Free resource · use it on this page
Loyalty Vendor Evaluation Scorecard
The nine benchmark criteria as a blank weighted scorecard — you set the weights, score each vendor 1 to 5, and end with a decision a committee can defend.
How this scorecard works
This is the nine-criteria benchmark from our POS loyalty guide, stripped of our weights so you can apply your own. Assign each criterion a weight, with all nine weights totaling exactly 100. Score every vendor from 1 to 5 on each criterion. A vendor's total is the sum of weight times score across all nine rows: the maximum is 500, the floor is 100. The number matters less than the trail it leaves — a weighted grid turns 'we liked them' into a record of what you valued, what you saw, and why the winner won.
The criteria are stable; the weights are your strategy. A drive-thru coffee bar should not weight counter speed the same as an appointment-based salon, and a business with 4,000 existing members should not weight migration the same as one starting from zero. Argue about the weights before the first demo, not after the scores exist.
Disclosure: this scorecard is published by PushNotice, a wallet-pass marketing platform. The criteria are deliberately vendor-neutral and apply to us exactly as they apply to anyone else — that is the point of handing you a blank grid instead of a verdict.
Use it directly on this page: print it, or copy the grid into a spreadsheet. Make one copy per stakeholder per vendor if a committee is scoring.
The nine criteria
Each criterion comes with anchors for the bottom and top of the scale, so two people scoring the same demo are using the same ruler. Scores of 2 through 4 sit between the anchors.
| Criterion | What you are scoring | What a 1 looks like | What a 5 looks like |
|---|---|---|---|
| 1. POS integration depth | How loyalty connects to the till you already run: native integration, middleware, or a separate device | A second tablet beside the POS; staff re-key each sale and the two systems never reconcile | Certified integration with your exact POS product; earning and redeeming happen inside the normal transaction flow |
| 2. Enrollment friction | What a customer must do to join, measured in steps at the counter | Download an app, create an account, verify an email — while a queue builds behind them | One scan or tap on their own phone, no app and no password; the pass or card is on their phone before payment completes |
| 3. Channel reach and message cost | Which channels reach members after enrollment, and what each additional message costs | One rented channel with per-message fees, so every campaign is a variable cost that scales with success | Owned surfaces with no marginal message cost, such as wallet pass updates and lock-screen notifications, plus email — frequency becomes a strategy choice, not a budget line |
| 4. Data ownership and portability | Who legally owns the member list and how completely you can take it out | The vendor owns the list; export is on request, aggregated, or behind a fee | The contract states you own member and transaction data, with self-serve export in open formats at any time |
| 5. Migration path | Whether existing members, balances, and tiers can move in without a reset | No import path — every current member restarts at zero and discovers it at the counter | Documented bulk import for members, balances, and tiers, with a dry run against a copy of your data before cutover |
| 6. Counter and redemption experience | What staff do when a member redeems, and how long training takes | Staff switch systems, look the member up, and apply the discount by hand while the line waits | Redemption is applied inside the POS tender flow in one step; a new hire can run it after one shift |
| 7. Pricing structure and exit cost | How fees scale as the program grows, and what leaving actually costs | Multi-year lock-in, per-member fees that punish growth, and vague overage and exit terms | Published pricing, notice periods you can live with, and a stated no-ransom path for your data and live passes if you leave |
| 8. Reporting and measurement | Whether you can tie the program to revenue rather than activity | Vanity totals only: points issued, members joined, messages sent | Member versus non-member visit and spend comparison, redemption rate, repeat-visit cohorts, and raw export for your own analysis |
| 9. Vendor stability and support | Whether the vendor will still be there, and reachable, in year three | No SLA, ticket-only support with no response commitment, and a changelog that went quiet months ago | Public status page, active release notes, a named support path, and reference customers in your vertical you can call |
Set your weights first
Write a weight next to each criterion before any vendor demo. All nine must total exactly 100. The hints describe when a criterion deserves a heavy weight — ignore them freely; they are prompts, not defaults.
Weight — POS integration depth
Weight heavily if your counter is high-volume and double-keying sales is unacceptable; lower if a standalone flow is tolerable at your pace.
Weight — Enrollment friction
Weight heavily if walk-in customers join at the counter, where every extra signup step costs enrollments; lower for appointment businesses that enroll during booking.
Weight — Channel reach and message cost
Weight heavily if you plan frequent outreach. Per-message fees compound every month; channels you own outright do not.
Weight — Data ownership and portability
Rarely weight this low. The member list is the asset the whole program exists to build, and it outlives any one vendor.
Weight — Migration path
Weight heavily only if you have an existing program with balances customers care about; near zero for a first program.
Weight — Counter and redemption experience
Weight heavily if staff turnover is high or staffing is seasonal — the program has to survive being run by someone hired last week.
Weight — Pricing structure and exit cost
Weight heavily if cash flow is tight or you are testing the loyalty category for the first time and want a cheap exit if it fails.
Weight — Reporting and measurement
Weight heavily if someone will actually review the numbers monthly. A program you cannot measure cannot be defended at renewal.
Weight — Vendor stability and support
Weight heavily if you have no technical staff and the integration simply has to keep working without you.
Check: the nine weights total exactly 100
Freeze the weights before the first demo. Re-balancing weights after scores exist is how committees quietly rationalize a favorite.
The scoring grid
Copy your weights into the second column, then enter each vendor's 1-to-5 score. A vendor's weighted total is the sum of weight times score across all nine rows: maximum 500, floor 100. One grid per scorer keeps individual judgments visible.
| Criterion | Your weight | Vendor A (1-5) | Vendor B (1-5) | Vendor C (1-5) |
|---|---|---|---|---|
| 1. POS integration depth | ____ | ____ | ____ | ____ |
| 2. Enrollment friction | ____ | ____ | ____ | ____ |
| 3. Channel reach and message cost | ____ | ____ | ____ | ____ |
| 4. Data ownership and portability | ____ | ____ | ____ | ____ |
| 5. Migration path | ____ | ____ | ____ | ____ |
| 6. Counter and redemption experience | ____ | ____ | ____ | ____ |
| 7. Pricing structure and exit cost | ____ | ____ | ____ | ____ |
| 8. Reporting and measurement | ____ | ____ | ____ | ____ |
| 9. Vendor stability and support | ____ | ____ | ____ | ____ |
| Weighted total: sum of (weight x score) | 100 | ____ of 500 | ____ of 500 | ____ of 500 |
Running it with a committee
The scorecard only makes the decision auditable if the process around it is disciplined. Six steps, in order.
- 1
Freeze the weights before any demo
Each stakeholder proposes weights independently, then the group negotiates one shared set totaling 100. Record who argued for what — that record is part of the audit trail.
- 2
Score independently, same day
After each demo, every scorer fills their own grid before anyone discusses the vendor. First impressions converge fast once the loudest person speaks; independent scores preserve real disagreement.
- 3
Surface the variance
For any criterion where individual scores spread by two or more points, stop and talk. A wide spread usually means people saw different demos, or the criterion definition drifted — both are worth catching now.
- 4
Demand evidence for every 4 and 5
A top score must cite something checkable from the evidence list below. 'The salesperson said so' caps a criterion at 3.
- 5
Compute the weighted totals
Multiply each score by its weight and sum the column. Close totals are a finding, not a failure — they tell you the decision turns on one or two criteria, so go re-verify those.
- 6
Write the one-page decision memo
Attach the final grid, name the winner, and state in one paragraph why it won and what the runner-up would have needed to change the outcome. This memo is what makes the decision defensible in a year.
Evidence rules: what a 4 or 5 must be backed by
Scores are only as good as what they rest on. Before any criterion gets a top score, check off the evidence that supports it.
The demo ran on your POS, not a slideshow
Insist the integration demo uses your exact POS product and version. 'We integrate with everything' collapses quickly at this step.
You completed a full earn-and-redeem loop yourself
Get a sandbox or trial account and go from enrollment to redemption as a customer would, on your own phone.
You timed the counter transaction
Run an earn and a redeem against a stopwatch and compare with your current average transaction time. Scores for criterion 6 should cite the number.
Data ownership is in writing
A verbal 'of course it is your data' is not evidence. The claim belongs in the contract or the data processing agreement.
You opened a real export file
Ask for a sample export and confirm members, contact details, balances, and transaction history are all present — not just an email list.
Migration got a dry run
Import a slice of your real member list into a test environment before scoring criterion 5 above a 3.
You called a reference, not read one
One phone call with a customer in your vertical and size range outweighs any number of logos on a website.
The quote covers year two
Get renewal pricing, overage rules, and exit terms in writing. A year-one discount is not the price of the program.
You found the status page yourself
Uptime history you can view without asking is evidence; an uptime claim in a sales deck is not.
You sent a real support ticket before signing
File a genuine pre-sales technical question through the normal support channel and note the response time and quality.
You read six months of release notes
An active, dated changelog is the cheapest reliable proxy for whether the product is still alive.
You asked the cancellation question directly
Ask exactly what happens to your members' passes, cards, and data on the day you cancel — then score criterion 7 on the specificity of the answer.
From the guide: POS Loyalty Program: Add Loyalty Without Switching POS
This resource accompanies the full article — worth reading before you commit to a tool.