1. What is a distributor loyalty program?
A defined system through which a distributor manages recognition, benefits, qualification and ongoing communication for its trade accounts, in order to change account-level buying behaviour it can name in advance and measure afterwards. It is built from seven components, and the mistake almost every distributor makes is buying the last two before deciding the first three.
Strip the retail vocabulary out and a distributor program has a small number of moving parts. Some are close to universal — a member card carrying the branch phone number and the account number helps everybody and costs nothing. Others, particularly anything that transfers margin to a customer, are the reason Section 6 is the longest section in this guide rather than a footnote.
| Component | Behaviour it targets | Example |
|---|---|---|
| Account qualification | Defines who is in the program and on what basis | Trade accounts in good standing, above a stated annual spend |
| Incentive structure | The earning mechanic — points, rebate, tier, or none | 1% back on eligible spend above a quarterly threshold |
| Tiering | Concentrates benefit on the accounts worth concentrating on | Three status levels with different pricing and service access |
| Reward fulfilment | Converts earned value into something the customer receives | Account credit, prepaid card, merchandise, or a service benefit |
| Account membership | Makes the relationship visible between orders | A digital card carrying account number, tier and branch contact |
| Communication | Delivers status, thresholds, offers and reminders | A message when an account is close to a quarterly threshold |
| Measurement | Establishes whether any of the above changed anything | Repeat-order rate and category breadth against a matched baseline |
Why distributor loyalty is not consumer loyalty
Five differences do all the work, and every design decision later in this guide traces back to one of them.
The buyer is not the beneficiary. In retail, the person who chooses, pays and enjoys the reward is one person. In distribution they are frequently three: a purchasing manager places the order, the business pays for it, and whoever the reward is directed at may be neither. That single structural fact is why merchandise and travel rewards persist in B2B long after retail moved on from them — and it is also the reason those rewards attract scrutiny that a discount does not.
The decision is collective. Gartner, surveying 632 B2B buyers in August–September 2024, reports that buying groups now range "from five to 16 people across as many as four functions." A program that speaks only to the person holding the card is speaking to a minority of the decision.
The margin is thinner and the order is bigger. A distributor's gross margin sets the ceiling on what any reward can cost, and in distribution that ceiling is genuinely low. Three publicly reporting distributors give the honest range: Fastenal reported gross margin of 45.0% of net sales for FY2025, W.W. Grainger 39.1%, and Watsco 27.5% in Q2 2026 — the last of those a single quarter, and down 180 basis points from 29.3% a year earlier. The first two are among the strongest full-year margins in the sector; the third is a quarterly figure in decline. Plenty of distribution runs below all three.
Purchases are already contractual. Most trade accounts already have negotiated pricing, credit terms and sometimes a rebate agreement. A loyalty program is not arriving in empty space — it is being layered on a commercial relationship that already transfers value, which means the first question is not "what should we offer?" but "what are we already paying for this behaviour?"
Customers rarely leave; they drift. Retail churn is usually binary. In distribution the typical failure is share erosion: the account still orders, but half of what it used to. That is why share of wallet, not retention, is usually the honest loyalty measure in this industry.
The entity chain a distributor program actually runs on, from the individual buyer through to the repeat order. It is drawn here rather than at the end because every later framework attaches to one of its layers, and because the two layers teams most often skip — qualification and measurement — are the two that decide whether the rest of it pays.
2. Distributor, wholesaler, reseller, retailer: why the words matter
A distributor buys from manufacturers, holds inventory, adds service, and sells to businesses that use or install the goods. A wholesaler is a broader Census category that includes distributors. A reseller sells the product on largely unchanged. A retailer sells to end consumers. A channel partner is a role in someone else's go-to-market, not a business type. These are used interchangeably in loyalty content, and it produces programs designed for the wrong customer.
This section exists because the practical consequence is expensive. A large share of the "distributor loyalty program" examples circulating online are actually retailer programs for trade customers — Home Depot Pro Xtra, Lowe's Pro, Sherwin-Williams PRO+. Those are excellent programs and there is a great deal to learn from them, and this guide draws on all three. But a retailer serving contractors is running a different business from a distributor serving contractors: different margin structure, different order size, different credit exposure, different competitive set. Copying the mechanic without adjusting for the model is how a program ends up with retail economics on distribution margins.
Within distribution there are two directions of incentive and they are constantly conflated. Manufacturer-to-distributor programs are funded by the manufacturer to move its product through your branches. Distributor-to-customer programs are funded by you to move your customer's behaviour. Several of the largest distributors in North America run only the first kind — they administer incentives they do not fund. Section 9 shows three live examples of each, and it changes how you read the whole competitive landscape.
- Decide first whether you are rewarding the account or the buyer. Almost every downstream design decision follows from that answer, and most programs never make it explicitly.
- Distributor gross margin sets the ceiling on the whole program. Publicly reported figures at three large distributors run from 27.5% to 45.0% — and much of the sector sits below that band.
- In distribution, customers erode rather than churn. Measure share of wallet, not just retention.
3. Why distributors lose repeat customers
Distributors rarely lose accounts outright. They lose share — quietly, one line item and one reorder at a time — usually to price, to a competitor's service, to a rep who moved firms, or to the simple fact that reordering somewhere else was easier. Six causes account for most of it, and only three of them are addressable by a loyalty program at all.
Being honest about which failures a program can fix is what separates a program that pays for itself from one that becomes a line item nobody will defend in the next budget review. Here are the six, with what the evidence actually says about each.
1. Price competition, and the limits of buying loyalty
The reflex response to share erosion is to make leaving expensive — deeper discounts, longer commitments, stickier terms. The research does not support it. Pick and Eisend's meta-analysis in the Journal of the Academy of Marketing Science (2014), drawing on 170 independent samples across 152 manuscripts, found that "switching costs have only a weak negative influence on switching" — and that the link between switching-cost antecedents and perceived switching costs is weaker in B2B settings than in B2C.
Dikcius and colleagues, in Engineering Economics (2019), surveyed 101 B2B service buyers and modelled attitudinal and behavioural loyalty separately. Switching costs had opposite signs on the two: they raised behavioural loyalty (t=2.05, p=0.043) while lowering attitudinal loyalty (t=−2.27, p=0.026). In plain terms: making it hard to leave keeps the orders coming and makes the customer like you less. That is a fragile position — it holds exactly until a competitor removes the friction.
Caveat honestly: n=101, single country, and a sample weighted toward B2B services rather than distribution. Directional, not definitive.
2. Rep dependency
This is distribution's oldest retention mechanism and its least examined risk. Palmatier and colleagues' meta-analysis in the Journal of Marketing (2006) — 94 studies, 38,000 relationships — found that relationship marketing is more effective "when relationships are more critical to customers (e.g., service offerings, channel exchanges, business markets)," and that relationships built with an individual person rather than the selling firm are more effective still.
Read that twice, because it cuts both ways. It is the strongest available evidence that the rep relationship really is the loyalty mechanism in distribution — and the strongest available warning that a firm whose loyalty lives entirely inside one person's phone contacts is one resignation away from losing it. A program's most defensible job here is not to replace the rep. It is to give the account a relationship with the firm that survives the rep changing.
3. Reordering somewhere else was easier
The most underrated cause of share loss in distribution is friction. McKinsey's 2026 B2B Pulse, surveying nearly 4,000 decision-makers across 13 countries, found that 71% of organisations now offer e-commerce, that roughly a third of total revenue flows through digital channels, and that 73% of buyers are comfortable placing orders above $50,000 online. Meanwhile a survey of senior B2B procurement leaders conducted by B2B Online Insights with OroCommerce, reported by Modern Distribution Management in June 2025, found that only 33% of buyers said they can reorder with a single click — the rest start from scratch each time. (Vendor-sponsored survey, no published sample size — treat as directional.)
4. Rebate complexity
A rebate a customer cannot forecast is not an incentive; it is a surprise. When an account cannot tell you, in the middle of a quarter, how far it is from its next threshold, the rebate is doing none of the behavioural work it was designed for and all of the margin damage. This is the single most fixable problem on this list, and it is the one place where a wallet card earns its keep with no argument required — a threshold you can see is a threshold you can act on.
5. Weak customer visibility
Distributors typically know their accounts by revenue and almost never by profit. The best evidence for how badly those two diverge comes from Texas A&M's Industrial Distribution program: Narayanan, Lawrence, Rao and Krishnadevarajan's customer stratification study, using data collected from November 2005 to October 2006, found in one branch analysis that the top 25 customers — about 10% — contributed 80% of sales and 75% of net profit, while a segment of ten "service-drain" accounts generated 18% of sales and 14% of gross margin but only 7% of net profit. The NAW Institute later ran the same stratification approach across 68 wholesaler-distributors, evaluating accounts on buying power, loyalty, profitability and cost-to-serve.
A widely circulated "40% of accounts produce 130–140% of profits, 40% destroy 30–40%" split appears in distribution content constantly. It traces to a single consultant's article attributing the pattern to his own engagements, with no underlying study cited. It may well be directionally right. It is not a statistic, and quoting it as one in front of a CFO is a bad trade. Use the Texas A&M numbers, which have a method attached.
6. Missed promotions and silent programs
The most common failure mode of a distributor loyalty program is not that customers reject it. It is that they forget it exists. Points balances sit in a portal nobody logs into, threshold notices arrive by email into a purchasing inbox that receives four hundred emails a day, and the quarterly statement lands after the quarter it could have influenced. This is a communication-architecture problem wearing a loyalty-program costume, and it is the one this guide's wallet sections are actually about.
It helps with three of the six: rebate complexity (a visible, current threshold), missed promotions (a channel that is not an inbox), and rep dependency (a persistent relationship with the firm, carrying the branch number, that survives a rep change).
It does nothing at all for the other three. A wallet pass will not win a price war, will not fix a broken reorder flow, and will not tell you which accounts are unprofitable. Any vendor — including us — suggesting otherwise is selling you a card and calling it a strategy.
- Switching costs are a weak retention lever in B2B and they buy compliance rather than preference. Design for preference.
- The rep relationship is real and evidence-backed — which is exactly why the firm needs a relationship of its own alongside it.
- Three of the six causes of share loss are program-addressable. Be explicit about which three, or the program will be blamed for the other three.
4. Distributor loyalty program models
Eight models are in live use in distribution, and they differ far more in economics and administrative load than in mechanics. Volume, frequency, tiered, rebate, category-growth, order and channel incentives, membership benefits, and hybrids. Most distributors should run one primary model with at most one accelerator layered on it — the failure mode is not choosing wrong, it is choosing three.
Eight models, six questions each, split across two tables so both stay readable on a phone. The columns that matter most are economic mechanism and margin exposure — the two every vendor comparison omits, and the two your finance director will ask about first.
| Model | Goal | Behaviour influenced | Best use case |
|---|---|---|---|
| 1. Volume-based | Grow total spend per account | Consolidation of spend already going elsewhere | Fragmented categories where accounts buy from three or four suppliers |
| 2. Frequency-based | Increase order cadence | More, smaller orders instead of fewer batched ones | Consumables and replenishment lines with predictable burn rates |
| 3. Tiered | Concentrate benefit on accounts worth keeping | Movement up a threshold; retention at the top | Portfolios with a wide spread of account sizes |
| 4. Rebate-based | Buy contractually committed volume | Hitting a defined purchase target in a defined period | Large accounts with negotiated agreements and a forecastable base |
| 5. Category-growth | Widen the range an account buys | First purchase in a category the account buys elsewhere | Broad-line distributors where most accounts buy narrowly |
| 6. Order & channel | Shift transactions to lower-cost channels | Ordering online or in-app instead of by phone or counter | Distributors with a live e-commerce channel and high cost-to-serve |
| 7. Membership & benefit | Deliver value without issuing a currency | Preference, driven by pricing, credit and service access | Trade suppliers whose real advantage is service, stock or terms |
| 8. Hybrid | Combine a base model with one accelerator | Whatever the base model targets, amplified selectively | Mature programs with the data to run them, and only those |
| Model | Economic mechanism | Margin exposure | Complexity | Wallet compatibility |
|---|---|---|---|---|
| 1. Volume-based | Pays a rate on qualifying spend, often including the base | High if linear — see Section 6 | Low | Strong — a running total and a threshold |
| 2. Frequency-based | Rewards order count, decoupled from order value | Moderate; can subsidise splitting orders | Low | Strong — a simple visible count |
| 3. Tiered | Fixed benefit blocks at defined thresholds | Moderate; steps out at each threshold | Medium | Strongest — tier is the ideal wallet field |
| 4. Rebate-based | Contractual entitlement; reduces reported revenue | High and accrued in advance | High | Partial — show progress, never the accrual |
| 5. Category-growth | Pays only on incremental category penetration | Low — narrow by construction | Medium | Partial — needs product-level context |
| 6. Order & channel | Trades reward cost against cost-to-serve saved | Low — often self-funding | Medium | Strong — the card is the channel prompt |
| 7. Membership & benefit | No currency issued; cost sits in price and service | Lowest — no liability to accrue | Low | Strong — status, benefits, contact |
| 8. Hybrid | Base rate plus a targeted multiplier | Depends entirely on the base | Highest — and the usual reason programs fail | Depends on the base model |
The eight models in practice
1. Volume-based. The default, and the one most often designed badly. The account earns a rate on qualifying spend. Whether that rate applies to all spend or only spend above a threshold is not a detail — on the worked example in Section 6 it is the difference between netting $4,600 and netting $600 — and, if growth lands at 5% rather than 10%, between still netting $2,300 and losing $1,700 while every engagement metric looks excellent throughout. Section 6 works the arithmetic.
2. Frequency-based. Rewards the number of orders rather than their value. Genuinely useful for consumables, where a customer who orders weekly is far less likely to shop around than one who orders quarterly. The failure mode is obvious once stated: an account that splits one order into four to earn four rewards has cost you three pick-and-pack cycles and given you nothing. Cap it, or key it to distinct order dates.
3. Tiered. The most natural fit for distribution, because distribution portfolios are steeply skewed and tiering is how you stop spending uniformly on a non-uniform customer base. Tiers also survive the wallet better than any other model — a status field is the single most legible thing you can put on a card. The discipline is that a tier must carry benefits the customer can name, or it becomes a label.
4. Rebate-based. A contractual entitlement rather than a marketing program, and it belongs to finance as much as to sales. Section 7 treats it properly, including why the accounting standards put it in a genuinely different category from a loyalty point.
5. Category-growth. Pays only when an account buys something it did not buy from you before. It is the most economically defensible model on this list because incrementality is built into the mechanic rather than assumed after the fact — you cannot accidentally pay for volume you already had. It is under-used because it needs product-level data most programs never wire up.
6. Order and channel incentives. The quiet winner. If a phone or counter order costs meaningfully more to process than a web order, an incentive that moves the transaction pays for itself out of cost-to-serve rather than out of gross margin. This is not theoretical: the two clearest examples in live trade programs are both channel-shift instruments wearing loyalty badges (see Section 9).
7. Membership and benefit. No points, no accrual, no liability. Value is delivered as pricing, credit access, service, training or priority. It is the most under-represented model in loyalty content and the one several serious trade suppliers have deliberately chosen — again, Section 9.
8. Hybrid. A base model plus one accelerator — for example a tier structure with a supplier-funded multiplier on selected lines. Hybrids are where the best programs and the worst programs both live. The rule that separates them: one base, one accelerator, and a written reason for the accelerator. Three stacked mechanics produce a program nobody in your own branch network can explain, and a program your own people cannot explain is not a program.
Ask what you want an account to do next week that it would not otherwise have done. If the answer is "spend more," you have not answered — spend more on what, ordered how, in which category? Models 5 and 6 exist because those specific answers are cheaper to buy than the vague one, and easier to prove afterwards.
5. Points, rebates, tiers, coupons and rewards
Seven reward mechanics are in common use in distribution and they are not interchangeable. Points are a discretionary currency you issue. Rebates are a contractual entitlement the customer earns. Coupons are a time-boxed price reduction. Tier benefits are conditional access. Exclusive pricing is a permanent margin decision. Order rewards buy a channel or a cadence. Category incentives buy range. They differ in who bears the cost, when the cost lands, and whether you carry a liability for it.
Seven mechanics, five dimensions. The column that decides most arguments is when the cost lands, because a mechanic that costs you in twelve months feels free for eleven of them — and that is precisely how programs get approved and then get killed.
| Mechanic | What it is | When the cost lands | Carries a liability? | Principal risk |
|---|---|---|---|---|
| Points | A discretionary currency issued on qualifying spend | On redemption, which may be years away | Yes — unredeemed points | Breakage flatters early results; redemption arrives later |
| Rebates | Contractual entitlement earned against stated criteria | Accrued as revenue is recognised | Yes — accrued, and estimated in advance | Pays for volume you would have had anyway |
| Coupons | A time-boxed reduction on a defined purchase | At the moment of use | No, if it expires | Trains accounts to wait for the next one |
| Tier benefits | Conditional access unlocked by reaching a threshold | Continuously, while the tier is held | Not usually, but hard to withdraw | Tier inflation — everyone reaches the top tier |
| Exclusive pricing | A permanently lower price for qualifying accounts | On every single order, immediately | No | It is a margin decision, not a marketing one |
| Order rewards | A benefit tied to how or when an order is placed | At order, offset by cost-to-serve saved | No | Rewarding a behaviour the account had already adopted |
| Category incentives | A benefit on the first or growing purchase in a range | At order, on genuinely new volume | No | Needs product-level data most programs lack |
Points versus rebates: the distinction that matters most
These two get treated as flavours of the same thing. They are not. The clearest way to see it is through the accounting standards, which put a rebate in a specific and demanding category.
IFRS 15.51 names rebates directly as a source of variable consideration: "An amount of consideration can vary because of discounts, rebates, refunds, credits, price concessions, incentives, performance bonuses, penalties or other similar items."
IFRS 15.70, and ASC 606-10-32-25 in materially identical wording, require that "an entity shall account for consideration payable to a customer as a reduction of the transaction price and, therefore, of revenue unless the payment to the customer is in exchange for a distinct good or service."
IFRS 15.56 then constrains the estimate: variable consideration is included only to the extent that "it is highly probable that a significant reversal in the amount of cumulative revenue recognised will not occur." ASC 606 applies the same constraint at a lower threshold — US GAAP says "probable," not "highly probable." The divergence is deliberate, and it matters if you report under both.
Note the precision: the consideration payable to a customer paragraphs do not enumerate "rebates," "coupons" or "volume discounts" as separate line items — these fall under the broader headings of cash amounts and credit or other items. Do not write "ASC 606 says rebates are…"; write that a rebate is consideration payable to a customer, which the standard requires to reduce the transaction price.
The commercial consequence is the interesting part. A rebate is an entitlement you must estimate and accrue before you know whether the customer will earn it, and it reduces reported revenue rather than sitting in a marketing budget. A points currency is a discretionary promise you issue and can, within the terms you wrote, change. Those are different instruments with different governance, different owners inside the business, and different failure modes. Distribution runs predominantly on rebates rather than points for exactly this reason: at the account sizes involved, the customer wants a contractual entitlement, not a currency.
Points programs look cheaper than they are in their early years, because points are issued long before they are redeemed. We are not going to tell you for how long: we could not find a figure with a method attached, and neither can anyone quoting one. Trade programs handle this with expiry rules that would be considered aggressive in retail. Ferguson's PRO Plus terms state that points are valid for the life of the program, but that "if you do not earn points for a period of 365 days, all unredeemed points will be forfeited." NAPA's Installer Rewards states that award dollars earned during 2026 "must be redeemed by 2/28/2027 when they will expire." Those clauses are not customer-hostile fine print; they are how a B2B points liability is kept finite. Write yours before launch, not after the balance builds.
The mechanic that is not a reward at all
Exclusive pricing deserves separating from the rest of this table because it is categorically different. It is not a program benefit that finance can model as a marketing cost — it is a permanent reduction in the gross margin on every order that account places, forever, and it is extraordinarily difficult to withdraw once given. That does not make it wrong. For a great many distributors it is the most efficient thing they can offer, because it is what the customer actually wants and it carries no liability, no fulfilment cost and no administration. It just needs to be recognised for what it is: a pricing decision with a loyalty label on it.
- Points and rebates are different economic objects. One is a discretionary currency you issue; the other is a contractual entitlement that reduces reported revenue and must be estimated in advance.
- Any points program needs an expiry rule written before launch. Live trade programs use 365-day inactivity forfeiture and hard annual redemption deadlines.
- Exclusive pricing is a margin decision. Treat it as one, and price it against the alternatives in Section 6.
6. How to design the economics of a distributor loyalty program
A distributor loyalty program pays for itself when the gross profit on behaviour the program actually caused exceeds the total cost of the program — not the cost attributable to that behaviour, the total. That single distinction is why a 2% rebate can quietly cost 22% of the revenue it created, and why two programs with identical headline rates can differ by nearly eightfold in what they contribute. This section works the arithmetic in full.
Six steps, one test and one formula. Built to be worked through in a spreadsheet before a program is approved, and to be re-run against actuals afterwards. The steps are ordered deliberately: gross margin comes before reward design, not after, because margin is the constraint and everything else is a choice inside it.
The chain, in order
- Customer value. Start with the account's current annual spend and its gross margin, not its revenue. Two accounts of identical size can differ by a factor of two in what they are worth to you, and a program that treats them identically is transferring margin from the profitable one to the unprofitable one.
- Desired behaviour. Name it in one sentence that includes a number and a time period. "Grow spend" is not a behaviour. "Move this account from four orders a quarter to eight, in the maintenance-consumables category, within two quarters" is.
- Incremental revenue. The revenue that behaviour produces above what would have happened anyway. This is the number everybody skips, and every subsequent number depends on it.
- Gross margin. Apply the margin on that specific incremental volume — which is often lower than your blended margin, because growth frequently comes in competitive lines.
- Maximum reward cost. The ceiling, derived below.
- Program economics. Total cost against total incremental gross profit, including administration, fulfilment and the cost of the people who run it.
Incremental gross profit > total incentive cost.
Note both sides carefully. The left side counts only behaviour the program caused. The right side counts everything the program paid out, including payments on volume you would have received regardless. Most distributor programs are evaluated on a version of this test with the same word missing from both sides, and pass comfortably.
Maximum total payout = Incremental revenue × Gross margin % × (1 − the share of incremental gross profit you intend to keep)
If an account's incremental revenue is $20,000 at a 25% margin, incremental gross profit is $5,000. If you intend to keep half of that, your entire program payout on that account must stay under $2,500 — not $2,500 on the incremental portion, $2,500 in total. Expressed as a rate on incremental revenue, the ceiling is gross margin × the share you give away: 12.5% here. That number is much larger than any rebate rate you would ever quote, which is exactly why the design question is never "what rate?" — it is "on what base?"
The Incremental Rate Trap
Here is the whole argument in one worked example. An account currently buys $200,000 a year at a 25% gross margin — $50,000 of gross profit. You want 10% growth: $20,000 of incremental revenue, $5,000 of incremental gross profit. You offer 2%. Two ways to structure it.
| Design A — linear 2% on all qualifying spend | Design B — target 2% on spend above $200,000 | |
|---|---|---|
| Account spend after growth | $220,000 | $220,000 |
| Incremental revenue | $20,000 | $20,000 |
| Incremental gross profit @ 25% | $5,000 | $5,000 |
| Total program payout | $4,400 | $400 |
| Net contribution | $600 | $4,600 |
| Effective rate on incremental revenue | 22.0% | 2.0% |
| Share of incremental gross profit consumed | 88% | 8% |
Design A pays 22% of the revenue it created, against a gross margin of 25%. It is not quite a loss — but it is a rounding error away from one, and it will be presented internally as a 2% program. The headline rate and the effective rate differ by a factor of eleven, and nothing in the program's own reporting will ever surface the difference.
Now make the growth 5% instead of 10%, which is a perfectly ordinary outcome:
Spend rises to $210,000. The linear payout is $4,200 against incremental gross profit of $2,500 — a net loss of $1,700, and an effective rate of 42% of incremental revenue on a 25% margin. The target rebate pays $200 and contributes $2,300. The program that loses money and the program that makes money are, on paper, the same 2% offer.
The Linear Rebate Break-Even
That fragility can be stated exactly. For a linear rebate paid on all qualifying spend, with rebate rate r and gross margin m, the account growth rate g* at which the program merely breaks even is:
g* = r ÷ (m − r)
Below g*, the program destroys gross profit. A target rebate paid only on volume above a baseline has no equivalent threshold: it is profitable at any positive growth rate, provided the rebate rate is below the gross margin on the incremental volume.
| Gross margin on incremental volume | 1% linear rebate | 2% linear rebate | 3% linear rebate |
|---|---|---|---|
| 20% | 5.3% | 11.1% | 17.6% |
| 25% | 4.2% | 8.7% | 13.6% |
| 30% | 3.4% | 7.1% | 11.1% |
| 40% | 2.6% | 5.3% | 8.1% |
The practical reading: a 3% linear rebate on a 20%-margin line needs the account to grow 17.6% to break even. Very few distribution accounts grow 17.6% in a year for any reason, let alone because of a rebate — which means that program was a price cut from the day it was signed. That may still be the right commercial call. It should just be made as a pricing decision, by the people who make pricing decisions.
Terry Taylor's "Supply Chain Coordination Under Channel Rebates with Sales Effort Effects" (Management Science, 2002) reaches the same conclusion formally. Target rebates — paid on volume above a threshold — can coordinate a distribution channel and produce a win–win for both parties. Linear rebates — paid per unit on everything — cannot: Taylor shows that a channel-coordinating linear rebate "is not implementable, as it requires the manufacturer to be worse off." The mechanism is precisely the one in Table 4: a threshold avoids paying for infra-marginal volume that would have moved anyway.
Taylor's model concerns manufacturer-to-reseller rebates rather than distributor-to-customer programs. The structural logic transfers; the specific parameters do not.
Qualification periods, and the thing they quietly do
The length of the qualifying period is a real design lever and it is usually chosen by administrative convenience. Shorter periods — monthly or quarterly — produce more frequent behaviour changes, more communication moments and faster feedback, and they suit consumables. Longer periods — annual — suit large negotiated accounts and reduce administrative load, but they concentrate all the behavioural pressure into the final weeks, which is how you end up with a fourth-quarter order surge that is pulled-forward volume rather than new volume.
The specific trap: an annual threshold produces a December that looks like growth and a January that looks like collapse. If your program has a period boundary, measure across it, not up to it.
Incrementality cannot be measured without a counterfactual, and in distribution you can usually build one cheaply. Three options, in descending order of rigour: (1) hold out a matched control group of accounts and do not enrol them; (2) stagger the launch across branches or regions and compare enrolled and not-yet-enrolled cohorts; (3) at minimum, capture a twelve-month pre-launch baseline per account and compare against the same accounts' trend, not against a flat line. Option three is weak — it cannot separate the program from the market — but it is infinitely better than the usual approach, which is to report enrolled-account growth and call it program performance.
- The Loyalty Economics Test: incremental gross profit must exceed total program cost, not the cost attributable to incremental volume.
- A linear rebate has a break-even growth rate of r ÷ (m − r). At 3% on a 20% margin that is 17.6% growth — a threshold most accounts will never clear.
- Target rebates are profitable at any positive growth rate provided the rate is below the incremental margin. This is both arithmetic and a published result in Management Science.
- Choose the qualification period deliberately, and measure across period boundaries so pulled-forward volume is not counted as growth.
7. Distributor loyalty program vs rebate program
A loyalty program is a broad, ongoing relationship and engagement system. A rebate program is a narrow financial instrument: a contractual entitlement to a portion of spend, earned against defined criteria, accrued in advance and treated as a reduction of revenue rather than a marketing cost. They answer different questions, sit with different owners, and work best together — the rebate carries the economics, the loyalty program carries the visibility.
| Dimension | Loyalty program | Rebate program |
|---|---|---|
| What it is | An ongoing engagement and recognition system | A contractual financial entitlement |
| Basis | Programme terms you set and can amend | An agreement, often negotiated per account |
| Trigger | Qualifying activity as defined by the program | Meeting a stated purchasing or performance criterion |
| Typical owner | Marketing, or commercial | Finance and sales leadership jointly |
| Accounting posture | Cost recognised on redemption; liability for unredeemed value | Consideration payable to a customer — reduces the transaction price and therefore revenue, estimated in advance |
| Visible to the customer | Usually, and continuously | Often only at settlement, which is the core problem |
| What it is good at | Attention, frequency, channel shift, range, engagement | Committing large, forecastable volume |
How they work together
The productive relationship is not either/or. In most mid-sized and large distributors the rebate already exists — it was negotiated, it is in the contract, and it is where the real money moves. What is almost always missing is the layer that makes it legible.
A rebate that a customer discovers at settlement has already failed at the only job an incentive has: changing behaviour while the behaviour is still changeable. An account that can see, in week seven of a thirteen-week quarter, that it is $18,000 short of its threshold, has a decision to make. An account that finds out in week fourteen has a complaint to make. This is the single most useful thing a distributor loyalty layer does on top of an existing rebate structure, and it requires no change to the rebate itself.
Show the customer progress toward a threshold — spend to date, distance to go, period end date. Do not show a calculated accrual, a projected payout figure, or anything a customer could reasonably read as a confirmed entitlement before the period closes and the criteria are verified. A wallet field is a communication surface, not a statement of account, and the gap between "you are $18,000 from your target" and "you have earned $4,400" is the gap between a helpful nudge and a dispute.
8. Distributor loyalty program vs incentive program
"Incentive program" is the widest term of the three: any mechanism designed to influence a specific commercial behaviour. A loyalty program is one kind of incentive program — the kind aimed at an ongoing customer relationship. Not every incentive program is a loyalty program, and several common ones are not aimed at your customer at all.
The reason this matters practically is that "incentive program" covers at least five different things in distribution, with different funders, different recipients and different governance. Confusing them is how a distributor ends up describing a manufacturer-funded contractor rebate as "our loyalty program" in a board pack.
| Program type | Funded by | Aimed at | Purpose |
|---|---|---|---|
| Customer loyalty program | The distributor | The customer account | Ongoing relationship, frequency, range, share |
| Customer rebate program | The distributor | The customer account | Committing forecastable volume |
| Manufacturer / channel incentive | The manufacturer | The distributor, or the distributor's customer | Moving a specific manufacturer's product |
| Sales incentive / SPIFF | Either party | Individual salespeople | Short-term selling behaviour on named lines |
| Buyer-directed incentive | Either party | An individual at the customer | Preference at the point of specification or order |
Incentives directed at an individual at a customer business — rather than at the business itself — sit in a different governance category from everything else on this list. Many customers have procurement policies restricting what their staff may accept; some sectors and jurisdictions regulate it; public-sector and regulated buyers frequently prohibit it outright. This guide takes no position on where those lines fall, because they vary enormously. But "who receives the reward, and are they permitted to receive it?" belongs on the design checklist, and it is a question worth putting to your own legal and compliance advisers before launch rather than after an account's procurement team asks it.
The most-cited number in the incentive industry comes from Condly, Clark and Stolovitch's meta-analysis in Performance Improvement Quarterly (2003): 45 qualifying studies, an overall 22% average gain in performance, with long-term programs (6+ months) at 44% versus short-term at 20%, and team-based at 48% versus individual at 19%.
It is routinely misquoted. That meta-analysis measures incentives applied to workers performing work tasks — not customer loyalty programs, and not channel partner programs. It is a legitimate and useful finding about the fourth row of Table 7. It says nothing whatever about the first two, and any vendor presenting "22% performance lift" as evidence for a customer loyalty program has either not read it or is hoping you have not.
9. What real distributor programs actually do
Three structurally different models coexist in live trade programs: distributor-funded points on distributor spend, distributor-administered but manufacturer-funded rebates, and an explicit refusal to run a currency at all in favour of pricing, credit and service. Every program described below was read on the company's own published page. No results, ROI, member counts or lift figures appear here, because with one exception none of these companies publishes any.
We searched specifically for published distributor loyalty benchmarks — redemption rates, enrolment rates, incremental lift, program ROI. There are effectively none in the public record. NAED's Performance Analysis Report and HARDI's Annual Benchmarking Survey both exist and both contain relevant metrics; both are participation- or purchase-gated with no public figures. Every "distributor loyalty increases retention by X%" statistic we encountered traced to vendor marketing with no sample, no method and no publication.
The consequence for this article is simple: the mechanics below are verified, and the outcomes are not reported, because nobody publishes them. Treat any article that gives you a distributor loyalty benchmark without naming a study as a work of fiction.
| Program | Business type | Published mechanic |
|---|---|---|
| Ferguson PRO Plus | Distributor (plumbing / HVAC) | "One point for every dollar ($1.00) spent on products ordered through ferguson.com." 1,500 welcome points; bonus multipliers on featured products. Points not redeemable for cash, gift cards or against Ferguson invoices. All unredeemed points forfeited after 365 days without earning. |
| Wesco CONNECT / EDGE Advantage / VIP Elite | Distributor (electrical / industrial) | Three separate approval-gated programs, not tiers of one. "Selected Wesco customers who are in good standing"; participants must be senior-level executives, subject to Wesco approval. Four purchase goals unlock points at each milestone; VIP Elite sets a goal from prior-year sales and rewards a trip. Supplier-funded multipliers at Diamond 15%, Gold 10%, Silver 5%. More than 35,000 redemption choices for CONNECT and EDGE Advantage members. |
| SiteOne Partners | Distributor (landscape supply) | "1 point for every $1 you spend on qualifying purchases at any SiteOne location" and "2 points for every dollar spent on SiteOne.com and the mobile app." Redemption unlocks at $5,000 annual purchases. No tiers — "No complex system. Just spend and unlock." |
| White Cap Rewards | Distributor (construction supply) | Rewards on every dollar spent, up to $2,000 in vouchers a year. Cash customers only — "customers with a credit account are not eligible for the program." |
| NAPA Installer Rewards | Distributor network (automotive) | A quarterly threshold, not a currency: installers hitting a "$1,950 per quarter purchase benchmark on eligible program products" earn "1% back" on program purchases for that quarter. Paid as prepaid Visa or retail gift card. 2026 awards expire 28 February 2027. |
| ABC Supply Manufacturer Rewards | Distributor (roofing / exteriors) | Administers programs it does not fund. "Our team submits rewards receipts to the manufacturers on your behalf." Nine named manufacturer programs including Atlas Pro Plus, CertainTeed Contractor Cash Back, GAF Rewards, IKO ROOFPRO Advantage and TAMKO The Edge. |
| Airgas manufacturer rebates | Distributor (gas / welding) | Hosts manufacturer-funded rebates rather than a distributor currency: ESAB Burn and Earn (up to $850), Miller Build with Blue (up to $1,000), Lincoln Electric Buy Red Save Green (up to $1,200, claims within 30 days of purchase). |
| Sherwin-Williams PRO+ | Manufacturer with company stores | Explicitly no currency: "No Points. No Minimums. All Benefits." Value delivered as PRO+ pricing, 0% interest credit subject to approval, a dedicated expert team, and in-person and on-demand training. |
| Home Depot Pro Xtra | Retailer with a trade program | Three tiers — Member, Elite, VIP. Elite adds a priority support line; VIP adds account management and preferred pricing. Spend thresholds are not published. Partner offers refresh quarterly and are available regardless of tier. |
| MyLowe's Pro Rewards | Retailer with a trade program | Three tiers — Gold, Platinum, Titanium — starting at Gold with no volume threshold to earn. Published thresholds: 20% member paint discount after $3,000 annual qualifying paint spend; member volume discount on eligible quotes of $1,500 or more. |
| Grainger KeepStock | Distributor (MRO) — not a loyalty program | Vending, point-of-use storage and managed inventory with a web interface offering role-based access, spend limits and consumption reporting. Included here because it is a retention mechanism, and a more powerful one than most programs on this list. |
The three funding models, and why they change the competitive picture
Four design details worth stealing
1. The channel-shift multiplier. Ferguson awards points only on orders placed through ferguson.com. SiteOne pays 1 point per dollar in a branch and 2 points per dollar online or in the app. Both are channel-shift instruments wearing a loyalty badge — and both are the economically strongest design on this list, because the reward is funded out of cost-to-serve rather than gross margin. If you take one idea from this section, take this one.
2. The redemption gate. SiteOne's points only become redeemable once an account reaches $5,000 in annual purchases, with earlier purchases counting toward the threshold. That is a genuine B2B qualification mechanic: it filters out marginal accounts without excluding them, and it defers liability on the long tail.
3. The credit-account exclusion. White Cap's rewards program explicitly excludes customers with a credit account. Read that as a design statement: accounts on credit terms already receive their economics through negotiated pricing; the rewards program exists to reach the cash segment. This is the single clearest published example of a distributor recognising that one program cannot serve two commercially different customer types.
4. The firewall between B2C and B2B. NAPA runs two structurally different programs and separates them explicitly. The consumer NAPA Rewards program awards one point per dollar and states that "NAPA commercial, wholesale, or other accounts receiving a standing discount are not eligible to participate." Trade accounts are routed instead to Installer Rewards: a $1,950 quarterly purchase benchmark that, once cleared, pays 1% back on that quarter's program purchases. That is a company saying, in its own published terms, that B2B and B2C loyalty are not the same instrument.
And read that NAPA structure carefully rather than admiringly. A hurdle that pays on the whole quarter once cleared is not the target rebate Section 6 recommends — it pays on the infra-marginal base as well as on the growth, which is Design A from Table 4 with a qualifying gate in front of it. The gate does real work: it excludes accounts below the benchmark entirely, capping exposure in a way an ungated linear rebate does not. But the structure is a hurdle, not a target, and the same design at 3% on a 20%-margin line would be the trap rather than the fix.
Sherwin-Williams PRO+ leads with "No Points. No Minimums. All Benefits." — pricing, 0% interest credit, expert support and training, with "savings and support start on day one." A serious trade supplier looked at the points model and declined it. Set that against Grainger, which appears to run no customer points program at all and instead embeds itself operationally through KeepStock vending and managed inventory. Two of the largest names in trade supply have concluded that the answer to distributor loyalty is not a currency. Any honest guide has to sit with that rather than write around it.
The Home Depot has published that professionals "make up about 10 percent of The Home Depot's customer base and approximately half of our sales." That is a retailer rather than a distributor, and it is a concentration statistic rather than a program result — but it is the only verified programme-adjacent performance number we could find anywhere in trade loyalty. It points the same way as the Texas A&M stratification finding in Section 3 without matching it: roughly a tenth of customers driving a disproportionate share of sales — about half at The Home Depot, 80% in the branch Texas A&M analysed. The direction is consistent; the magnitude is not, which is precisely why you should stratify your own book rather than borrow anybody else's ratio.
10. Setting up a distributor loyalty program in Apple & Google Wallet
Yes — a trade account membership card works on both platforms, with no app install for you or your customer. But there is a platform detail specific to B2B that almost nothing published gets right: on Google Wallet, a points-free B2B membership card belongs in the generic vertical, and generic passes have no field-update notification mechanism at all. That single fact should shape your pass design before you choose a colour.
A wallet pass is a credential and a communication surface. It is not a loyalty system. A loyalty system still requires rules, customer and account data, qualification logic, reward calculation, redemption handling, expiry management, dispute resolution and an operational process behind all of it. Neither Apple nor Google supplies any of that, and neither does any pass-issuing platform on its own. If a vendor's diagram puts "wallet" in the same box as "loyalty program," that is the moment to ask what happens on the day an account disputes a threshold.
Apple Wallet
Apple documents exactly five pass styles: boarding pass, coupon, event ticket, store card and generic. There is no "membership" style. Apple names the store card style for "discount, gift, and store loyalty cards" — it renders points and rewards balances automatically — and the generic style for passes that do not fit the other categories, with Apple's own documentation giving "a gym membership card, library card, valet claim ticket" as examples and using pass.com.example.membership-card as its sample identifier. For a trade account card with no points balance, generic is the style Apple's own wording points to; store card is the one to use if you are displaying a balance.
On the app question Apple is unambiguous: "Users can add a pass without installing the related app, or add a pass that doesn't have a related app at all." Three distribution routes are documented — from an app or App Clip, as a download on a web page, or as an email attachment. Passes can also be bundled: up to 10 passes or 150 MB in a .pkpasses file, which is genuinely useful if you issue one card per authorised contact on a multi-site account.
Apple's documented update flow is: the user installs an updatable pass; the device registers with your server and provides a push token; the pass changes and your server sends a push; the device queries your server and requests each changed pass.
The step that gets misdescribed is the push itself. Apple's documentation specifies "an empty JSON dictionary for the payload." The push carries no message content whatsoever — it is a silent wake-up. Wallet then fetches the new pass, compares fields, and surfaces a notification only for fields carrying a changeMessage. Apple states it plainly: "You need to provide a value for the system to show a change notification."
The consequence for a distributor program is specific and useful: you cannot push arbitrary marketing copy through Apple's pass channel. The message is a side effect of a data change. "You're $18,000 from your Q3 target" works because the threshold field genuinely changed. "Check out our new range" does not, because nothing on the card moved.
Also worth knowing before you build: Apple notes that a pass update push "works only in the production environment" — there is no sandbox for it.
Two further Apple details matter for a trade card. relevantDate is deprecated in favour of relevantDates (iOS 18 and later), so most published guidance on wallet relevance is written against a superseded key. And a pass can carry up to ten relevant locations — Apple: "If your pass needs more locations, such as a coupon for a chain of stores, start with the best ones" — with maxDistance able only to shrink the system's default radius, never widen it. Apple does not publish that default, so no honest article can tell you the geofence size in metres. For a distributor with forty branches, ten is a real constraint and it is worth deciding early which ten.
sharingProhibited removes the Share button from the back of a pass. Apple is explicit that it "has no effect in earlier versions of iOS, nor does it prevent sharing the pass in some other way." It is a UI affordance, not an access control. If your program has an economic reason to care whether a card is passed around a customer's yard — and if it carries pricing tier or account status, it might — the control has to live in your systems, not in the pass file.
Google Wallet
Google organises passes into verticals — loyalty, offers, gift cards, tickets, transit, access, generic — plus a separate generic private pass for sensitive categories such as health insurance and government ID. A trade account number is not sensitive data in that sense, so a distributor card does not need the private-pass programme, and you would not want the friction: the private-pass vertical is access-gated, requires eligibility documentation and explicit permission from Google, and distributes the pass as a self-contained JWT rather than through the ordinary class-and-object model.
Passes are saved through an "Add to Google Wallet" link — a signed JWT in a URL — which Google documents in these words: "This link can be embedded into websites, email, chat, SMS, or anywhere else that supports hyperlinks." No issuer app is required.
This is the most consequential platform detail in this article and we have not seen it stated anywhere for a B2B context.
Google's loyalty vertical supports notifyPreference: NOTIFY_ON_UPDATE, but Google states that notifications "will only be sent to users if the field is part of an allowlist" — and the allowlist is tiny. On the class: rewardsTier, secondaryRewardsTier, programName. On the object: loyaltyPoints.balance and secondaryLoyaltyPoints.balance. That is the entire list.
Google's generic vertical has no notifyPreference field at all. Generic passes use the Add Message API with TEXT_AND_NOTIFY instead; the only built-in generic notifications are upcoming (24 hours before an interval starts) and expiry (48 hours before it ends).
So a points-free B2B membership card faces a genuine fork. Put it in the generic vertical, which fits the product, and field updates will never notify — your only push channel is Add Message. Put it in the loyalty vertical to get field-update pushes, and the only fields that will fire are tier and points balance — and in practice the notification typically appears only after the customer next opens Wallet. If threshold progress matters to your program, model it as a points balance or a tier, or accept that every notification goes out as a message.
One honest inconsistency: Google's generic-pass FAQ states that "developer authored push notifications are not currently supported by Google Wallet," which contradicts the Add Message documentation on the same doc set. We cannot resolve that from the documentation. Test it on a real device before you design around either answer.
"You may send a maximum of 3 updates that trigger a push notification in a 24 hour period" and "a maximum of 3 messages that trigger a push notification in a 24 hour period." The messages cap appears on Google's trigger-push pages for both loyalty and generic passes; the updates cap appears only on the loyalty page, because generic passes have no update-notification path at all. Both pages carry a note that Google "may throttle your push notification delivery quota if it deems you are spamming your users." Any vendor promising unlimited wallet push on Android is describing something Google's own documentation does not permit.
Location works through merchantLocations — up to ten on the class and ten on the object, so twenty in effect — with Google documenting a notification when a user enters and dwells within a radius Google sets and does not publish — but delivery is controlled entirely by Google and is not dependable; treat location surfacing as an Apple Wallet capability. Note that the older locations field explicitly does not trigger geo notifications; use merchantLocations.
| Question | Apple Wallet | Google Wallet |
|---|---|---|
| Does the customer need an app? | No — a pass can have no related app at all | No issuer app; the user needs Google Wallet |
| Right pass type, no points | Generic | Generic |
| Right pass type, with points | Store card | Loyalty |
| Update mechanism | Web service + APNs; the device re-fetches the pass | REST update (replaces) or patch (merges) on class/object |
| What makes the customer see it | changeMessage on the changed field — otherwise silent | Allowlisted field + NOTIFY_ON_UPDATE (loyalty only), or Add Message |
| Notification on a points-free card | Works — any field with a changeMessage | Field updates cannot notify; Add Message only |
| Published notification cap | None published; you run your own APNs | 3 notification-triggering updates or messages per 24 hours |
| Branch locations on the card | Up to 10; maxDistance can only shrink the default | 10 per class + 10 per object; Google-set radius |
| Barcodes safe on both | Seven formats are documented on both: QR, PDF417, Aztec, Code 128, Code 39, Codabar and EAN-13. Apple alone adds Interleaved 2 of 5; Google alone adds Data Matrix, EAN-8, UPC-A and ITF-14. | |
| Documented B2B account concept | None — a trade card is an ordinary membership pass | None — issuer sub-accounts are about your org, not the customer's |
We checked both sets of documentation specifically for this. Apple's pass personalisation collects only name, postcode, email and phone — there is no company, VAT or account-number field. Google's multi-business language concerns your issuer account structure, not a pass representing a customer business. There is no seat model, no sub-user model, and no organisation-held credential on either platform.
A B2B trade card is, per the documentation, an ordinary membership pass. Everything B2B about it — the account it belongs to, the contacts authorised to hold it, the tier it reflects — lives in your own data model. That is not a limitation to work around; it is simply where the line falls, and knowing it stops you searching for a platform feature that does not exist.
11. When a wallet pass makes sense for a distributor
A wallet pass earns its place when your program has state that changes, when that state changes slowly enough to matter and often enough to notice, and when the person holding the phone is the person whose behaviour you want to change. If any of those three is false, the card is decoration. Four gates decide it.
Four gates, run in order, with the third being the one that eliminates most candidate programs. Designed to be answered in a meeting in under ten minutes and to produce a defensible "no" as readily as a "yes."
Where a distributor card genuinely earns its place
- Tiered programs. A status level is the single most legible thing a card can carry, and it changes at exactly the right cadence — a few times a year, visibly, with consequences.
- Threshold programs. Progress toward a quarterly or annual target is useless in a portal nobody opens and valuable on an iPhone lock screen in week seven.
- Multi-branch accounts. An account number and a lookup barcode that work at any counter, plus branch contact details, removes real friction at the trade desk.
- Cash and small trade accounts. The segment least likely to have a portal login and most likely to be standing at a counter with a phone in hand.
- Programs where the rep changes. The card is a relationship with the firm. It survives a resignation in a way a rep's phone contacts do not.
Where it does not
- Pure annual rebate agreements with no interim state. If nothing changes for eleven months, a contract and an email are the right instruments.
- Programs whose only content is a discount already applied at invoice. The customer receives the value without needing to be told; a card adds an object to manage, not information.
- Accounts where all ordering runs through an EDI or punchout integration. Nobody is standing at a counter and no human sees the prompt.
- Anything requiring a confirmed financial figure. See the warning in Section 7.
12. Push notifications that bring distributor customers back
A wallet push is a short, opted-in message to someone who chose to keep your card on their phone. In distribution it suits threshold progress, tier changes, expiring value, stock and branch notices, narrow seasonal prompts and reactivation — six things that are time-sensitive and actionable. It does not suit general promotion, and the platform limits will stop you trying.
Membership → wallet → state change → notification → action → new state. The loop only turns when the state change is real. This is the practical reason the channel resists being used for advertising: on Apple the message is the data change, and on Google the cap is three a day.
The six message types that work in distribution
- Threshold progress. "You're $18,000 from your Q3 target — period ends 30 September." The single highest-value message this channel can carry, because it is information the customer wants and cannot easily get elsewhere.
- Tier change. Both directions. Moving up is a celebration and a reason to call. Moving down, delivered before it happens, is the most effective retention message in a tiered program and the one most companies are too nervous to send.
- Expiring value. If your points expire on 365-day inactivity, or awards must be redeemed by a hard date, that deadline is a legitimate, welcome, time-sensitive message. It is also the one your terms probably require you to communicate.
- Stock, branch and logistics. A back-ordered line arriving, a branch changing hours, a new trade counter opening, a delivery cut-off moving before a holiday. Highest tolerance of any message type in B2B, because it is operationally useful.
- Seasonal and category prompts. Narrow and calendar-driven — the start of a season, a code change, an annual maintenance window. Useful when tied to the account's actual buying pattern; noise when broadcast.
- Reactivation. One message to an account whose order cadence has broken, and one only. Not a sequence.
1. Every message must be actionable inside the period. If the customer cannot do anything about it before the quarter closes, it is not a notification, it is a newsletter.
2. Respect the published limits. Google caps notification-triggering updates and messages at three in a 24-hour period. Apple requires a changeMessage on the changed field or the update is entirely silent.
3. Two to four a month is a defensible ceiling. A purchasing manager is not browsing. Write it into the program rules and give someone the authority to enforce it.
4. Never put a confirmed financial figure on the card. Progress toward a threshold, yes. A calculated accrual or projected payout, no. See Section 7.
5. Assume it is read at a counter. Pricing tier, discount level and account status are commercially sensitive in a trade environment where competitors' customers share job sites. Decide deliberately what is visible on the front of the card versus the back.
13. The distributor loyalty KPI framework
Measure in a chain: objective → target behaviour → leading KPI → lagging KPI → financial outcome — and never report a number from the middle of that chain as though it were the end of it. Enrolment, redemption and engagement are program activity. Incremental gross profit is program value. Almost every distributor loyalty dashboard reports the first three and none of the fourth.
Six objectives, each traced from the behaviour it targets to the money it produces. Built so that a program manager and a finance director can read the same row and agree on what it means. The rule that makes it work: every row must terminate in the financial-outcome column, or the objective does not belong in the program.
| Objective | Target behaviour | Leading KPI | Lagging KPI | Financial outcome |
|---|---|---|---|---|
| Keep accounts ordering | Orders continue at historic cadence | Days since last order, by account | Repeat-order rate; active-account count | Retained gross profit |
| Increase order frequency | More orders per period | Orders per account per month | Order-frequency trend vs baseline | Incremental gross profit, net of pick-and-pack |
| Grow order size | More lines or units per order | Lines per order; average order value | AOV trend vs matched control | Gross profit per order, less any split-order effect |
| Widen the range bought | First purchase in a new category | New categories entered per account | Category breadth; share of wallet | Incremental gross profit at category margin |
| Shift channel | Ordering online instead of by phone or counter | Share of orders placed digitally | Digital share of account revenue | Cost-to-serve saved per order moved |
| Reactivate lapsed accounts | An order after a defined dormant period | Reactivation message response rate | Reactivated accounts, and their survival at 90 days | Recovered gross profit, less reactivation cost |
The five measurement layers, and what each one lies about
| Layer | Example metrics | What it tells you | How it misleads |
|---|---|---|---|
| Adoption | Enrolled accounts, activation rate, card saves by branch | Whether customers accepted the program | Enrolment without activation is a vanity number |
| Engagement | Pass installs and removals, opt-out rate | Whether the communication is welcome | Opt-out rate is the honest metric here, and it is the one nobody puts on the slide |
| Behavioural | Repeat-order rate, frequency, AOV, category breadth, digital share | Whether buying patterns actually moved | Enrolled accounts grow faster than non-enrolled because growing accounts enrol — selection, not causation |
| Program cost | Reward cost, redemption rate, fulfilment, administration, headcount | What the program actually costs | Redemption lag makes years one and two look cheap |
| Financial | Incremental gross profit, margin contribution, cost-to-serve saved | Whether the program is worth running | Meaningless without a counterfactual — see Section 6 |
The most common analysis in distributor loyalty is a comparison of enrolled against non-enrolled accounts. It is close to worthless, and it always flatters the program. Accounts that enrol are systematically different from accounts that do not: they are larger, more engaged, more digitally active, more likely to have a rep relationship, and — critically — more likely to have been growing already. The measured gap is mostly selection.
The fix is not complicated. Compare enrolled accounts against a matched group on prior-period spend, category mix and branch, or stagger the launch across branches and compare cohorts. Either takes a week to set up before launch and is impossible to reconstruct afterwards. This is the argument for doing it in the right order.
Why incremental behaviour is the only thing that counts
The nuance that separates a credible program from a well-decorated one: an increase in program activity is not evidence of an increase in revenue, and an increase in revenue is not evidence that the program caused it. A loyalty program can raise engagement, raise redemption, raise enrolment and raise measured account growth while destroying gross profit — that is precisely what the linear rebate in Table 4 does, and every metric except the last one looks excellent while it happens.
The literature is more measured about program effects than program vendors are. Belli and colleagues' meta-analysis in the Journal of the Academy of Marketing Science (2022), synthesising 429 effect sizes from thirty years of research, finds strong evidence that loyalty programs enhance customer loyalty — but with an important asymmetry: programs "particularly enhance behavioral loyalty, [while] shifting consumers' attitudinal loyalty is more challenging." Effectiveness "differs systematically depending on LP design characteristics (LP structure, reward content and delivery) and industry characteristics."
Read against Kim, Steinhoff and Palmatier's observation in the same journal that "real-world efforts often fail," the honest position is this: a well-designed program reliably moves what people do. It much less reliably moves what they prefer. In distribution, where the account is buying from three suppliers and will keep doing so, moving what they do is worth a great deal — provided you measure it properly and price it against the ceiling in Section 6.
- Every objective must trace to a financial outcome. If a row in your KPI table stops at "engagement," delete the objective or find its money.
- Opt-out rate is the most honest metric in the program. Track it weekly; treat a rise as a stop signal, not a rounding error.
- Enrolled-versus-non-enrolled comparison is selection bias with a chart on it. Build a matched control or a staggered launch before you go live.
- Loyalty programs move behaviour more reliably than they move preference. Design for behaviour and claim only behaviour.
14. How to build a distributor loyalty program, step by step
Twelve steps, and the order is the point. Segment before you design, model the economics before you choose a mechanic, and capture a baseline before you launch. Every one of those three, done late, is either impossible to recover or expensive to unwind.
- Write the objective in one sentence with a number and a period in it. "Grow trade sales" is not an objective. "Move 200 mid-tier accounts from four orders a quarter to six, within two quarters, in consumables" is. If nobody can write that sentence, the program does not have an objective — it has a budget looking for one.
- Segment and stratify the account base. By spend, by gross profit, by cost-to-serve, and by category breadth. The Texas A&M stratification work in Section 3 is the model: buying power, loyalty, profitability and cost-to-serve, evaluated together rather than one at a time. Programs designed on revenue alone reliably subsidise the accounts that are already costing you money.
- Decide whether you are rewarding the account or the buyer. Decision zero from Figure 1. Write down the answer. It determines eligibility, fulfilment, communication and the compliance questions in Section 8.
- Name the target behaviour and check it is worth buying. Frequency, breadth, channel or share — one of them, primarily. Then estimate what a unit of it is worth in gross profit before you decide what to pay for it.
- Model the economics before you choose the mechanic. Run the Loyalty Economics Test and the Reward Ceiling from Section 6 at three growth scenarios, including one where the account grows less than you hope. If the program only works in the optimistic scenario, it does not work.
- Choose one model and at most one accelerator. Framework 1 in Section 4. Write down why the accelerator exists.
- Define eligibility and exclusions precisely. Which account types, which product lines, which channels, which order types. Exclusions matter as much as inclusions — look at White Cap excluding credit accounts and NAPA excluding standing-discount accounts from the consumer program. Both are deliberate, and both prevent a program from paying twice for the same customer.
- Write the terms, including expiry, before launch. Earning rate, qualifying period, thresholds, redemption rules, expiry, forfeiture, what happens on account closure or acquisition, and the right to amend. Any points currency without an expiry rule is an open-ended liability.
- Design the communication architecture. Message categories, frequency ceiling, who approves wording, what appears on the card versus what stays in the portal, and the opt-out path. Decide here — not later — that no confirmed financial figure goes on a pass.
- Capture the baseline. Twelve months of per-account history, and a matched control group or a staggered branch rollout. This is the step that cannot be added retroactively, and skipping it means the program can never be evaluated, only defended.
- Launch narrow. One region or one branch cluster, one account segment, a defined period. Train the trade counter before the customers arrive — in distribution the counter staff are the enrolment channel, and a program the counter cannot explain in one sentence will not be enrolled.
- Review on a set schedule. Monthly on engagement and opt-outs, quarterly on behaviour and cost, annually on terms, economics and whether the program still targets a behaviour you still want.
Fourteen items to clear before a single account sees the program. Print it. Item five is the one that gets skipped and item ten is the one that cannot be recovered afterwards.
- Objective written with a number, a segment and a time period, and agreed by commercial leadership
- Accounts stratified on spend, gross profit, cost-to-serve and category breadth — not revenue alone
- Account-versus-buyer decision made explicitly and recorded
- Target behaviour named, and its gross-profit value per unit estimated
- Economics modelled at three growth scenarios, including a pessimistic one, against the Reward Ceiling
- One model chosen, with at most one accelerator and a written reason for it
- Eligibility and exclusions defined — account types, product lines, channels, order types
- Programme terms written including expiry, forfeiture, amendment rights and account-closure handling
- Finance sign-off obtained on liability, accrual treatment and how program cost will be reported
- Baseline captured — twelve months per account, plus a matched control or staggered rollout
- Communication rules written — categories, frequency ceiling, approval owner, opt-out path
- Pass field list reviewed — no confirmed accruals, and a decision on what is commercially sensitive
- Both platforms tested on real devices — add and update, iOS and Android, including a notification
- Trade counter trained and able to explain the program in one sentence, before launch day
15. Where PushNotice fits
PushNotice is a wallet-based engagement layer: it creates Apple Wallet and Google Wallet passes and sends pass-level push notifications, without the business or the customer needing a mobile app. In a distributor program that makes it the membership and communication layers of Figure 1 — not the rules engine, not the rebate calculation, and not the system of record.
PushNotice is not an ERP, a CRM, a distribution management system, an order-management system, a rebate-calculation engine, a pricing system or an accounting platform. As of this review date it publishes no public API and no named POS, CRM, ERP or DMS integration on its product pages or pricing. It does not calculate rebates, hold your pricing matrix, or connect to your order data on its own. If you need automatic rebate accrual or a live feed from your ERP, that is a different category of product and you should buy one.
We would rather say that here than have you find it in week three of an implementation. Two of our own older comparison posts are less clear on this than they should be, and we are correcting them.
What it does, verifiably. A trade contact saves a membership or loyalty pass to Apple Wallet on iPhone or Google Wallet on Android from a single link or QR code — at the counter, on an order confirmation, or in an email. Pass content can be updated and a notification sent with no companion app. On Apple Wallet, passes can be associated with places, so the card can surface on the iPhone Lock Screen when someone is near a branch — a paid-tier feature named Smart Location; Google Wallet holders are not affected. Higher tiers add tagging and segmentation, and multiple workspaces for multi-brand or multi-region operations. Pricing runs from a free plan at $0 through Starter at $29 per month and Pro at $79 per month, with an agency tier priced by client count for white-label use.
Where that fits a distributor program. Three places, honestly. It carries the account credential — account number, tier, branch contact, lookup barcode. It carries state the customer wants to see — tier, threshold progress, expiry date — which you update through the platform, not through an integration with your ERP, because as said above there is no published API to build one on. And it carries the notification when that state changes, inside the platform limits set out in Section 10. Everything upstream of that — who qualifies, what they earn, what it costs you, whether it worked — is your program, and this article is mostly about getting that part right first.
On templates. The platform offers pass templates for loyalty cards, coupons, memberships and rewards, which is where a distributor card starts. A preset distributor configuration is a natural thing to build on top of that and we would like to; we are not going to describe one here as though it already exists.
Which customer and account data the platform will hold, and where. Whether pass state can be updated programmatically from your systems, and by what method. What the segmentation model actually supports at your account count. Whether the notification behaviour you are relying on works in the pass vertical you need — test it on real devices, because Section 10 shows the answer differs sharply between Google's generic and loyalty verticals. What the data export and deletion path is. And whether any integration claim is backed by named systems and a reference customer. If we cannot satisfy those for your operation, we would rather you used something that can.
16. Future research opportunities
The public record on distributor loyalty is close to empty, and that is an opportunity rather than a complaint. Five research assets do not exist and would be cited immediately if they did. We are publishing the list because we would rather the work existed than that we were the only ones who noticed it was missing.
In researching this article we looked specifically for published benchmarks, effectiveness data and program inventories in wholesale distribution. What exists is either paywalled association research with no public figures, or vendor marketing with no method. Five gaps stand out.
- A State of Distributor Loyalty report. A survey of distributors on program prevalence, model type, funding source (distributor versus manufacturer), qualification period and administrative cost. Nothing comparable exists publicly. The nearest analogue — Distribution Strategy Group's annual e-commerce survey, 401 respondents with published field dates — shows the format works and gets cited.
- A distributor loyalty benchmark study with a published method. Enrolment, activation, redemption and opt-out rates by distribution vertical. Every number currently in circulation is unsourced. The first credible one becomes the citation everybody uses.
- A B2B loyalty economics calculator. An interactive tool implementing the Reward Ceiling and the Linear Rebate Break-Even from Section 6, letting a distributor enter baseline spend, margin and rebate rate and see the break-even growth rate. The formula is in this article; the tool is not, and the tool is the linkable asset.
- A distributor loyalty program database. A maintained, verified inventory of live trade programs with mechanics, funding source and business type — an expanded, updated version of Table 8. Reference material of this kind attracts links for years because it saves everyone else the research.
- Wallet adoption research in distribution. Nobody has measured how trade customers actually use digital credentials at a counter — save rates by enrolment channel and retention of the pass over time. We have a partial view from our own platform and it is not a substitute for a study.
Publishing a research agenda you have not completed is unusual, and it is deliberate. Two of these — the calculator and the program database — are things we intend to build. The other three would be better done by an association or an academic group than by a vendor, and saying so is more useful than pretending we are about to do all five. If you are working on any of them, we would rather link to your version than duplicate it.
Frequently asked questions
Sixteen questions distributors, B2B marketers, channel managers and loyalty consultants actually ask about trade programs — grouped by what you are trying to decide.
The core questions
What is a distributor loyalty program?
A structured, ongoing system through which a wholesale distributor recognises, rewards and communicates with its business customers in order to influence account-level purchasing behaviour — order frequency, category breadth, channel choice and share of wallet — and to make the commercial relationship visible to the customer between orders. In practice it is built from some combination of account qualification, an incentive structure, tiering, reward fulfilment, an account membership credential, a communication layer and a measurement discipline. Most distributors should run three or four of those seven components, not all of them.
How does a distributor loyalty program work?
An account qualifies on defined criteria — usually trade status, good standing and often a minimum annual spend. Qualifying activity in a defined period earns something: points, a rebate percentage, a tier, or access to pricing and service benefits. The account's status is communicated back to the customer, ideally continuously rather than at settlement, so they can act on it while the period is still open. At period end the earned value is calculated, verified and fulfilled. Behind all of that sits the part vendors rarely show you: rules, account data, qualification logic, redemption handling, expiry management and a dispute process.
How is a distributor loyalty program different from a rebate program?
A loyalty program is the broad, ongoing engagement and recognition system. A rebate program is a narrow financial instrument: a contractual entitlement to a portion of spend, earned against defined criteria. The difference is not cosmetic. Under IFRS 15 and ASC 606, a rebate is consideration payable to a customer, which reduces the transaction price and therefore reported revenue, and must be estimated and accrued in advance under a constraint IFRS 15 sets at "highly probable" and ASC 606 at the lower "probable" threshold. A loyalty point is a discretionary currency you issue under terms you set. They sit with different owners inside the business — finance for the rebate, commercial or marketing for the program — and they work best together: the rebate carries the economics, the loyalty layer makes it visible in time for the customer to act on it.
What is a distributor incentive program?
"Incentive program" is the broadest of the three terms: any mechanism designed to influence a specific commercial behaviour in the channel. It covers at least five distinct things — a customer loyalty program, a customer rebate program, a manufacturer-funded channel incentive, a sales incentive or SPIFF aimed at your own salespeople, and an incentive directed at an individual at the customer business. A loyalty program and a rebate program are both kinds of incentive program; the reverse is not true. The distinction matters because the funder, the recipient and the governance differ in every row, and the last of the five carries considerations the others do not.
Choosing the mechanic
What rewards work best for distributors?
The evidence from live trade programs points in a consistent direction, and it is not toward merchandise catalogues. The strongest designs in the market reward behaviour that is cheap for the distributor to serve: Ferguson awards points only on orders placed through its website, and SiteOne pays double points online and in-app versus in-branch. Both are channel-shift instruments, funded out of cost-to-serve rather than gross margin. Beyond that, account credit, prepaid cards and service or pricing access all avoid the fulfilment cost and liability that merchandise carries. And it is worth noting that two of the most credible trade suppliers — Sherwin-Williams with PRO+ and Grainger with KeepStock — deliver value through pricing, credit, service and embedded inventory rather than any reward currency at all.
Should distributors use points or rebates?
It depends on account size and on what the customer wants. Large accounts generally want a contractual entitlement they can forecast — a rebate. Smaller and cash accounts respond better to a visible, simple currency. The more important question is not points versus rebates but linear versus target: a rebate paid on all qualifying spend funds volume you would have received anyway, while one paid only above a baseline does not. On a 25% gross margin, a 2% linear rebate needs 8.7% account growth simply to break even; a 2% target rebate is profitable at any positive growth. Peer-reviewed supply-chain research reaches the same conclusion formally — Taylor (2002) shows target rebates can coordinate a channel where linear rebates cannot.
How do distributor loyalty tiers work?
An account reaches a status level by meeting a threshold — usually annual spend, sometimes a combination of spend and category breadth — and holds that level for a defined period, with benefits attached. Live examples run from three published tiers (Home Depot Pro Xtra's Member, Elite and VIP; MyLowe's Gold, Platinum and Titanium) to Wesco's three separate approval-gated programs — CONNECT Rewards, EDGE Advantage and VIP Elite — where participation is by Wesco approval and participants must be senior-level executives. Tiers suit distribution because account portfolios are steeply skewed, and a tier is the single most legible field you can put on a membership card. The two disciplines that matter: each tier must carry a benefit the customer can name, and you must be willing to move an account down as well as up — or the structure inflates until everyone is at the top and it means nothing.
How much does a distributor loyalty program cost?
Four cost lines, and most business cases include only the first. Reward cost — what you pay out, which for a linear rebate is far larger than the headline rate implies. Fulfilment — cards, merchandise, shipping, prepaid card issuance. Administration — the software, and the people who run enrolment, verification, disputes and reporting, which in a rebate program is substantial. Liability — the value issued but not yet redeemed, which makes a points program look cheap in its early years and expensive once redemption catches up. Software is usually the smallest of the four: wallet pass platforms start at $0 and run to tens or low hundreds of dollars a month. The reward line is where the money is, and it is set by the design in Section 6, not by the vendor you choose.
Retention and measurement
How can distributors improve customer retention?
Start by recognising that distribution customers rarely leave outright — they erode, ordering less while still ordering. Six causes account for most of it: price competition, dependence on a single rep, friction in reordering, rebate complexity the customer cannot forecast, poor visibility into which accounts are actually profitable, and programs the customer forgets exist. A loyalty program can address three of those six. Making switching expensive is a weak lever: a meta-analysis of 170 independent samples found switching costs have "only a weak negative influence on switching," and the effect is weaker in B2B than B2C. Fixing the reorder flow, making thresholds visible in time to act on them, and giving the account a relationship with the firm that survives a rep changing are all more durable than lock-in.
What KPIs should distributors track?
Track in five layers and never report a middle layer as though it were the last. Adoption: enrolled accounts, activation rate, card saves by branch. Engagement: message response and — most importantly — opt-out rate, which is the most honest number in any loyalty program. Behavioural: repeat-order rate, order frequency, average order value, category breadth, digital order share, and share of wallet. Cost: reward cost, redemption rate, fulfilment, administration. Financial: incremental gross profit, margin contribution and cost-to-serve saved. In distribution, share of wallet is usually more meaningful than retention, because most trade accounts buy from several distributors at once and the failure mode is share erosion rather than departure.
How do you calculate distributor loyalty ROI?
Incremental gross profit minus total program cost, divided by total program cost — with both words doing real work. Incremental means only the behaviour the program caused, measured against a counterfactual: a matched control group of accounts you did not enrol, a staggered rollout across branches, or at minimum a twelve-month pre-launch baseline compared against trend rather than against a flat line. Total means everything the program paid out, including reward on volume you would have received anyway — which for a linear rebate is most of it. The most common error is comparing enrolled accounts against non-enrolled accounts. That comparison always flatters the program, because accounts that enrol were already larger, more engaged and more likely to be growing. It measures selection, not effect.
Wallet, cards and push
Can distributor loyalty programs use Apple Wallet?
Yes. Apple documents five pass styles and names the store card style for loyalty and rewards cards that display a balance, and the generic style for membership cards — Apple's own example identifier is a membership card. For a trade account card with no points balance, generic is the style Apple's wording points to. Apple states that users can add a pass without installing the related app, or add a pass that has no related app at all, and documents three distribution routes: from an app, as a web page download, or as an email attachment. Passes update through a web service and the Apple Push Notification service. One detail worth knowing before you build: the update push payload is, in Apple's words, an empty JSON dictionary — for the customer to see anything, you must set a change message on the field that changed.
Can distributors use Google Wallet?
Yes, with one important design consequence. A trade account card with no points belongs in Google's generic vertical, and generic passes have no notifyPreference field at all — field updates cannot trigger a notification, and your only push channel is the Add Message API with TEXT_AND_NOTIFY, and even then the notification typically appears only after the customer next opens Wallet. Google's loyalty vertical does support field-update notifications, but only for an allowlist of fields: rewards tier, secondary rewards tier and program name on the class, and points balance on the object. So if threshold progress needs to notify, model it as a points balance or a tier, or accept that every message goes out as a message. Either way, Google caps notification-triggering updates or messages at three in any 24-hour period. A trade account card does not need the generic private pass programme, which is for sensitive categories such as health insurance and government ID.
What should a distributor loyalty pass include?
The distributor name and logo, the account name and account number, the authorised contact's name, tier or program status, progress toward the current threshold with the period end date, the home branch's phone number and address, and a barcode or QR code that resolves to a record you control for counter lookup. On the back: programme terms, a link to full statements, and the opt-out path. What should not appear: a calculated rebate accrual or projected payout, since a pass is a communication surface rather than a statement of account; and anything commercially sensitive you would not want read over a shoulder at a busy trade counter — discount level and pricing tier deserve a deliberate decision rather than a default.
Can a wallet pass replace a mobile app?
For a membership credential and a notification channel, largely yes — and that is the point. Neither Apple nor Google requires an issuer app for a customer to save and receive a pass, so a distributor gets a persistent presence on the customer's phone without asking a purchasing manager to install anything. What a pass cannot replace is a functional app or portal: ordering, catalogue browsing, quotes, invoices, statements, punchout and account administration all need somewhere real to live. The honest framing is that a wallet pass is the credential and the prompt; your e-commerce site or portal is the destination. It is also worth noting that neither platform has any concept of a business account — a B2B trade card is, per the documentation, an ordinary membership pass, and everything B2B about it lives in your own data model.
How can push notifications support distributor loyalty?
Six message types work in a trade setting, and all six share one property — they are actionable inside the current period. Threshold progress ("you're $18,000 from your Q3 target, period ends 30 September"), which is the highest-value message this channel carries. Tier changes, in both directions, with a downgrade warned before it happens. Expiring value, which your terms probably require you to communicate anyway. Stock, branch and logistics notices, which have the highest tolerance of any B2B message type because they are operationally useful. Narrow seasonal or category prompts tied to the account's actual buying pattern. And a single reactivation message — one, not a sequence — to an account whose cadence has broken. Two to four messages a month is a defensible ceiling for a purchasing audience, and the platform limits will enforce restraint whether you intend it or not.
Methodology, sources & disclosure
This guide is published by PushNotice, reviewed by its editorial team, and written to be useful whether or not you use our product. Every platform statement comes from Apple's or Google's developer documentation. Every research finding is quoted with its sample and its limits. Every program mechanic was read on the company's own published page. No distributor loyalty benchmarks, retention percentages, redemption rates, lift figures or ROI numbers are asserted anywhere in this article, because none exist in the public record that survive checking.
What is sourced, and how
Four categories of claim appear here and they are treated differently. Peer-reviewed research is cited with its sample size, journal and year, and reported with its own caveats rather than its headline. Platform documentation — Apple's Wallet Passes reference, pass update web service, Lock Screen relevance and barcode documentation; Google's Wallet generic and loyalty class and object references, update, trigger-push and merchant-location documentation — was consulted 4 September 2026 and quoted rather than paraphrased where precision matters. Company programme mechanics and financial figures come from the company's own published pages, terms and results releases. Everything else — the seven frameworks, the matrices, the decision tree, the checklist, the Reward Ceiling, the Linear Rebate Break-Even and the Incremental Rate Trap — is PushNotice analysis and labelled as such.
What was deliberately excluded, and why
Five things, and the exclusions are as much a part of this article as the inclusions.
"It costs five times more to acquire a customer than to retain one." No retrievable primary source. Traced by the authors of Loyalty Myths (Ipsos Loyalty) to unpublished late-1980s consulting research at the Technical Assistance Research Project, with their own verdict that "there is currently enough contrary information to bury or significantly qualify this truism." Pfeifer's peer-reviewed analysis in the Journal of Targeting (2005) shows the ratio is analytically inert unless you separate average from marginal cost. No distribution-specific version exists at all.
"A 5% increase in retention increases profits by 25–95%." A conflation of three separate claims. Bain's published brief says "in financial services, for example, a 5% increase in customer retention produces more than a 25% increase in profit" — one industry, and a lower bound rather than a range. Bain's own gloss on the 1990 Harvard Business Review article "Zero Defections" is a different claim again ("almost 100%," with MBNA as the worked example). The familiar "25% to 95%" range traces to a 2014 HBR blog post citing no study. We could not read the 1990 article, which is paywalled, and therefore assert nothing about what its chart contains.
The "22% performance lift from incentives" figure. Real, peer-reviewed and correctly reported in Section 8 — but it measures workplace performance incentives applied to workers, not customer loyalty programs. It is excluded from every part of this article that concerns customer programs.
Gartner's "six to ten decision makers" figure. Widely quoted and not present on any live Gartner page we could find. The current published figure, from a Gartner press release dated 7 May 2025 with a stated sample of 632 B2B buyers surveyed August–September 2024, is "five to 16 people across as many as four functions." That is the figure used here.
"19% of earned rebates never get claimed," and every figure for MDF and co-op fund levels. The first is a rebate-software vendor's self-reported analysis of 47 unnamed companies with no published methodology. For the second we could find no Tier 1 or Tier 2 source at all. Both are omitted.
Where we could not verify something
Stated in the text rather than smoothed over. US Census Bureau gross-margin-by-line-of-business data exists — Annual Wholesale Trade Survey Table 4, and Economic Census table EC2242GRMARGPROF — but was not retrievable in this research environment, so this article uses company-reported gross margins from three publicly listed distributors instead and says so. Google's own generic-pass FAQ contradicts its Add Message documentation on whether developer-authored push notifications are supported; we report the contradiction rather than resolving it. Neither platform publishes its default geofence radius, so this article gives no figure in metres — treat any article that does with suspicion. Home Depot's Pro Xtra tier thresholds are not published, so no dollar figures appear for them. And no distributor publishes program outcome data: enrolment, redemption, lift and ROI figures are absent from this article because they are absent from the public record.
Disclosure and limitations
PushNotice sells wallet marketing software, and Section 15 describes our own product. We have tried to make that section unusually explicit about what PushNotice is not, including the absence of any public API and of any named POS, CRM, ERP or DMS integration. The author is not an accountant, auditor, tax adviser or trade-law specialist, and this article is not accounting, tax or legal advice. The accounting standards are quoted to establish that a rebate and a loyalty point are different economic objects — not to tell you how to book one. The worked examples in Section 6 use an illustrative 25% gross margin and a $200,000 account; substitute your own figures, and take them to your finance team before you take them to a customer.
Primary sources
Peer-reviewed research: Taylor, T.A., "Supply Chain Coordination Under Channel Rebates with Sales Effort Effects," Management Science 48(8), 2002. Belli, A., O'Rourke, A-M., Carrillat, F.A., Pupovac, L., Melnyk, V. & Napolova, E., "40 years of loyalty programs: how effective are they? Generalizations from a meta-analysis," Journal of the Academy of Marketing Science 50(1), 2022. Kim, J.J., Steinhoff, L. & Palmatier, R.W., "An emerging theory of loyalty program dynamics," JAMS 49(1), 2021. Palmatier, R.W., Dant, R.P., Grewal, D. & Evans, K.R., "Factors Influencing the Effectiveness of Relationship Marketing: A Meta-Analysis," Journal of Marketing 70(4), 2006. Pick, D. & Eisend, M., "Buyers' perceived switching costs and switching: a meta-analytic assessment of their antecedents," JAMS 42(2), 2014. Dikcius, V., Kirse, S., Casas, R. & Koncanina, A., "Drivers of Attitudinal and Behavioural Loyalty in B-2-B Markets," Inzinerine Ekonomika–Engineering Economics 30(1), 2019. Keiningham, T.L., Perkins-Munn, T. & Evans, H., "The Impact of Customer Satisfaction on Share-of-Wallet in a Business-to-Business Environment," Journal of Service Research 6(1), 2003. Condly, S.J., Clark, R.E. & Stolovitch, H.D., "The Effects of Incentives on Workplace Performance: A Meta-analytic Review of Research Studies," Performance Improvement Quarterly 16(3), 2003. Pfeifer, P.E., "The optimal ratio of acquisition and retention costs," Journal of Targeting, Measurement and Analysis for Marketing 13(2), 2005.
Industry and university research: Narayanan, A., Lawrence, F.B., Rao, B. & Krishnadevarajan, P., "Customer Stratification: Understanding Customer Profitability," Department of Industrial Distribution, Texas A&M University (POMS proceedings; data collected November 2005 – October 2006). NAW Institute & Texas A&M, "Customer Stratification: Best Practices for Boosting Profitability" (68 wholesaler-distributors). Distribution Strategy Group, "2024 State of eCommerce in Distribution" (Jonathan Bein, 401 respondents, surveyed August–September 2024). McKinsey & Company, "The surprising economics of B2B growth," 28 May 2026 (nearly 4,000 decision-makers, 13 countries). Gartner press release, 7 May 2025 (632 B2B buyers). Ipsos Loyalty, Loyalty Myths, Myth 8 excerpt. Bain & Company, "Prescription for cutting costs."
Accounting standards: IFRS 15 Revenue from Contracts with Customers, paragraphs 50, 51, 53, 56, 70, 71 and 72 (IFRS Foundation issued standard). ASC 606-10-32-11 (the constraint, at the "probable" threshold); ASC 606-10-32-25 and 606-10-32-27 (consideration payable to a customer).
Platform documentation: Apple Developer — Wallet Passes; Creating a store card pass; Creating a generic pass; Distributing and updating a pass; Adding a web service to update passes; PassFieldContent; Pass.Barcodes; Pass.RelevantDates; Showing a Pass on the Lock Screen; Getting Started with Apple Wallet. Google — Wallet overview; Generic passes; Issuing passes for web, email, SMS; Update passes, classes and objects; Trigger push notifications (loyalty and generic); Generic pass notifications; loyaltyclass, loyaltyobject, genericclass, genericobject, MerchantLocation and BarcodeType references; Generic private pass; Generic pass FAQ. All consulted 4 September 2026.
Company programme and financial sources: Ferguson PRO Plus programme page and FAQs. Wesco loyalty programmes page. SiteOne Partners programme page. White Cap Rewards. NAPA Installer Rewards; NAPA Rewards programme details. ABC Supply Manufacturer Rewards Support. Airgas rebate programmes (ESAB, Miller, Lincoln Electric). Sherwin-Williams PRO+. The Home Depot corporate releases on Pro Xtra (5 January 2023; 6 July 2026). Lowe's corporate release on MyLowe's Pro Rewards (18 February 2025). Grainger KeepStock. Fastenal 2025 Annual and Fourth Quarter Earnings (19 January 2026). W.W. Grainger fourth quarter and full year 2025 results (3 February 2026). Watsco second quarter 2026 results (29 July 2026).
About the author
Sajid Ali is Founder and CEO of PushNotice, a wallet marketing platform for small, multi-location and B2B businesses. He writes about wallet passes, loyalty mechanics and customer communication from a technology and business perspective. He is not an accountant, auditor, tax adviser or trade-law specialist, and the accounting and regulatory material in this article is a summary of publicly available primary sources, provided so that distributors and their advisers can have a better-informed conversation rather than as a substitute for one. Connect on LinkedIn.