Most loyalty dashboards look impressive in a board meeting and tell you almost nothing about what’s about to happen to your customer base like rewards issued this quarter, total members enrolled, redemption rate as a single blended percentage, etc. These numbers move up and to the right, they photograph well in a slide deck but none of them predict which members are about to stop showing up. The metric that actually predicts churn sits between two numbers most operators already track separately but rarely look at together: how long it takes a member to go from earning a reward to redeeming it, call it redemption velocity. It is the single clearest early signal of member intent that most restaurant loyalty programmes have access to and simply aren’t measuring.
Why the standard metrics miss the point
Rewards issued tells you how much activity happened but it doesn’t tell you whether that activity is converting into anything the member actually values. A customer can rack up rewards every week and never once feel like the programme is working for them becausOur e rewards sitting unused in an account create no loyalty on their own.
Redemption rate, as most dashboards report it, is a single aggregate number across the entire member base. It’s the kind of figure that can hold steady at a healthy-looking 40% while masking a split between members redeeming within days of earning and members who have three years of accumulated rewards they’ve never touched. Averaging those two behaviours together produces a number that is technically accurate and operationally useless and it tells you nothing about which members are moving toward the door. Member counts and enrollment growth are worse offenders. A programme can add ten thousand new members in a quarter while its existing base quietly disengages and the dashboard will still show growth. Enrollment measures acquisition, not health.
Redemption velocity measures something different: the gap in time between when a member earns enough for a reward and when they actually claim it. That gap is a direct read on intent because it reflects a decision the member is actively making, whether they’re conscious of it or not. A short gap means the programme is front of mind and a long or widening gap means it’s becoming background noise and that is exactly what precedes a member lapsing without ever formally cancelling anything.
Three velocity profiles, three different problems
Once you start tracking the earn-to-redeem gap at the individual member level rather than as a blended average, members sort into three distinct groups and each one needs a different response.
- Fast redeemers: These members claim rewards within days of becoming eligible. They’re engaged, they check their balance and the programme is functioning as designed for them. The risk with this group is complacency on the brand’s part: fast redeemers are often assumed to be safe and left alone, when in reality they’re the segment most likely to respond well to being asked for more, whether that’s a referral, a higher-tier upgrade or a request for feedback.
- Slow redeemers: This is the group carrying the earliest churn signal in the entire programme. They still earn rewards, visits haven’t stopped, but the gap between eligibility and redemption is stretching, quarter over quarter and that stretch usually shows up well before visit frequency itself starts to decline. A slow redeemer is in the process of becoming disengaged, which is precisely the window where an intervention still has a chance to work. Catching this group requires watching the trend in the gap, not the gap’s current value, since a member who’s always taken three weeks to redeem is behaving normally, while a member who’s gone from one week to five is telling you something.
- Never redeemers: These members have rewards sitting in their account that they have never claimed at all. Some of them don’t know the programme has rewards worth claiming, some enrolled for a one-time signup bonus and never engaged further an some have simply forgotten the account exists. This group is frequently written off as lapsed, but the more useful way to think about them is unaware rather than lost because the intervention that works for a genuinely lapsed customer (a win-back discount) is different from the intervention that works for someone who just needs to be shown clearly what they’re sitting on.
How Starbucks used redemption patterns
Starbucks Rewards is one of the few loyalty programmes that has been public about restructuring itself around redemption behaviour rather than simple participation numbers. Analysis of how members were actually earning and cashing in Stars showed that the original tiering created friction at exactly the point where members were most likely to disengage: the gap between crossing a rewards threshold and finding a reward worth redeeming it for.
The redesign moved toward more frequent, smaller redemption opportunities rather than fewer large ones, shortening the distance between earning and claiming across the member base. The logic wasn’t about issuing more rewards or advertising the programme more heavily but about reducing the earn-to-redeem gap itself, on the understanding that a shorter gap keeps the programme active in a member’s decision-making rather than something they engage with occasionally and forget about the rest of the time. The lesson generalizes past coffee: the redemption structure itself shapes velocit and velocity is more diagnostic than the redemption rate it eventually rolls up into.
Acting on velocity without a manual analysis project
Tracking redemption velocity at the individual member level, across three profiles, updated continuously, is not a report most teams can build and maintain by hand. It requires the same data most F&B groups already have (earn dates, redemption dates, rewards balances) restructured into a metric nobody’s dashboard is built to surface by default.
This is where segmentation and campaign automation do the work that would otherwise sit on an analyst’s desk. Instead of a quarterly manual pull to identify slow redeemers, Como can maintain that segment continuously as a live filter: members whose earn-to-redeem gap has widened over a defined period, updated as behaviour changes rather than recalculated once a quarter. The same applies to never redeemers who can be automatically routed into a campaign that simply surfaces what they’ve already earned and to fast redeemers, who can be automatically flagged for upsell or referral campaigns instead of being left alone by default. The point is removing the manual work of recalculating these profiles member by member, campaign by campaign, so the response to a widening redemption gap can happen while the member is still a slow redeemer and not after they’ve become a lapsed one.
In practice, this comes down to what Como is already recording for every gift and punch card in the system: the date it was issued or started, every punch or visit logged against it, the date it was completed and the date it was actually redeemed, if it has been at all. That data is what turns “redemption rate” from a single quarterly percentage into a live per-member gap. A completed card sitting at day 3 past completion looks different from one sitting at day 40 and Como can flag that difference automatically rather than requiring someone to export a list and check dates by hand. The same tracking applies to gifts issued outside a punch structure entirely, birthday rewards, win-back offers, referral bonuses, since each one has an issue date and a redemption date that behaves exactly like a completed card waiting to be claimed. Once that gap is visible per member and updating on its own, building a segment around it, whether that’s everyone who finished a card in the last two weeks and hasn’t come back or everyone sitting on a gift issued a month ago becomes a filter rather than a research project.
Rewards issued and redemption rate will keep showing up in every loyalty report because they’re easy to calculate and easy to present. Redemption velocity is harder to compute and far more useful, precisely because it measures the gap where a member’s relationship with the programme is actually decided.
