Loyalty Cards: What Actually Changes Customer Behaviour (And What Doesn’t)

The loyalty card proposition sounds mutually beneficial: you spend money, you collect points, you get rewards. The business encourages repeat visits and learns more about your preferences.

But if you look carefully at most loyalty programmes, they’re not really about changing behaviour. They’re about rewarding behaviour that was already happening. The supermarket shopper who was going to spend £200 per week anyway is now collecting points — the business is paying for data and the appearance of loyalty without creating any additional loyalty.

The programmes that actually work — the ones that generate genuine incremental revenue — are the ones that understand this distinction and design around it.

The Basic Types

Points accumulation programmes: The classic model. Spend money, earn points, redeem points for discounts or goods. Tesco Clubcard, Boots Advantage Card, Nectar. The mechanics are simple; the design question is whether the earn rate and redemption value are enough to change what you would have done anyway.

Tiered programmes: Spending crosses thresholds that unlock status levels with progressively better benefits. British Airways Executive Club (Bronze/Silver/Gold), hotel programmes (Marriott Bonvoy’s Silver/Gold/Titanium). The psychological mechanism is different from flat points: people spend to reach the next tier, to maintain a tier they’ve achieved, and to avoid dropping below one. The loss-aversion around tier status is a more powerful motivator than points accumulation alone.

Stamp or punch card programmes: Spend ten times, get something free. Common in coffee shops and local restaurants. Simple, no technology required, easily understood. The constraint is that there’s no data capture and no ability to personalise.

Coalition programmes: One loyalty currency that works across multiple partners — Avios, Amex Membership Rewards, Nectar across multiple retailers. The value is the earn breadth; you accumulate faster because the programme is useful in more contexts.

Subscription-based loyalty: Amazon Prime, Costco, various premium loyalty tiers at retailers. You pay an upfront fee in exchange for ongoing benefits. The psychology is different again — having paid the fee, you’re incentivised to use it to justify the cost. These generate predictable revenue and high engagement from members who have opted in.

The Behavioural Economics of Points

The reason loyalty programmes work as well as they do is psychological, not purely rational. Points produce several effects that cash discounts don’t.

The endowment effect and loss aversion: Once you have points, you don’t want to lose them. Expiry policies that threaten to remove accumulated points create urgency. The fear of losing something you already have is a stronger motivator than the prospect of gaining something equivalent.

Reward deferral: Points create a reward you can see accumulating but that you don’t take immediately. This deferred gratification has a different psychological texture from a 5% discount applied immediately — it feels more like saving and earning than discounting.

Aspiration and hedonic rewards: The classic study on this showed that people who accumulated credit card points preferred to redeem them for luxury or experience items — restaurant meals, travel, spa treatments — rather than cash equivalent. The points create psychological permission to spend on things people wouldn’t otherwise feel justified buying.

The near miss effect: The psychological pull of being close to the next reward, the next tier, or the next threshold is disproportionate to the actual value of getting there. Programmes that show you how close you are (“you need 200 more points for your next reward”) leverage this consistently.

When Loyalty Programmes Actually Drive Incremental Behaviour

The research on whether loyalty programmes generate incremental revenue is more mixed than the industry likes to admit. Several conditions seem to make the difference:

Meaningful rewards that require real effort to obtain, but aren’t too distant: If the reward is too easy (effectively just a discount), you’re not changing behaviour. If it’s too remote (thousands of purchases before any meaningful redemption), engagement drops. The design challenge is calibrating earn rates and redemption thresholds to the right zone.

Category exclusivity or near-exclusivity: Loyalty programmes work better when there’s a limited number of competitors in the category. A coffee shop loyalty card works partly because people who commit to it go to fewer competing shops. Grocery loyalty in a market with three comparable options is fighting harder for the same effect.

Status with social visibility: Tier programmes where your status is visible — lounge access, priority boarding, visible tier names — create social signals that reinforce the value of maintaining status. This is why airline programmes generate the intense engagement that grocery programmes don’t.

Personalised offers based on actual behaviour: A loyalty programme that knows you buy a specific brand of wine every Friday and sends you a wine offer on Thursday evening is genuinely useful. A programme that sends the same email to everyone is treating its data as a cost centre, not an asset.

The Data Question

For many businesses, the real value of a loyalty programme isn’t the repeat purchase behaviour — it’s the data.

Knowing that a specific person bought nappies last month, a new car seat two months ago, and children’s books most recently tells you something specific about that customer’s life stage that anonymous transaction data doesn’t. Tesco’s Clubcard data is consistently cited as one of its most valuable strategic assets — the insight into purchasing behaviour across 18 million households is worth significantly more than the cost of the points it distributes.

This creates a responsibility. Customers who participate in loyalty programmes implicitly trust that their data is being used in ways they’d broadly accept. Programmes that use purchase data to set higher prices for loyal customers (a pattern that’s emerged in some research on dynamic pricing), or that share data in ways customers wouldn’t expect, are eroding the trust that makes the loyalty relationship work.

Getting Value as a Consumer

If you’re evaluating whether to participate in a loyalty programme:

Calculate the actual earn rate: A programme offering 1 point per pound spent with 100 points = £1 reward is returning 1%. Whether that’s worth the effort of carrying a card, providing your data, and modifying any behaviour depends on your spending volume and alternatives.

Understand the redemption landscape: Points have different values in different redemption contexts. Airline miles transferred to partner airlines for business or first class redemptions can be worth 3–5× more per point than the same miles redeemed for economy flights or statement credits. Understanding the highest-value redemptions in a programme unlocks significantly more value.

Tier programmes deserve genuine calculation: Before spending to reach or maintain a tier, calculate what you’d need to spend, what the tier benefits are worth to you specifically (not the headline value, but the benefits you’d actually use), and whether the incremental spending is justified. For frequent travellers, airline tier status is often clearly valuable. For occasional travellers, the calculation is less clear.

Don’t change behaviour that costs you more than the reward: If you’re buying from a loyalty partner instead of a cheaper alternative to earn points, confirm the points value exceeds the price difference. Often it doesn’t.

The Design View

For businesses considering loyalty programme investment, the key design questions are:

1. Are you trying to attract new customers, retain existing ones, or both? These are different problems and require different mechanics.
2. What behaviour are you specifically trying to reinforce? Purchase frequency? Category trial? Referral? Design the earn structure around the behaviour, not around the points.
3. What data do you need, and what are you willing to give customers in exchange for providing it?
4. What does your redemption economics look like? Points that are never redeemed don’t cost you anything but also don’t create loyalty. Points that are all redeemed can destroy margin. The float on unredeemed points is real, but breakage rates that are too high suggest the programme isn’t valuable enough to drive engagement.

The best loyalty programmes feel like a genuine exchange — valuable enough that customers participate by choice rather than by inertia, data-informed enough to deliver experiences that feel relevant rather than generic. Most programmes achieve this inconsistently. The ones that do it consistently tend to be the ones customers think about as a feature rather than a card in their wallet they forget is there.