | Are your sends segmented, and are deletions flat? | Growth stalls; store and web still separate | Unify online and in-store identity |
| 4 · Orchestrated | Does the program run without weekly intervention? | Complacency; slow decay in relevance | Review cohorts quarterly; prune messages that no longer earn their place |
A retailer at level 1 who buys a personalisation engine gets nothing, because personalisation needs the behavioural history that level 2 produces. A retailer at level 2 who launches five simultaneous campaigns gets deletions, because level 3 is about restraint and timing rather than volume. Each level's tooling only works on the foundation the previous level built — which is why the honest advice is almost always "do the boring level below the one you were about to buy."
21. Who should — and should not — use wallet-based retail loyalty
Wallet loyalty suits retailers with plausible repeat purchase, an identifiable customer, and no appetite for building an app — which describes most independent and multi-site retail. It does not suit businesses with genuinely one-time customers, businesses that cannot obtain consent, or retailers whose loyalty proposition is too weak to be worth carrying.
| Profile | Fit | Why |
|---|---|---|
| Small independent retailers | Strong | Zero development, launches in days, works with the scanner already on the counter |
| Boutiques | Strong | Makes an existing personal relationship explicit and measurable without making it transactional |
| Fashion brands | Strong | Tier and access mechanics suit the category; artwork carries the brand |
| Multi-location retailers | Strong | One identity and one balance across sites is the core advantage |
| Franchises | Strong | The only layer that can be standardised when operators choose their own tills |
| Retail plus Shopify or another ecommerce platform | Strong | The pass bridges online enrolment and in-store redemption |
| Retailers with a POS but no loyalty feature | Strong | Barcode identification requires nothing from the till |
| Retailers without an app | Strong | The wallet is the app you do not have to build or maintain |
| Retailers with a well-used existing app | Complementary | Run both; the pass reaches the customers who never open the app |
| Retailers whose POS loyalty is included and adequate | Consider carefully | If it works, is paid for, and you will not change till, use it — see Section 24 |
Who should not use it
1. Businesses with almost no repeat customers. Airport concessions, tourist-district gift shops, one-off destination retail. There is no repeat behaviour to reward, and the enrolment effort is a tax on the transaction. Spend it on conversion and basket size instead.
2. Purely one-time purchases. If the product is genuinely bought once — a category where a second purchase would be irrational — a loyalty balance is a promise you cannot keep. A membership or a service relationship may fit; a points card does not.
3. Businesses that cannot obtain proper consent. If your enrolment process cannot support a clear, freely given, informed and specific consent, and a plain way to leave, do not launch. This is not a formality; it is the difference between an asset and a liability.
4. Retailers without a meaningful loyalty proposition. If the best reward you can afford is 2% back after twelve visits, the customer is being asked to carry a card for almost nothing. Fix the proposition first. A weak program is worse than no program, because it teaches customers that your loyalty offer is not worth attention.
If your till already includes loyalty in a plan you pay for, you run one site, and you have no intention of changing till — use it. If you need certified NFC tap-to-earn on enterprise terminals, you need a platform with a certification programme and terminal partnerships. If you need deep programmatic control over pass issuance at very high volume, a developer-first wallet infrastructure platform is a better foundation. Saying so is the point of writing an honest guide rather than a brochure.
22. The cost of retail loyalty
Eight cost categories: software, design, POS integration, implementation, staff training, promotion, customer acquisition and maintenance — plus the one that is usually largest and usually omitted, the actual margin cost of rewards redeemed. Do not compare vendors on subscription price alone; model the whole thing over twenty-four months at the size you expect to be, not the size you are.
| Category | What it covers | Usually underestimated because… |
|---|---|---|
| Software | The loyalty or wallet platform subscription | It is priced at today's size, not the size you will be in month 24 |
| Design | Pass design, in-store collateral, signage, receipt artwork | It is treated as a one-off, but seasonal refreshes recur |
| POS integration | Connector fees, developer time, partner programme costs | Vendor approval timelines are outside your control |
| Implementation | Rule configuration, data import, testing, QR placement | Testing properly takes longer than configuring |
| Staff training | Hours × wage × headcount × turnover multiplier | It is set to zero, and retail turnover means it recurs several times a year |
| Promotion | Announcing the program in store and online | A program nobody knows about enrols nobody |
| Customer acquisition | Any spend to drive enrolment specifically | Often absorbed invisibly into general marketing |
| Maintenance | Ongoing configuration, integration upkeep, data hygiene | Assumed to be zero until something breaks |
| Reward cost | The margin cost of every reward actually redeemed | Frequently the largest line, and routinely left out of the software comparison entirely |
Model over a 24-month horizon:
TCO = Software + Design + Integration + Implementation + Training + Promotion + Acquisition + Maintenance + Reward cost
Two rules that prevent most bad decisions. First, price the plan you will need in month 24, not the one you need this week — and pay particular attention to which meter grows fastest when the program succeeds. A price per member punishes success; a flat price does not. Second, value redeemed rewards at marginal cost, not at retail price. A free £30 item does not cost you £30; it costs you what it cost you. Overstating reward cost is the most common way a healthy program gets killed at its first budget review.
A two-site homewares retailer, 24-month horizon. Software at £62/month = £1,488. Pass and collateral design, plus one seasonal refresh = £600. Integration: none, because identification is a counter scan = £0. Implementation and testing = £450. Staff training: 12 people × 1 hour × £13, three times over two years for turnover = £468. Promotion: in-store collateral and an announcement = £350. Customer acquisition specifically for enrolment = £0, absorbed into general marketing — name the line even when it is zero, or it will never be questioned. Maintenance allowance = £240. Reward cost: 1,300 rewards redeemed over two years at £6.50 marginal cost = £8,450.
Two-year TCO ≈ £12,046 — of which the software is roughly 12%. Every number here is invented to demonstrate the method. Substitute your own, especially the reward line.
We are not quoting competitor prices in this guide. Loyalty and wallet pricing changes without notice, varies by country, and several vendors publish nothing at all — so any table of figures here would be stale before it was useful. Check each vendor's own pricing page on the day you evaluate. PushNotice's own published plans are described in Section 24, with the date checked.
- Nine cost lines, of which reward cost is usually the largest and the most often omitted.
- Staff training is set to zero far more often than it is actually zero.
- Model the plan you need in month 24, and watch which meter grows with success.
- Value rewards at marginal cost. Retail-price accounting kills healthy programs.
23. Customer data, consent and privacy
Collect the minimum your reward rule needs, record enrolment consent and marketing consent separately, say plainly what you are doing in language a shopper can read in the queue, and know how you would delete someone's record before you are asked. A loyalty database is a first-party asset — which is exactly why it carries obligations.
Loyalty programs are one of the few remaining ways a physical retailer builds a genuine, consented, first-party relationship with its customers. That is their strategic value in a world where third-party tracking is contracting. It is also precisely why they attract scrutiny: a loyalty database is, by definition, a record of identifiable individuals and their behaviour.
| Field | Why you might want it | Privacy weight | Guidance |
|---|---|---|---|
| Pseudonymous member ID | Links visits without naming anyone | Low | Prefer this as the primary key. For a stamp card it is often all you need. |
| First name | Personalisation and counter recognition | Low | Optional field, never required. |
| Email address | A second channel and account recovery | Medium | Only collect it if you will actually use it. Store consent separately from the address. |
| Phone number | Counter lookup and an SMS channel | Medium | Useful as an identifier, but SMS marketing consent is a separate permission with stricter rules. |
| Date of birth | Birthday rewards | Medium | Day and month only. You almost never need the year. |
| Transaction amounts | Spend-based earning, basket analysis | Medium | Aggregate where you can; per-line detail is rarely necessary. |
| Itemised basket | Product-specific rewards | High | Purchase histories can be revealing — in some categories acutely so. Collect only with a clear stated purpose. |
| Store visited | Multi-site attribution and local segmentation | Medium | Store-level is fine; continuous device location is a different matter entirely. |
| Device location | Geofenced campaigns | High | See Section 10. Prefer behavioural signals; if you use location, say so explicitly and cap frequency. |
| Communication consent | Lawful, wanted marketing | Required | Record what they agreed to, when, and how. Make opting out a single tap. |
Five practices that keep a retail loyalty database defensible
- Minimise at the design stage. Under the UK and EU GDPR, personal data must be "adequate, relevant and limited to what is necessary in relation to the purposes for which they are processed" — Article 5(1)(c), the data-minimisation principle. Decide the reward rule first, then collect only its inputs.
- Separate enrolment consent from marketing consent. Joining a stamp card is not agreement to weekly promotions. Two permissions, two records. Note that the ICO states you "must have consent to send electronic mail marketing to individual subscribers (unless one of the 'soft opt-ins' apply)," and that its definition of electronic mail expressly includes SMS.
- Write a plain-English purpose statement on the enrolment screen: what you collect, why, how long you keep it, and how to leave. Two clear sentences beat a linked policy nobody opens.
- Know and test your deletion path. If a customer asks to be erased, can you do it without breaking your reporting? Test it once, before you need it.
- Set a retention limit. A member who has not visited in three years is not a marketing opportunity; they are a liability. Define an inactivity horizon and honour it.
This section describes general data-protection principles, not legal advice, and requirements vary by jurisdiction — UK and EU GDPR, PECR and the ePrivacy Directive, US state privacy laws, PIPEDA in Canada, and others. Two specifics worth flagging: the ICO's guidance was updated in April 2026 to list seven lawful bases rather than six, following the addition of "recognised legitimate interest"; and the EDPB is clear that marketing cannot be run on the "performance of a contract" basis. Confirm your own position with a qualified adviser before launching, particularly around SMS and email consent.
- A pseudonymous member ID covers most stamp and visit programs.
- Enrolment consent and marketing consent are different permissions — record both.
- Data minimisation is a legal principle and an engineering simplification simultaneously.
- Test your deletion path before a customer forces you to.
- Itemised purchase data and device location are the two highest-weight fields. Justify them or skip them.
PushNotice, the benchmark and the toolkit
Where our own product fits and where it does not, an editorial benchmark that discloses its own bias, and ten companion resources.
24. PushNotice for retail
PushNotice can act as the wallet-based customer engagement layer of a retail loyalty program — the card the customer carries, the balance they see, the segments you target and the messages you send — sitting alongside the till, ecommerce platform and other systems a retailer already runs. It is not a point-of-sale system, a CRM, an ERP or an ecommerce platform, and we do not claim a native integration with any named POS.
Having spent this guide arguing that in-store marketing needs a persistent layer, it would be poor form to describe our own product as something it is not. Here is the accurate version, checked against what is published on pushnotice.io in August 2026.
Where it fits
Retailer → Loyalty program → Apple Wallet / Google Wallet → Customer engagement → Notifications → Retention
What PushNotice does
PushNotice is a wallet marketing platform. A retailer designs a pass — a loyalty or stamp card, a membership or store card, a coupon, an event ticket, or a generic pass — and customers save it to Apple Wallet or Google Wallet with one tap, with no app to download and no account to create. Every pass carries a scannable barcode or QR code, which is what makes it work at a counter: the code can be read by any 2D-capable scanner or imager, with no special hardware and no dependency on which point-of-sale system you run. If your counter still runs a 1D-only laser scanner, test it against a real pass before you launch, or use a 1D barcode format instead.
The platform includes a customer database, tagging and segmentation, campaign broadcasts with targeting and scheduling, an analytics dashboard covering installs and engagement, and white-label workspaces for agencies managing multiple clients. Published plans start at a permanently free tier and run to $79 per month for Pro, with an Agency tier quoted on request. Notifications work the way both wallet platforms intend: a push prompts the device to fetch the latest pass, and the updated pass content is what appears on the lock screen. Once installed, passes are stored on the device and work offline.
| Capability | Status | Detail |
|---|---|---|
| Apple Wallet passes | Yes | Built on Apple's PassKit framework, with remote updates and lock-screen delivery |
| Google Wallet passes | Yes | Android users are served the Google Wallet version of the same link. Confirm the current behaviour with us rather than assuming it for a specific device mix |
| Loyalty and stamp cards | Yes | A core pass type |
| Membership and store cards | Yes | A core pass type — suits high-ticket retail where status beats a balance |
| Coupons and offers | Yes | A core pass type — issue a coupon separately so redeeming it does not disturb the loyalty card |
| Event tickets | Yes | A core pass type — useful for the in-store event mechanic in Section 7 |
| Works with any 2D-capable scanner | Yes | Every pass carries a scannable code. 1D-only laser scanners cannot read QR — test yours, or use a 1D format |
| Customer database, tagging and segmentation | Yes | Tagging and segmentation on Pro; install and engagement tracking across plans |
| Campaign broadcasts | Yes | With targeting and scheduling; campaign volumes differ by plan |
| Multi-location and multi-client | Yes | Workspaces — 1 on Free and Starter, 5 on Pro, white-label on Agency |
| Named native POS integrations | No | We do not currently publish an integration with any named point-of-sale system. If you need automatic earning driven by till transactions, ask us about your specific setup rather than assuming |
| Geofenced notifications | Ask us | Published plans list a feature named "Smart Location" from Starter upwards. We are not going to describe its behaviour here beyond that. Any geofencing capability is bounded by the platform constraints in Section 10 — particularly Apple's documented position that pass relevance posts no notifications |
| Public developer API / webhooks | Not published | No public developer API or webhook documentation is published at the time of writing. Do not assume programmatic access without checking with us |
| Point-of-sale, CRM, ERP or ecommerce replacement | No | PushNotice is a layer alongside those systems, not a substitute for any of them |
Where PushNotice fits the architecture in this guide is the loose-coupling end: the till keeps doing its job, identification happens by scanning the pass at the counter, and the loyalty, wallet, segmentation and messaging layers sit on our side. That is the right fit for a retailer who wants a modern loyalty card in Apple Wallet and Google Wallet without an integration project, and the wrong fit for a retailer who requires points to post automatically from an enterprise point-of-sale system with no human step. Both are legitimate requirements. Only one of them is ours.
25. The PushNotice Retail Loyalty Benchmark 2026
This is a PushNotice editorial evaluation of in-store marketing approaches against seven criteria drawn from documented platform behaviour and general retail practice. It is not independent research, it contains no survey data, and it scores approaches rather than ranking named products.
We do not hold proprietary industry research on retail loyalty, and we are not going to invent any. What follows is a PushNotice Editorial Framework: a structured assessment of six in-store marketing approaches against seven criteria, drawn from platform documentation, published regulatory guidance and general retail practice, all cited in the Sources section. Scores are our editorial judgement, on a 1–5 scale. They are reasoning made explicit, not measurement. Individual products within any approach will vary widely.
| Criterion | Signage & merch. | Printed coupons | SMS | Retail app | Wallet loyalty | |
|---|---|---|---|---|---|---|
| Campaign design flexibility | 3 | 3 | 5 | 3 | 5 | 4 |
| Loyalty mechanic support | 1 | 2 | 3 | 2 | 5 | 5 |
| Customer experience / friction | 5 | 4 | 4 | 3 | 2 | 5 |
| Persistence between visits | 1 | 2 | 3 | 2 | 3 | 5 |
| Notification capability | 1 | 1 | 3 | 5 | 5 | 3 |
| Measurability | 2 | 3 | 4 | 4 | 5 | 4 |
| Implementation simplicity (5 = simplest) | 5 | 5 | 4 | 3 | 1 | 4 |
| Unweighted total (max 35) | 18 | 20 | 26 | 22 | 26 | 30 |
How to read this responsibly. The totals are unweighted, which means they are almost certainly wrong for your business. Weight the criteria yourself. A grocery retailer should weight "customer experience / friction" and "implementation simplicity" heavily, because a queue is unforgiving. A luxury retailer should weight "notification capability" at close to zero, because the right number of promotional notifications in that category is approximately none. A retailer that already has a well-used app should weight "loyalty mechanic support" differently again. The value of this table is the criteria list, not the arithmetic.
Our disclosed bias. PushNotice sells a product in the wallet loyalty category, which scores highest here. We have tried to counteract that in three ways: by scoring wallet's notification capability at 3 rather than 5, because both platforms cap it and Apple's relevance posts nothing at all; by scoring signage and merchandising at the maximum on customer experience, because nothing beats a well-merchandised shelf for a shopper standing in front of it; and by scoring the retail app equal to email on total, because where an app is genuinely used it is a formidable channel. Read the criteria, apply your own weights, and discount our judgement accordingly.
26. Downloadable resources
Ten companion resources turn this guide into a working toolkit: a program planner, an in-store marketing checklist, a campaign calendar, an ROI calculator, a KPI dashboard, a pass planning worksheet, an offer planning template, a retention planner, a vendor evaluation checklist and a launch checklist.
A one-page worksheet that forces the six decisions in order: objective, mechanic, threshold, identification method, expiry policy and primary KPI.
For: owners and marketing leads designing a program for the first time. Why link to it: it is the only planner we know of that requires the identification decision before the reward decision — the order in which programs actually succeed or fail.
Preview: six numbered boxes, each with a single question and a worked example beneath it.
Download the planner →A store-walk checklist covering the seven disciplines from Section 1 — merchandising, signage, promotions, staff selling, checkout marketing, loyalty and post-purchase — with a scoring column.
For: store managers and multi-site operators doing a periodic audit. Why link to it: most retail audit templates stop at the shop floor; this one includes what happens after the customer leaves.
Preview: 48 checks across seven sections, scored out of 5.
Download the checklist →An editable twelve-month planning grid with the always-on triggered campaigns separated from the calendar moments, plus space for local events.
For: anyone planning a retail year. Why link to it: it separates triggered from calendar campaigns, which is the distinction most retail calendars miss.
Preview: twelve rows, four columns — moment, loyalty role, wallet campaign, primary metric.
Download the calendar →A spreadsheet implementation of the Measurement Framework™, with baseline capture fields, marginal-cost reward accounting, and a matched-cohort comparison tab.
For: owners and finance. Why link to it: it models selection bias instead of quietly ignoring it, which almost no vendor calculator does.
Preview: three tabs — baseline, behaviour, value — with the formula exposed rather than hidden.
Open the calculator →A monthly reporting template covering the twelve metrics in Section 14, with the primary metric highlighted and vanity metrics deliberately excluded.
For: operators reporting to owners or a board. Why link to it: it names the metrics that should not be reported, which is the harder half.
Preview: one page, four blocks — adoption, behaviour, value, health.
Download the dashboard →A field-by-field worksheet for the pass: what goes on the front, what goes on the back, which field carries the change message, and which barcode format you will use.
For: whoever designs the card. Why link to it: it prevents the most common pass design failure — terms on the front, balance on the back.
Preview: a card outline with seven numbered slots and a back-of-pass list.
Download the worksheet →One page per offer: objective, segment, mechanic, deadline, wallet implementation, the notification it justifies, and the single KPI — the structure used for the twenty ideas in Section 7.
For: marketers running more than one campaign a month. Why link to it: it makes the "which KPI?" question mandatory before launch rather than optional afterwards.
Preview: a seven-field brief, one side of A4.
Download the template →A planner built around the Retail Engagement Ladder™: count your customers by rung, then plan one promotion move per rung for the quarter.
For: owners who suspect their marketing is stuck at the top of the funnel. Why link to it: the rung-count diagnostic reframes the problem faster than any campaign review.
Preview: five rungs, a count box and a single action per rung.
Download the planner →The questions that predict regret rather than the ones that predict a good demo: identification workflow, data export, portability, notification limits, consent handling and exit terms.
For: anyone shortlisting vendors. Why link to it: it puts exportability and exit above features, which is the reverse of how most buyers evaluate.
Preview: 30 questions across six sections, with space for three vendors.
Download the checklist →The 30-day plan from Section 15 as a task list, including the failure test, the both-platforms device test, and the paper-card conversion step.
For: whoever runs the launch. Why link to it: it contains the tests that most launches skip and then regret.
Preview: four weekly blocks, 34 checks.
Download the checklist →The links above are marked as placeholders because the assets are specified here but not yet published. We would rather say that than present dead links as live downloads. If you want one of these before it ships, ask us via the contact page and we will tell you honestly when it is ready.
Reference
Fifty-four questions answered, the methodology and disclosures behind this guide, and every primary source it draws on.
27. Frequently asked questions
The three questions that come up most: yes, both Apple Wallet and Google Wallet can carry a retail loyalty card with no app installed; no, neither wallet connects to your till by itself; and the two platforms genuinely differ on location — Apple's pass relevance posts no notifications, while Google's merchant locations do trigger them, on Google's terms.
Strategy and fundamentals
What is in-store marketing?
In-store marketing is every deliberate action a retailer takes inside a physical store to influence what a shopper notices, chooses, buys and remembers. It covers merchandising, signage and wayfinding, in-store promotions and sampling, staff-led selling, checkout marketing, loyalty enrolment, and post-purchase engagement. The first five stop working when the customer leaves the building; the last two are the only ones that create an asset you can use again.
What are the best in-store marketing strategies?
The strategies that repay the effort do one of two things: capture a customer's identity, or give an identified customer a specific reason to return. In practice that means a welcome reward at enrolment, a checkout moment that recognises members, bounce-back offers that can only be redeemed on a later visit, replenishment and milestone reminders, member-only access to sales and events, and a win-back offer for customers who have stopped coming. Merchandising and signage remain essential for converting the shopper who is already present, but they leave no residue.
How can retailers increase repeat customers?
By making three things true. First, the store must be able to recognise a returning customer, which requires an identification step at the counter. Second, the customer must carry something that keeps the relationship visible between visits, which is what a wallet loyalty card does. Third, the retailer must reach out at moments that are useful to the customer rather than convenient for the marketing calendar, such as a reward becoming available, an offer expiring, or a product being due for replacement. Without the first, the other two are impossible.
What is retail loyalty?
Retail loyalty is a structured program that identifies customers and rewards repeat purchase or repeat visits. It has three components: an identity mechanism that links a transaction to a person, a set of rules that turn that activity into points, stamps, tiers or rewards, and a customer-facing card that shows the balance and can be presented at the counter. Modern retail loyalty puts that card in Apple Wallet or Google Wallet rather than in plastic or in a mobile app.
Does in-store marketing still matter now that people shop online?
Yes, and the data is unambiguous. The US Census Bureau's Q2 2026 e-commerce release put e-commerce at 17.1% of total retail sales, which means 82.9% of retail sales were not ordered online. That figure is a floor rather than a ceiling on store involvement, because the Census counts a sale as e-commerce based on where the order was placed, so buy-online-pickup-in-store orders sit inside the 17.1% even though the customer physically visits the shop.
What is the difference between in-store marketing and retail advertising?
Advertising's job is to get someone to the door. In-store marketing's job starts after they arrive. They are bought differently, measured differently, and fail differently: a store with excellent advertising and weak in-store execution converts expensive traffic into browsing. The two are complementary, but treating them as one budget usually means the in-store half is under-specified.
How much should a retailer spend on in-store marketing?
There is no defensible universal percentage, and anyone quoting one is guessing. The more useful question is about allocation rather than amount: what share of your current spend creates something that outlives the visit? Most independent retailers find that almost all of it is promotional and none of it is spent on identification. Shifting even a small proportion towards capturing identity usually produces more improvement than increasing the total.
Designing a loyalty program
How does wallet loyalty work?
A customer saves a digital loyalty card to Apple Wallet or Google Wallet, usually by scanning a QR code at the counter. The card carries a unique member identifier encoded in a barcode. At checkout, staff scan that barcode, which tells the loyalty platform who is buying. The platform applies the earning rules and updates the pass, and the customer's card changes on their phone without them doing anything. No app is installed at any point.
What should a retail loyalty card contain?
On the front: the brand mark and store name, the single number that matters to the customer such as points or stamps, the distance to the next reward, a tier if you run one, and a scannable barcode with the member identifier. On the back: full terms, the expiry policy, store addresses and hours, a customer service contact, the privacy notice, and a clear way to leave the program. If a shopper does not need it in a two-second glance, it belongs on the back.
What are examples of retail loyalty programs?
The common patterns are points per pound or dollar spent, redeemed against a reward; stamp cards where a fixed number of purchases earns a free item; spend thresholds that release a credit at a cumulative amount; visit-frequency rewards; tiered programs where cumulative activity unlocks status and access rather than discount; and paid membership schemes where customers pay for ongoing benefits. Most retailers should run one of these as a base and add birthday, referral or seasonal mechanics as layers on top.
Points or stamps — which is better for a retail store?
It depends on basket variability. If your customers spend roughly the same amount each visit, stamps are better: they are simpler, need no transaction data, and are easier for staff to explain. If baskets vary widely, points are better, because stamps would reward a small basket exactly as much as a large one. A useful shortcut is to look at the spread between your typical smallest and largest transactions; if it is more than about three to one, use points.
How many purchases should a reward take?
Divide the threshold by how often a typical customer actually visits, and ask whether anyone would wait that long. A ten-stamp card in a shop someone visits twice a week pays out in five weeks, which works. The same card in a shop visited every six weeks pays out in over a year, which is not a loyalty program. If the wait exceeds a few months, either lower the threshold or change the mechanic.
Should a small shop run a tiered loyalty program?
Usually not at first. Tiers need enough active members for the top tier to be a visible minority rather than a rounding error; a Gold tier that three people reach is not aspirational, it is invisible. Run a single flat mechanic until the member base is large enough, then introduce tiers when there is a genuine spread of customer value worth recognising.
Should loyalty rewards expire?
Expiry is legitimate and common, and it serves a real purpose: unexpired balances accumulate as an open-ended liability against your margin. The requirements are that you decide the policy before launch, disclose it clearly at enrolment, and warn customers before it applies. Introducing expiry retroactively, or applying it silently, converts a loyal customer into a complaint.
How do I stop loyalty fraud in a retail store?
Four controls cover most of it. Cap earning, most commonly at one earning event per customer per day. Verify redemption so a reward can only be used once. Set staff permissions with an audit trail on any manual balance change. And monitor for anomalies such as a single member earning at an implausible frequency. Design these in at launch; retrofitting them after someone has exploited a gap is an uncomfortable conversation.
Apple Wallet
Can Apple Wallet be used for retail loyalty?
Yes. Apple's storeCard pass style is described in Apple's own developer guidance as appropriate for store loyalty cards, discount cards, points cards and gift cards. The pass can carry a member identifier, a scannable barcode, a balance and a status, and it can be updated remotely so the balance changes without the customer doing anything. Apple Wallet does not connect to your till by itself, though: identification and synchronisation still have to be provided by you or your loyalty platform.
How does an Apple Wallet loyalty card update?
Your loyalty platform hosts a pass web service. When a balance changes, it sends a push to every device registered for that pass. Apple's documentation specifies an empty JSON dictionary as the payload, so the push carries no content; it simply tells the device to come and fetch the updated pass. One documented trap catches many teams during development: Apple states that a push notification for a pass update works only in the production environment.
Does Apple Wallet send location-based notifications?
No, and this is widely misreported. A pass can list up to ten relevant locations, and Apple's Wallet developer guidance is explicit that relevance is passive: it helps users find passes by putting them on the lock screen, and it does not present alerts or post notifications. If a vendor tells you Apple Wallet will send a push when a customer walks past your shop, that claim does not match Apple's own documentation.
What barcode formats work on an Apple Wallet pass?
Apple's pass barcode reference lists QR, PDF417, Aztec and Code 128, and the current reference also includes Code 39, Codabar, EAN-13 and ITF. Apple notes that Code 128 is not supported on watchOS. QR is the usual default for retail because 2D-capable scanners and phone cameras read it reliably. Note that 1D-only laser scanners, still common on older counters, cannot read QR at all — test yours, or choose a 1D format such as Code 128.
Do customers need an app to use an Apple Wallet loyalty card?
No. Apple Wallet is pre-installed on every iPhone, so a customer saves the card in a single tap from a QR code, a link or a web page, with no download and no account creation. This matters because it reaches customers who would not install an app at all. We do not quote an app-abandonment figure, because we found none from a named publisher disclosing a sample and a period.
Google Wallet
Can Google Wallet be used for retail loyalty?
Yes. Google Wallet's Loyalty API provides a loyalty class as the shared template and a loyalty object for each customer, with native fields for a points balance, a secondary balance, an account identifier and an account name, plus rewards tier fields on the class. Objects are updated server-side through Google's REST API. As with Apple, the wallet is a container: it has no connection to your point-of-sale system on its own.
How does a Google Wallet loyalty card update?
The pass object lives on Google's servers. Your platform calls the REST API to update or patch the loyalty object, and the change applies to that holder without a per-device handshake. Google's documentation notes that changes made to a class propagate immediately across all pass objects that reference it, so a single class edit affects everyone holding that card. That is powerful and worth respecting.
Does Google Wallet send location-based notifications?
Yes, unlike Apple Wallet, but on Google's terms rather than yours. Google's MerchantLocation reference states that when a user is within a set radius of the location and dwells there, Google will trigger a notification, and that the notification is hidden when the user exits. Google's loyalty guidance adds that Google decides how close the user must be and how long they must stay, and that Google also controls the text of the notification. It applies to users who have enabled notifications and granted precise, always-on location access to the Google Wallet app. Note also that the older locations field on the loyalty object is deprecated and carries a note that it is not supported for triggering geo notifications; the working field is merchantLocations.
How many notifications can I send through Google Wallet?
Google publishes a hard limit: a maximum of three messages that trigger a push notification in any 24-hour period, and a maximum of ten messages stored on an object. Google also allow-lists which field changes can trigger an update notification for loyalty passes: rewardsTier, secondaryRewardsTier and programName on the class, and loyaltyPoints.balance and secondaryLoyaltyPoints.balance on the object. Treat these as architectural constraints when planning campaigns, not as guidelines.
Does Google Wallet work on iPhone?
Google's developer documentation describes Google Wallet passes on Android-powered devices and does not document iOS support. We checked the loyalty overview, the FAQs and the web-issuing documentation and found no explicit statement either way, so we are not going to assert that it does or does not work on iPhone. In practice, a retailer should issue an Apple Wallet pass to iPhone users and a Google Wallet pass to Android users. Many wallet platforms serve the right one from a single link — confirm that with your own vendor rather than assuming it.
Wallet passes generally
What is a wallet pass?
A wallet pass is a small structured file that a customer saves to the wallet app on their phone. It contains text fields, artwork and a barcode. Once saved it is stored on the device, so it works with no signal, and it remains addressable by the issuer, so the balance shown today can be different tomorrow without the customer doing anything. Those two properties together are what make it useful for retail loyalty.
What happens if a customer deletes the pass?
Their balance is not lost, because it lives in your loyalty platform rather than on the phone. They can re-save the pass using the same enrolment link or QR code and their record is intact. That said, treat deletion as a serious signal: unlike an unopened email, a deleted pass ends the channel until the customer chooses to re-enrol, so it is worth tracking deletions after every campaign.
Can a wallet pass hold a coupon as well as a loyalty card?
They are usually better as separate passes. Apple provides a coupon pass style described as appropriate for coupons, special offers and other discounts, and Google provides an offer class and object with a redemption channel setting. Keeping the coupon separate means redeeming or expiring it does not disturb the customer's loyalty card, which should be a permanent object rather than a campaign artefact.
Do wallet passes work without an internet connection?
Yes. Once installed, the pass is stored on the device, so the card and its barcode are available with no signal. What requires connectivity is updating the pass, so a balance change made while the customer is offline will appear when their device next syncs. For a shop with poor mobile reception, this is a significant practical advantage over a web-based loyalty page.
Can customers tap their phone instead of scanning?
Technically yes, but it requires certification on both platforms. Apple's Value Added Services protocol requires an NFC certificate from Apple, a VAS-certified terminal, and point-of-sale software that supports VAS modes. Google's Smart Tap requires certification, an eight-digit collector ID, a key exchange, and specific fields enabled on the loyalty class and object. A barcode requires none of this and works with an ordinary 2D-capable scanner.
Notifications
How do retailers use push notifications?
The messages that work in retail are narrow: a reward earned, a reward about to expire, a tier achieved, a milestone reached, an event or appointment reminder, a genuinely relevant new collection or restock, member-only access, and a win-back for lapsed customers. The message that does not work is a general promotion sent to everyone, which is also the message most likely to cause a customer to delete the pass.
How often should a retailer send notifications?
Transactional updates such as a balance or tier change can follow every qualifying event, because the customer caused them and they are confirmations rather than marketing. Deliberate broadcasts should sit at roughly two to four per member per month at most, always segmented, never to the entire base, and never outside trading hours. Google's documented cap of three push-triggering messages per 24 hours makes part of this a technical limit as well as a matter of taste.
What is notification fatigue?
It is the decline in attention and tolerance that follows repeated low-value messages. In email it produces silence: the customer stops opening and the list quietly rots. In wallet marketing it produces deletion, which is worse, because the channel ends permanently and the customer must re-enrol from scratch. That asymmetry should govern the whole send strategy.
What is the best notification a retailer can send?
The balance update itself. On both platforms the pass changing on the customer's phone is already a notification, it is caused by something the customer just did, it is true only for them, and it consumes no campaign budget. Reserve deliberate broadcasts for the things a balance cannot say: an expiring reward, a new collection, an event, or a closure.
Do wallet notifications require the customer's consent?
Saving a pass is an affirmative action by the customer, but you should still treat enrolment consent and marketing consent as two separate permissions and record both. Requirements vary by jurisdiction. The ICO, for example, states that you must have consent to send electronic mail marketing to individual subscribers unless a soft opt-in applies, and its definition of electronic mail expressly includes SMS. Confirm your own position with a qualified adviser.
Geofencing and location
What is geofencing in retail?
Geofencing means triggering an action when a device enters, dwells in or leaves a defined geographic area, such as the area around a store. In retail it is used to surface a loyalty card when a customer is nearby, or in some implementations to send a message. It depends on the operating system's location services and on the permission the customer has granted, which is why what is possible differs sharply between platforms.
How accurate is retail geofencing?
Less accurate than most campaign plans assume. GPS.gov states that GPS-enabled smartphones are typically accurate to within a 4.9 metre radius under open sky and that accuracy worsens near buildings, which is precisely where shops are. Google's own geofencing documentation recommends a minimum geofence radius of 100 to 150 metres, notes that accuracy with Wi-Fi available is usually 20 to 50 metres, and warns that if Wi-Fi is turned off an app might never receive geofence alerts. In a town centre, a 100 metre radius covers your competitors as reliably as it covers you.
Is geofencing legal for retail marketing?
It is lawful where it is done with a proper lawful basis and genuine consent, and it is heavily scrutinised where it is not. Location data is named in GDPR Article 4(1) among the identifiers that make information personal data, and consent under Article 4(11) must be freely given, specific, informed and unambiguous. In the United States the FTC has brought a series of actions against location data companies since 2024, including X-Mode/Outlogic, InMarket, Gravy Analytics and Venntel, Mobilewalla and, in May 2026, Kochava. The FTC's guidance puts the principle plainly: consent to one use without an explanation of other uses is no consent at all. None of this is legal advice.
Should a small shop use geofencing?
Usually not as a first move. If your customers are local, they pass your door constantly, so a naive geofence triggers repeatedly and pointlessly at exactly the people you most want to keep. Passive relevance, where the card simply surfaces on the lock screen when it is useful, is a better starting point: it costs the customer nothing, requires no message, and carries far less privacy weight. Behavioural signals such as home store, visit frequency and category history are usually more accurate for targeting than device location anyway.
Point of sale and customer data
Do I need to change my POS to run a loyalty program?
Almost never. A point-of-sale system is the system of record for transactions; a loyalty platform is the system of record for customers. Those are different jobs and they do not have to live in the same product. The simplest architecture requires no software connection to the till at all: the customer shows a barcode, a staff member scans it, and the loyalty platform records the visit. Nothing about the point-of-sale system changes.
How does the till know a customer is a member?
Generally it does not; something else tells the loyalty platform. There are four common methods: scanning the barcode or QR code on the customer's wallet pass, looking up a phone number or member ID, an NFC tap on a certified terminal, or matching against an online order record. This is the customer identity layer, and it is the part most often overlooked when a program is planned.
What customer data should a retailer collect?
The minimum your reward rule needs. A pseudonymous member identifier is often enough for a stamp or visit card. Add a first name for personalisation if you want it, day and month of birth only if you run a birthday reward, and one contact channel with recorded consent. Itemised purchase histories and device location carry real privacy weight and should only be collected with a clear, stated purpose. Under UK and EU GDPR, personal data must be adequate, relevant and limited to what is necessary.
Can I export my customer data?
You should verify this before you enrol a single customer, not afterwards. Ask three specific questions: is export self-service or a support request, is it in an open format such as CSV or JSON, and does it include loyalty balances or only customer records? A first-party data asset you cannot take with you is not an asset you own; it is one you are renting.
What happens to loyalty if I change POS or loyalty vendor?
It depends entirely on where the customer identifier, the balances and the saved card live. If they live outside the till, changing point-of-sale system is a reconnection exercise and the customer keeps the same card with the same balance. If they live inside a point-of-sale vendor's own loyalty feature, a change of system is usually a program restart. Changing loyalty vendor is a separate question: balances can generally be exported and imported, but a saved wallet pass normally points at the platform that issued it, so ask each vendor directly whether an existing pass can be re-pointed or whether customers must save a new one.
Omnichannel
What is omnichannel retail marketing?
It is treating a customer's store and online activity as one relationship with one identity, one balance and one history, rather than as two separate customers. The practical test is simple: if the same person buys from your website and then walks into your shop, does anything in either system know that? In most retail businesses the honest answer is no, which means the customer is acquired twice, marketed to twice, and understood half as well.
How do I connect in-store and online loyalty?
Through a single customer identity that works in both places. A wallet card is unusually good at being that identity, because it is issued digitally, carried physically and scanned at a counter. Issue it with online orders, offer it at the till, and make sure both routes create or match the same member record rather than two. The click-and-collect handover is the single best moment to do this, and it is almost universally unused.
Can a wallet card be issued from my ecommerce store?
Yes. The usual routes are a link or button on the order confirmation page, a link in the confirmation email, and a QR code printed on the packing slip. The order confirmation page is the strongest of the three, because the customer has just paid and is at maximum willingness to accept something from you. Google's web save link is a signed JSON Web Token, with a documented safe length of 1,800 characters.
Should online and in-store rewards be the same?
In most cases yes, because the customer experiences one brand and will notice any inconsistency as unfairness. The exception is where the economics genuinely differ, such as when delivery costs materially change the margin on an online order. If you do differentiate, make the difference explicit and easy to explain rather than leaving customers to discover it at the counter.
Measurement
How do you measure in-store marketing?
Different disciplines need different measures. Merchandising and signage are measured by conversion rate, sales per linear metre and average transaction value. Promotions are measured by uplift, sell-through and realised margin. Loyalty and post-purchase engagement are measured by repeat visit rate, redemption rate, reactivation rate and member share of revenue. The measurement that matters most for the loyalty half is repeat visit rate among members compared with a matched group of non-members.
What KPIs should a retail store track?
Pick one primary metric and no more than three supporting ones, or nobody will review any of them. For the program overall, the primary metric is repeat visit rate among members. Good supporting metrics are wallet save rate, scan rate at the till, redemption rate and member share of revenue. Passes issued is not a KPI: it measures your counter script, not whether anybody came back.
How long before a loyalty program shows results?
Long enough for a typical customer to complete two or three visit cycles. For a bakery that is weeks; for a homewares retailer visited three times a year it can be twelve months or more. Judging a program before one full reward cycle has completed tells you about enrolment, not about retention. What you can and should judge early is adoption: save rate, scan rate and whether the counter workflow is holding up at peak.
Why do before-and-after loyalty comparisons overstate results?
Because of self-selection. The customers who join a loyalty program tend to be the ones who were already visiting most often, so some of the apparent lift would have happened anyway. The rigorous fix is to compare enrolled members against a matched group of non-enrolled customers with similar prior behaviour over the same period, rather than comparing members against their own past. If you cannot do that, present the figure as an estimate with a wide error bar and say so.
Cost and implementation
How much does retail wallet loyalty cost?
The subscription is usually the smallest line. A full cost model has nine components: software, design, any point-of-sale integration, implementation, staff training, promotion, customer acquisition, maintenance, and the actual margin cost of rewards redeemed. In most small programs the reward cost is the largest single line, which is why comparing vendors on monthly price alone is misleading. Value redeemed rewards at marginal cost rather than retail price, and model the plan you will need in month 24 rather than the one you need this week.
How long does it take to launch a retail wallet loyalty program?
With no integration, using a wallet pass identified by barcode, a functional program can be live in a day, and a well-tested one in about four weeks including baseline measurement, staff training and a soft launch. If you want automatic earning driven by a point-of-sale integration, add whatever partner approval and certification timeline that vendor requires, which can run to months. Start at the loosest coupling that supports your mechanic and tighten only when a measured problem justifies it.
28. Methodology, EEAT, author and sources
This guide is published by PushNotice, a company that sells a product in the category it describes. Every platform claim comes from Apple's or Google's own developer documentation; every statistic comes from a named primary source with the period and publisher stated; every example is labelled as hypothetical. There are no invented statistics, no customer case studies and no claimed results anywhere in this article.
Disclosure and conflict of interest
PushNotice publishes this article and PushNotice sells a wallet marketing platform that can act as the customer engagement layer of a retail loyalty program. That is a direct commercial interest and it should shape how you read us. Four things we have done to keep the guide useful anyway: we state plainly where signage, email, SMS and apps beat wallet; we name four categories of retailer who should not use wallet loyalty at all; we disclose our bias in the benchmark and score our own category down on the criteria where it is genuinely weaker; and we state explicitly that we publish no native point-of-sale integration rather than implying one.
Source policy
Where a technical claim is made about Apple Wallet or Google Wallet, it comes from Apple's or Google's own developer documentation. Where a statistic appears, it comes from a named primary source — the US Census Bureau, the Bureau of Labor Statistics, the Federal Reserve Bank of Atlanta, Pew Research Center, the FTC, the ICO, GPS.gov or a named research publisher with a stated methodology. We have not cited third-party SEO articles, affiliate roundups or unattributed statistics listicles for any factual claim.
Fact-checking approach and what is deliberately absent
Every verifiable claim in this guide was checked against a primary source in August 2026 and dated in the text. Where a figure could not be verified, we say so rather than filling the gap. Specifically:
- The "five times cheaper to retain than acquire" claim is not used, because we could not trace it to a primary source. Section 2 explains what we found instead.
- No BOPIS statistic is quoted beyond Adobe's curbside figure with its full caveats, because everything else we found came from paywalled market-sizing vendors or unattributed listicles.
- No SMS engagement benchmark is quoted, because we found no named publisher disclosing both sample period and message volume.
- No competitor pricing is quoted, because loyalty pricing changes without notice and varies by country; any table here would be stale before it was useful.
- No claim is made that Google Wallet does or does not work on iPhone, because Google's documentation does not state it either way.
- No numeric Apple notification cap is quoted, because Apple publishes none. The accurate phrasing is that Apple publishes no limit.
- No "up to N featured actions" or per-feature OS version claims are made about recent Apple Wallet additions, because those numbers appear only in conference session material and not in reference documentation.
Editorial review
This guide was reviewed by the PushNotice Editorial Team for factual accuracy, for fair treatment of competing channels and approaches, and specifically to confirm that no capability is claimed for PushNotice that is not published on pushnotice.io. Dotted placeholder markers in the text indicate items to confirm against live product data before publication.
Update policy and version history
This page is reviewed at least quarterly, and whenever Apple or Google materially change wallet loyalty capabilities, or a cited statistical source publishes a new release. Because several figures here are quarterly — the Census e-commerce share in particular — expect those to move. The canonical URL always holds the current version. Corrections are welcome via the PushNotice contact page and are made in the text rather than silently.
- v1.0 — 20 August 2026. First publication. Platform documentation, Census Q2 2026 e-commerce release, Pew mobile data, Atlanta Fed payments diary, FTC enforcement record and ICO guidance all checked on this date.
About the author
Sajid Ali is the Founder and CEO of PushNotice, where he leads the company's category strategy and brand voice around wallet marketing. His focus areas are wallet marketing, retail marketing, customer retention, digital loyalty and brand strategy. Author profile: pushnotice.io/blog/authors. LinkedIn: linkedin.com/in/sajid-ali-wajid.
Reviewed by the PushNotice Editorial Team, which checks platform claims against primary documentation, verifies that competing approaches are represented fairly, and confirms that recommendations follow buyer fit rather than commercial preference.
Cite this guide
- APA: Ali, S. (2026). In-Store Marketing: Wallet Loyalty for Retail Stores. PushNotice. https://pushnotice.io/blog/in-store-marketing
- MLA: Ali, Sajid. "In-Store Marketing: Wallet Loyalty for Retail Stores." PushNotice, 20 Aug. 2026, pushnotice.io/blog/in-store-marketing.
Related guides
Sources
All sources below are primary — platform documentation, government statistical releases, regulator publications, or named research with a stated methodology. Retrieved and checked 20 August 2026. Platform behaviour, statistics and regulatory guidance all change; verify at the source before relying on any specific figure.
Apple — Wallet and PassKit
- Apple Developer — Pass (pass styles including
storeCardandcoupon;webServiceURL; up to 10 locations;relevantDates). developer.apple.com/documentation/walletpasses/pass - Apple Developer — Adding a web service to update passes ("an empty JSON dictionary for the payload"; production-only push). developer.apple.com — updating passes
- Apple Developer — PassFieldContent ("You need to provide a value for the system to show a change notification"). developer.apple.com — PassFieldContent
- Apple Developer — Pass barcodes (supported
PKBarcodeFormatvalues, including the more recent 1D formats). developer.apple.com — pass barcodes - Apple Developer — Showing a pass on the lock screen (10 relevant locations; up to ten beacon UUIDs;
maxDistance). developer.apple.com — lock screen relevance - Apple Developer Library (archived) — Wallet Developer Guide: Creating and Updating Passes (store card and coupon descriptions; "Relevance information is passive… It doesn't present alerts or post notifications"; small and large radius interpretation). developer.apple.com/library/archive — Wallet Developer Guide
- Apple Developer — Loyalty passes (Value Added Services; NFC certificate; VAS-certified terminals; "VAS Only" and "Payment and VAS" POS modes). developer.apple.com/wallet/loyalty-passes
- Apple Developer — What's new in Wallet. developer.apple.com/wallet/whats-new
Google — Wallet Loyalty and Offers APIs
- Google for Developers — How classes and objects work ("Changes made to a Passes Class instance will propagate immediately across all Passes Object instances that reference it"). developers.google.com — classes and objects
- Google for Developers — LoyaltyObject reference (
loyaltyPoints,secondaryLoyaltyPoints,accountId,accountName,notifyPreference, 10-message maximum;locationsdeprecated;merchantLocationsreplaces LatLongPoints). developers.google.com — LoyaltyObject - Google for Developers — LoyaltyClass reference (
rewardsTier,secondaryRewardsTier,programName). developers.google.com — LoyaltyClass - Google for Developers — MerchantLocation reference ("When a user is within a set radius of this lat/long, and dwells there, Google will trigger a notification"). developers.google.com — MerchantLocation
- Google for Developers — Trigger push notifications (maximum of 3 push-triggering messages per 24 hours; the five allow-listed loyalty fields; "Google decides how close a user needs to be"; precise, always-on location requirement; 10 locations per class and 10 per object). developers.google.com — trigger push notifications
- Google for Developers — OfferClass reference (
redemptionChannelvalues). developers.google.com — OfferClass - Google for Developers — Save to Google Wallet on the web ("The safe length of an encoded JWT is 1800 characters"). developers.google.com — save on the web
- Google for Developers — Smart Tap ("You must be certified in order to use this protocol"; 8-digit collector ID; key exchange;
enableSmartTap). developers.google.com/wallet/smart-tap - Android Developers — Geofencing API (100 geofences per app per device user; recommended 100–150 m minimum radius; Wi-Fi accuracy 20–50 m; latency; re-registration after reboot). developer.android.com — geofencing
- Android Developers — Location permissions (approximate vs precise accuracy figures; background location requirements from Android 10 and 11). developer.android.com — location permissions
Retail and consumer statistics
- US Census Bureau — Quarterly Retail E-Commerce Sales, Q2 2026 (released 18 August 2026; e-commerce 17.1% of total retail sales, $340.2bn of $1,986.5bn, seasonally adjusted). census.gov/retail/ecommerce
- US Census Bureau — Advance Monthly Retail Trade Survey (July 2026 retail and food services sales). census.gov/retail/sales
- US Census Bureau — E-commerce general FAQs (definition: e-commerce is classified by where the order was placed). census.gov — e-commerce FAQs
- US Bureau of Labor Statistics — Retail Trade: NAICS 44-45 (employment in retail trade). bls.gov — retail trade
- Pew Research Center — Mobile Fact Sheet (91% of US adults own a smartphone; NPORS, 5,022 US adults, fielded February–June 2025). pewresearch.org — mobile fact sheet
- Federal Reserve Bank of Atlanta — 2025 Survey and Diary of Consumer Payment Choice, published May 2026 (mobile phone payment adoption; in-person share of purchases). Report PDF · series landing page. The Atlanta Fed site restricts automated retrieval; open the PDF directly to verify.
- Adobe Analytics — 2025 holiday shopping season (17.1% of online orders used curbside pickup among retailers offering it; methodology stated). news.adobe.com — 2025 holiday season
- Mailchimp — Email Marketing Benchmarks (all-industry and e-commerce open and click rates; Mail Privacy Protection caveat). mailchimp.com — benchmarks
The retention-cost literature
- Keiningham, Vavra, Aksoy & Wallard — Loyalty Myths, published by Ipsos Loyalty (2005). The source that traces and disputes the "five times cheaper to retain" claim. ipsos.com — Loyalty Myths excerpt
- Reichheld & Sasser — Zero Defections: Quality Comes to Services, Harvard Business Review, September 1990. hbr.org — Zero Defections · Bain summary
Privacy, consent and location regulation
- US Federal Trade Commission — order prohibiting X-Mode Social / Outlogic from selling sensitive location data (January 2024; finalised April 2024). ftc.gov — X-Mode / Outlogic
- US Federal Trade Commission — order banning InMarket Media from selling precise location data (January 2024). ftc.gov — InMarket
- US Federal Trade Commission — actions against Gravy Analytics / Venntel and Mobilewalla (December 2024). ftc.gov — Gravy Analytics / Venntel · ftc.gov — Mobilewalla
- US Federal Trade Commission — settlement with Kochava and subsidiary on sensitive location data (May 2026). ftc.gov — Kochava
- US Federal Trade Commission Business Blog — How location, location, location can lead to enforcement, enforcement, enforcement (January 2024): "'consent' to one use without an explanation of other uses is no consent at all." ftc.gov — business guidance
- EU General Data Protection Regulation — Article 4(1) (location data as an identifier), Article 4(11) (definition of consent), Article 5(1)(c) (data minimisation), Article 6 (lawful bases). eur-lex.europa.eu — GDPR
- UK Information Commissioner's Office — A guide to lawful basis (updated 2 April 2026; now seven lawful bases including recognised legitimate interest). ico.org.uk — lawful basis
- UK Information Commissioner's Office — Electronic mail marketing (consent requirement; definition of electronic mail includes SMS). ico.org.uk — electronic mail marketing
- UK Information Commissioner's Office — Location data under PECR (PECR location rules do not generally cover GPS data from smartphones). ico.org.uk — location data
- European Data Protection Board — Process personal data lawfully (marketing cannot rely on the contract basis). edpb.europa.eu — lawful processing
- GPS.gov — GPS accuracy ("GPS-enabled smartphones are typically accurate to within a 4.9 m (16 ft.) radius under open sky"). gps.gov/gps-accuracy
- Apple Developer — Requesting authorization to use location services (When In Use vs Always; full vs reduced accuracy). developer.apple.com — location authorization
PushNotice
- PushNotice — Plans and pricing. pushnotice.io/#pricing
- PushNotice — Wallet marketing. pushnotice.io/wallet-marketing
- PushNotice — Authors. pushnotice.io/blog/authors
All statistics, platform behaviour and regulatory guidance described in this guide reflect the state of each source in August 2026 and may have changed. Verify at the source before relying on any specific figure. Nothing in this guide is legal, tax or financial advice.