The dark side of loyalty schemes: when rewards actually cost you money

Start with the commercial objectiveLink

Do not launch a scheme because a competitor has one.

Loyalty schemes look harmless. A stamp card. A points app. A free dessert after six visits. A members-only room rate. Then the margin starts leaking.

The problem is not loyalty. Repeat customers are valuable. The problem is paying existing customers to do what they were already going to do. That is not loyalty. It is an unnecessary discount.

For independent pubs, restaurants, boutique hotels and B&Bs, the order matters. Build a reward scheme before you understand your numbers and you are not creating retention. You are moving money from your till into someone else's pocket.

Decide what behaviour you need to change. Choose one objective. One scheme cannot fix everything.

A pub trying to fill Sunday lunch needs a different mechanism from a B&B trying to increase direct repeat stays. A restaurant with a strong Friday trade should not reward Friday visits. Those customers are already buying.

Reward the gap. Not the habit.

• Fill quiet Tuesday and Wednesday services.
• Increase direct hotel bookings.
• Bring lapsed guests back within 60 days.
• Increase average spend per visit.
• Drive repeat bookings for events or private dining.
• Reduce dependence on commission-heavy booking channels.

Calculate the real cost before offering anythingLink

The headline reward is not the total cost.

A “free main course” costs more than the ingredient cost. It uses a table, staff time, utilities and capacity. If redeemed during a busy service, it may displace a full-paying customer.

Use this basic calculation: incremental contribution = additional profitable visits and spend − reward cost − technology and administration − cannibalised revenue.

Here is a simple restaurant example. The venue serves 1,000 transactions per month, with average spend of £32, a gross margin after food and drink cost of 68% (contribution per transaction before labour of £21.76), a loyalty reward cost of £4 per redemption and software and administration of £250 per month.

If the scheme creates 100 genuinely additional transactions, that produces approximately £2,176 in contribution. If 180 existing transactions also claim the reward, the reward cost is £720. Add the £250 operating cost. That leaves around £1,206 before extra labour, fraud, over-redemption or capacity costs.

Now remove the additional visits. If the same 180 customers would have visited anyway, the scheme has created no demand. It has simply cost the venue £970.

That is the dark side. The venue confuses activity with growth.

Stop rewarding full-price behaviourLink

A reward must change behaviour.

Do not offer 10% off every visit if customers already visit every week. Do not give points on Saturday night when the room is full. Do not discount your best-selling, highest-margin items without checking the consequences.

Use rewards where they have a job to do.

• Pub: reward a midweek food booking, offer a soft drink or bar snack with a quiet Monday meal, give early access to quiz nights or live events, reward a second visit within 30 days.
• Restaurant: offer a complimentary side with a booking on a low-demand day, give a fixed-value credit after a profitable spend threshold, reward direct bookings rather than third-party bookings, create a set-menu benefit during controlled service periods.
• Boutique hotel or B&B: offer a late checkout when operationally possible, add breakfast, parking or a welcome drink, provide a direct-booking credit on the next stay, give returning guests first refusal on popular dates.

These benefits often have lower marginal costs than a percentage discount. They also protect your published rate.

A £10 discount on a £100 room rate removes £10 from revenue. A late checkout may cost almost nothing on a quiet Sunday, provided housekeeping can absorb it. Do the operational maths first.

Protect your margin with contribution, not turnoverLink

Revenue is not profit.

A loyalty customer spending £50 is not automatically better than a non-member spending £35. The question is what remains after the cost of serving them.

Track contribution by customer and daypart. Do not celebrate points issued. Points are a liability until they produce profitable behaviour.

Do not celebrate redemptions either. A redeemed reward may represent a successful conversion or an avoidable discount. Your booking and EPOS data must tell you which.

HMRA's restaurant and hotel revenue management services are built around this distinction: fill demand, but protect the money left after serving it.

• Average spend.
• Food and drink cost.
• Labour cost during the redemption period.
• Booking channel commission.
• Reward cost.
• Visit frequency.
• Cancellation and no-show rate.
• Gross profit by offer.
• Profit from the next visit.

• See our restaurant revenue management service
• and our hotel revenue management service

Design the reward around a profitable actionLink

“Spend £100, get £10 off” is easy to understand. It can also be financially lazy.

A better structure creates a clear, profitable next action.

• “Book a Tuesday table within 21 days and receive a complimentary dessert.”
• “Stay direct twice in 12 months and receive £25 dining credit.”
• “Visit three times between January and March and unlock a midweek food benefit.”
• “Spend £45 per head and receive a £5 credit for your next off-peak visit.”

Set the threshold above normal behaviour, but not above realistic behaviour.

If the average spend is £42, a £45 threshold may increase basket size. A £75 threshold may simply frustrate customers. If the average hotel stay is two nights, rewarding the fourth night may be irrelevant for most guests.

The scheme should stretch behaviour slightly. It should not demand a different customer.

Do not build a discount-hunting machineLink

Bad loyalty schemes attract people who only appear when the reward does.

That creates three problems: they buy at the lowest possible price, they disappear when the offer ends, and they train your team and customers to treat the full price as negotiable.

Avoid permanent blanket discounts. Use targeted, time-limited benefits instead. A quiet-night reward is useful. A permanent 20% discount is a pricing error with an app attached.

Also avoid rewards that force unnecessary spend.

• Visit five times in a month to receive a reward.
• Spend £80 to unlock £5.
• Buy an expensive bottle to earn points.
• Book a room at an inflated rate to receive future credit.

These mechanics may increase short-term transaction value. They can also destroy trust and encourage customers to chase a reward rather than value the venue.

Your scheme should make profitable behaviour easier. It should not make the customer feel manipulated.

Keep non-members out of a pricing trapLink

Do not punish guests who refuse an app.

Some customers do not want another account, another notification stream or another data trail. Others may not have a smartphone. They are still potential customers.

The UK Information Commissioner's Office explains that loyalty programmes can collect data about what people buy, how often they buy and how much they spend. Read the ICO's guidance before collecting more information than you can protect or justify.

Use loyalty benefits as additional value. Do not make the standard price look artificially inflated.

A transparent member benefit is defensible. A “normal” price that exists mainly to make the member price look attractive is a trust problem.

• Read the ICO's guidance on data gathered by loyalty cards

Make the booking and identification process frictionlessLink

A loyalty scheme fails if staff cannot identify members quickly.

Do not force the team to search five screens while a queue forms. Do not make guests explain a complicated points rule at the till. Do not make the benefit dependent on a code buried in an email.

The customer should understand the benefit in one sentence. The team should apply it in seconds.

This is where the wider digital journey matters. A loyalty reward cannot rescue a broken booking flow. PubLandlord makes the practical point clearly for pubs: mobile booking buttons, tap-to-call and straightforward customer journeys remove unnecessary friction.

The same rule applies to restaurants and hotels. If customers cannot book, redeem or return without effort, the scheme becomes another abandoned process.

For broader landing-page and local conversion principles, JetAds offers a useful example of focusing on enquiries rather than attractive but inactive websites. And VU1 demonstrates the value of translating technical audits into plain-English actions. Your loyalty reporting should work the same way.

No jargon. No vanity metrics. Just covers, room nights, repeat bookings and margin.

• Email address.
• Phone number.
• Direct booking profile.
• Digital or physical card.
• Table booking reference.

• See PubLandlord
• and JetAds
• and VU1

Measure cohorts instead of cheering sign-upsLink

Sign-ups are not success.

A customer who joins, claims one reward and never returns has not become loyal. They have become an acquisition cost.

Compare members with similar non-members over the same period.

• Visit frequency before and after joining.
• Spend before and after joining.
• Margin before and after joining.
• Redemption rate.
• Second-visit rate.
• 60-day and 90-day retention.
• Direct booking share.
• Offer-led versus full-price visits.
• Incremental revenue by daypart.

Separate customers into groups: true loyalists who are already frequent, profitable customers; persuadable regulars likely to increase frequency with the right prompt; lapsed guests who need a reason to return; deal hunters who only appear for discounts; and low-value members who create administration without contribution.

Do not give every group the same offer. Your best customers may need recognition, not discount. Your lapsed customers may need a timely reminder. Your quiet-day prospects need a controlled reason to visit.

Review the scheme every 90 daysLink

Loyalty programmes become expensive because nobody shuts them down.

Review the numbers quarterly.

• Did profitable visits increase?
• Did average contribution rise?
• Did the scheme shift demand into quiet periods?
• Did it reduce third-party commission?
• Did customers redeem rewards they would have earned anyway?
• Did the offer create operational pressure?
• Did full-price sales fall?
• Are customers returning after the reward?

If the answer is no, change the mechanic or stop the scheme.

Do not defend a bad programme because you have already paid for the software. Sunk cost is not a strategy.

Use this loyalty ruleLink

A loyalty scheme is worth keeping only when it does at least one of these things.

Independent hospitality operators do not have supermarket-scale margins, data teams or spare capacity. You cannot afford a programme that looks busy while the P&L gets thinner.

• Creates an additional visit.
• Increases profitable spend.
• Moves demand into an underused period.
• Reduces commission or acquisition cost.
• Improves direct customer data with proper consent.
• Strengthens retention without damaging your base price.

If it does none of these, it is a discount with better branding.

Build the scheme in order: set the commercial objective, calculate contribution, identify the behaviour to change, design a low-cost reward, remove unnecessary friction, test it against a control group, measure repeat profit, and kill what does not work.

That is loyalty economics without the nonsense.

• Start with our free marketing audit