Start with a behaviour, not a Wallet metric
Measurement begins before the pass is designed. Define one observable behaviour and a period: a second visit within 30 days, a repeat purchase, a booking or use of a benefit. The number of issued passes helps explain the funnel, but it does not prove loyalty or return on its own.
- Target behaviour and time window
- Identifier connecting enrolment and outcome
- Benefit and operating cost
- Pre-agreed rule to keep, change or stop
Build the complete funnel
Separate each step to locate friction: QR or link exposure, enrolment start, issuance, available save signal, interaction, validation and commercial outcome. Wallet providers do not expose identical signals, so the measurement contract must distinguish observed, inferred and unavailable events.
- Attributed visit or scan
- Enrolment start and completion
- Pass issuance and technical state
- Click, code, redemption or business event
Do not confuse an update with attention
Apple supports creating, distributing and updating passes, while Google supports issuance through signed links and updates to defined fields. A successful system change does not mean every holder viewed the pass or a notification. Useful measurement therefore continues on an owned surface: a tagged link, validated code, booking or connected purchase.
Use a small event contract
Every event needs a name, exact trigger, identifier, source of truth and owner. A small stable set is usually more reliable than dozens of ambiguous events. For demand generation, GA4 recommends events such as generate_lead and qualify_lead; loyalty measurement should keep the same principle and name each action after what actually occurred.
- pass_enrolment_started: enrolment has begun
- pass_issued: the issuer created the pass
- benefit_redeemed: the business validated the benefit
- repeat_purchase: the commercial source confirms a new purchase
Compare cohorts, not accumulated totals
Group people by date, source, campaign or mechanic and compare their progress over the same window. In the K-Kare success story, Vuelvy connected post-purchase enrolment, the visible benefit and the later order. Recurrence moved from approximately 3–4% to 14–15%; the 250% result uses the conservative comparison between the previous 4% and the current 14%. Total issued passes were not presented as incremental revenue.
- The same window for every cohort
- Source and mechanic stored at enrolment
- Outcome confirmed by the commercial source
- Full cost included in the analysis
Turn learning into a decision
Review the funnel with one specific question. Exposure without enrolment points to value or capture placement. Enrolment without use points to the promise and next action. Use without economic return points to the benefit, audience or cost. A valid improvement changes one hypothesis at a time and retains the history needed for comparison.