Loyalty tier design: thresholds people can actually reach

Aug 22
3 min read

Updated: 3 days ago

Introduction

A flat loyalty scheme treats a customer who visits twice a year the same as one who visits weekly. Tiers exist to fix that, and they work by giving people something to move toward.

They also fail in predictable ways: thresholds nobody can reach, rewards nobody wants, and complexity nobody can explain at the counter.

1. Loyalty tier design begins with your actual spend distribution

Thresholds must come from your own data, not from a competitor's scheme.

Pull a year of customer spend and find the quartiles. The entry tier should include a large share of existing customers, the middle tier should be reachable by someone who increases slightly, and the top tier should describe your genuinely best customers.

Setting the top tier above what anyone currently spends is the most common mistake. It reads as decorative, and a threshold nobody believes they can reach has no motivating effect at all.

2. Three tiers, not five

Each additional tier adds explanation and reduces the clarity of the next step.

Three works: one that most customers enter easily, one that requires a modest increase, one that rewards your best. Everyone can hold that in their head, including your staff.

Five-tier schemes are usually designed to look substantial. In practice customers cannot say which tier they are in, and a benefit nobody can recall does not influence behaviour.

3. Make the next step visible and close

The mechanism is proximity. A customer who knows they are near a threshold behaves differently.

Show progress every time they interact: how far into the current tier, what the next one gives, what remains. "Two more visits" is a specific, achievable instruction; "join our rewards programme" is not.

If you cannot display progress, tiers will underperform a simple stamp card, which at least shows a half-full card.

4. Reward with access and certainty, not just discounts

Discounts are the default and the weakest option, because they reduce margin on customers who were already loyal.

Better rewards cost less and mean more: priority booking, skipping a queue, a known contact, early access to something new, a held table, free delivery, an extended return window.

For service businesses, the strongest reward is usually responsiveness — a guaranteed slot or a faster turnaround. It costs scheduling, not margin.

5. Price each tier's benefit against its margin

Every tier needs a cost per customer, calculated before launch.

Take the average annual margin from a customer in that tier and check the reward is a small fraction of it. Then check the worst case: the customer who uses every benefit to its limit.

Tiers usually fail on margin at the top, where the rewards are most generous and the customers most active. That combination is exactly where the arithmetic needs to be checked.

6. Decide how status is lost, and say so upfront

Tiers require a reset rule or everyone accumulates permanently and the top tier fills with people who spent heavily two years ago.

The standard approach is a rolling twelve-month window: status reflects the last year. Alternatives include annual resets or a slow decay.

Whatever you choose, state it clearly at sign-up. Silently downgrading someone produces more resentment than never having offered the tier.

7. Keep it simple enough for staff to explain

The scheme lives or dies at the point of interaction, and the person explaining it is usually busy.

Test it: ask a member of staff to describe the tiers, the thresholds and the benefits without notes. If they cannot, customers certainly cannot, and the scheme will not be mentioned.

One sentence per tier. If a tier needs a paragraph, simplify it before launch rather than training harder.

8. Know when tiers are the wrong answer

They suit businesses with frequent purchases and variation in customer value. They do not suit everything.

If most customers buy once, or once every few years, tiers have nothing to accumulate — a referral incentive or a straightforward guarantee will do more. If nearly all customers spend similar amounts, tiers create administration without differentiation.

Run it as a test with an end date and a comparison group before committing. Retention schemes are easy to launch and awkward to withdraw, so establishing that it works while it is still small is worth the discipline.

Conclusion

Set thresholds from your own spend quartiles so every tier is believably reachable, and stop at three tiers.

Show progress toward the next one, reward with access and certainty rather than margin-eroding discounts, cost each tier against worst-case usage, publish the rule for losing status, keep it explainable in one sentence per tier, and test whether tiers suit your purchase frequency at all.