Do loyalty programs work? Why some quietly stop
Do loyalty programs work? On average, yes. Small-shop programs that fail tend to do it quietly, in one of four predictable ways, each with a fix.
On this page
- What the research says about whether loyalty programs work
- What “working” actually means for a small shop
- Four ways a loyalty program quietly stops working
- 1. Nobody asked anyone to join
- 2. The reward is too far away
- 3. It became a discount for people who were coming anyway
- 4. The relationship ends at the reward
- How to tell which one you have
- Is a loyalty program worth it for a small business?
- What keeps a loyalty program working
- Frequently asked questions
- Where to go next
Do loyalty programs work? On average, yes. A meta-analysis of thirty years of research found strong evidence that they make customers more loyal. But an average hides the programs that do nothing, and in a small shop those tend not to fail loudly. Nobody complains and nothing breaks. The card just stops changing anyone’s behavior.
The question worth asking is not whether loyalty programs work in general. It is whether yours is doing any work, and if not, which of the usual four things went wrong.
Key takeaways
- The research is broadly positive: a 2022 meta-analysis of 429 effect sizes found loyalty programs strengthen loyalty, more so in what people do than in how they feel.
- A program “works” when it changes a decision, usually bringing the next visit forward. A big member count proves nothing on its own.
- Programs that fail tend to die quietly: nobody asks customers to join, the reward is too far away, the card is a discount for people already coming, or the relationship ends at the reward.
- Each failure has a specific fix, and most do not require starting again.
What the research says about whether loyalty programs work
The best single answer comes from a meta-analysis by Alex Belli and colleagues in the Journal of the Academy of Marketing Science in 2022. They pooled 429 effect sizes from studies published between 1990 and 2020 and found “strong evidence that LPs enhance customer loyalty” (Belli et al., 2022).
Two details in that finding matter for a shop owner. First, programs move behavioral loyalty (visits and purchases) more readily than attitudinal loyalty (how much people like you). Second, the effect varied with how the program was designed and the industry it ran in. Design is the bigger lever, not the existence of a card alone.
A second study looked at the money side. Malika Chaudhuri, Clay Voorhees and Jonathan Beck followed 322 publicly traded firms that introduced a loyalty program between 2000 and 2015. Sales and gross profits rose in the first year and held for at least three, though the profit effect took until the second quarter after launch to show (Chaudhuri, Voorhees and Beck, 2019).
Neither study is about independent shops specifically, so read them as direction, not as a forecast for your shop. What they rule out is the cynical view that loyalty programs never do anything. What they leave open is the more useful question: why do some do nothing at all?
What “working” actually means for a small shop
A loyalty program works when it changes a decision a customer would otherwise have made differently. In practice that means coming back sooner, choosing you over the place next door on a busy morning, or returning at all after a first visit.
That rules out the numbers people usually reach for. A cumulative member count only goes up, whether the program is working or not. And members visiting more often than non-members proves less than it seems, because the customers who join tend to be the ones who were coming most already. That is selection, not cause.
A program that rewards people for what they were doing anyway is a discount, not a loyalty program. The four numbers that tell the two apart, and how to read them, are in loyalty program metrics.
Four ways a loyalty program quietly stops working
None of these tends to announce itself. Each shows up first as a number that stays flat while the owner assumes the card is doing its job.
1. Nobody asked anyone to join
A common failure happens before the program has a chance. A QR code by the counter with no one mentioning it converts a small share of customers, and a program with few members cannot change many decisions.
The sign: enrollment is low, but the people who did join use the card. The fix is a sentence, not a new program. Staff ask once, at payment, every time. How to promote a loyalty program covers the line, the sign, and the launch.
2. The reward is too far away
People work harder for a reward the closer they get to it. Ran Kivetz, Oleg Urminsky and Yuhuang Zheng showed this in a real café program: members bought coffee more often as they approached a free one (Kivetz, Urminsky and Zheng, 2006).
The flip side is that a reward which feels distant changes nothing today. A twenty-stamp card for a customer who visits twice a month is a ten-month promise, and nobody changes their morning for one of those.
The sign: people join, then stall in the first few stamps. The fix is a card length matched to how often your customers actually come. How many stamps should a card have? works through the arithmetic for each trade.
3. It became a discount for people who were coming anyway
Some programs have healthy numbers on the surface. Lots of members, lots of rewards claimed. But the regulars who already came daily are the ones redeeming, and nobody else has changed their habits.
Picture a bakery where the daily regulars redeem every reward while the weekly customers never finish a card. The sign: redemption is high while visit frequency is flat. The program is costing margin and buying nothing. The fix is usually a reward people value but that costs you little, and a card built to change the occasional customer’s behavior rather than thank the daily one. What a stamp card is actually for makes that case in full.
4. The relationship ends at the reward
This is the easiest one to miss. The customer finishes the card, claims the free coffee, and then nothing happens. Kivetz, Urminsky and Zheng saw this in the same café data: after the first reward, purchase rates reset to a lower level. They also found that customers who sped up more on the way to a reward were more likely to stay in the program and to start again sooner.
A card that is complete has no reason to pull anyone back. If the only thing connecting you to a customer is a finished card, the program has already done its last piece of work.
The sign: visits dip after redemption and some members simply disappear. The fix is what happens next: the new card started right away, a thank-you, and a message if the customer goes quiet. The first thirty days after joining and win-back messages that work cover both ends.
How to tell which one you have
Each failure has a different fingerprint, which is why one number is rarely enough to diagnose a program.
| What you see | Most likely failure | First fix | Why it matters |
|---|---|---|---|
| Few members, but they use the card | Nobody asked anyone to join | One line from staff at payment | Every other number is computed on members you do not have |
| Members stall in the first few stamps | Reward too far away | Shorten the card | A distant reward changes no decision today |
| High redemption, flat visit frequency | A discount for regulars | Rethink the reward and who it targets | You pay margin and buy no new behavior |
| Visits dip after the reward, members vanish | Relationship ends at the reward | Start the next card, follow up when they go quiet | A finished card has no reason to pull anyone back |
Give a program two full card cycles before judging it, or, for slower trades such as salons, a few months of watching how quickly members collect stamps. A change made in week three tells you nothing about what the first version would have done.
Is a loyalty program worth it for a small business?
In our view, for most independent shops with repeat trade, yes, on two conditions. The reward is priced at what it costs you rather than what it sells for, and the program is run rather than left by the register.
The cost is often smaller than owners expect. A free coffee costs you its ingredients and a minute of time, not its menu price, which is why the reward on a sensible card is a small share of the revenue that earned it. The arithmetic is in loyalty program metrics.
There are businesses where it is a harder case. If customers buy from you once every few years, a card has little to work with, and a good follow-up message or a referral reward may do more. If you already discount heavily, a program stacked on top of that can make the giveaway bigger without adding a visit.
What keeps a loyalty program working
The card gives a customer a reason to come back and a reason to tell you who they are. What keeps the program working after that is being able to reach them between visits.
That is also the honest answer to the fourth failure. A program that only lives on a card stops at the reward. A program connected to a way of reaching the customer keeps going, because the next visit has a reason even when the card does not.
Each channel has a different job here:
- Push, for the timely nudge. A reminder on the lock screen when someone is one stamp from a reward: “One more visit and your next coffee’s on us. See you this week?”
- Email, for staying connected between visits. A short monthly note with the new menu, the opening hours over the holidays, and a thank-you to members.
- SMS, for the message that matters. A win-back after weeks of silence: “We haven’t seen you since September. Your card’s still here, two stamps in. Come say hello.”
All three only reach customers who have opted in, which is why getting marketing opt-in is part of running the program, not an extra. Push, email or SMS covers which one earns its place for each message.
This is the part Passumo is built for, so take this as our recommendation and not a neutral one. Passumo gives you a wallet card, plus push, email, and SMS from the same place, with automations that send the reminder, the post-reward thank-you, or the win-back when the customer’s behavior calls for it. The card gives you the relationship. The follow-up keeps it alive.
Frequently asked questions
Do loyalty programs actually increase sales? On average, yes. A 2019 study of 322 publicly traded firms that launched programs (Chaudhuri, Voorhees and Beck, 2019) found sales and gross profits rose in the first year and held for at least three. Those were public companies, not independent shops, so treat it as direction for a small shop rather than a figure to plan around.
Are loyalty programs worth it for customers? When the reward is reachable and something they want, yes. If customers find your card is too long to finish, they stop collecting, and the program stops working for you as well.
Why do loyalty programs fail? Usually for one of four reasons: nobody asked customers to join, the reward is too far away, the card rewards people who were coming anyway, or nothing happens after the reward. Each has a different symptom and a different fix.
How long before I know whether mine works? Two full card cycles, at your customers’ real visit frequency. A ten-stamp card filled once a week takes about twenty weeks to cycle twice. For a salon or barber, where visits are weeks apart, judge it on the first few months of stamp pace instead.
Do I need to start again if mine is not working? Rarely. Most fixes are a change to the card length, the reward, the line at the counter, or what happens after redemption.
Where to go next
If you suspect one of the four failures above, the diagnosis starts with loyalty program metrics. If you have not launched yet, the full set of decisions, in order, is in the guide to digital loyalty cards.
