Loyalty programmes and subscribe-and-save at a glance
A loyalty programme is, in the words of the loyalty programme design guide, a trade: the business returns value to customers in exchange for behaviour it wants, such as more frequent purchases, spending consolidated from competitors or a first purchase in a new category. That guide compares the points, tiers, paid membership and community models, sets out what is worth rewarding, and works through reward liability and breakageThe share of issued loyalty rewards that customers never redeem, because points expire, balances are too small to use or members forget.. Its central warning is that programmes tend to recruit customers who already buy often, so much of the reward goes to behaviour that would have happened anyway.
Subscribe-and-save, as the subscribe-and-save guide describes it, lets a customer agree once to receive a product on a regular cycle, usually at a lower price, so the replenishment order happens without a reminder. It suits consumables used up at a steady rate, where running out is a nuisance and the customer wants the same item each time. Its costs last as long as the subscription does: a standing discount paid on every delivery, churnShare of customers at the start of a period who left during it. that has to be counted honestly and a cancellation route that regulators expect to be easy.
So the first question is not which scheme is more modern but how customers already buy. If they come back for the same product on a steady rhythm, the store is mostly trying to stop them lapsing or drifting to a competitor at the moment of reorder, and a subscription removes that moment. If they buy different things at irregular times, there is no cycle to agree in advance, and the open question is whether a reward can make them buy more often or more widely. The repeat purchase and loyalty guide maps those paths, from replenishment to cross-category buying, and is the place to find which ones a catalogue has.
Compare table
Loyalty programme vs Subscribe-and-save
| Row label | Loyalty programme | Subscribe-and-save |
|---|---|---|
| What the customer agrees to | To join and identify themselves when buying; each purchase is still their decision | To receive a product on a cycle and be charged for each delivery until they change or cancel |
| What is rewarded | Behaviour after it happens: spend, frequency, new categories, reviews or referrals | The commitment itself: a standing discount or perk on every scheduled delivery |
| Which products suit it | A range bought on the customer's own schedule, where they have a choice of where to buy | Consumables used up at a steady rate, where the same item is wanted each time |
| Cost on orders that would have come anyway | Rewards earned by regulars buying at their usual frequency | The discount on deliveries the customer would have reordered at full price after a reminder |
| Liability | Rewards earned but not yet redeemed, owed to members until used or expired | No balance owed, but a promise to supply every scheduled delivery |
| Cancellation and terms | Clear rules on earning, expiry and changes to the scheme | Clear terms before sign-up, consent to repeated charges and cancellation as easy as joining |
| Data gained | Identified purchases across the range, joined up over time | A forecast of demand for the subscribed lines and a clear churn signal |
| Effect on purchase frequency | May raise it where customers have a choice; often rewards frequency that already existed | Fixes it to the agreed cycle, which can be slower or faster than the customer's own |
| How it is measured | Incremental margin among members against a comparison group, net of rewards and running costs | Subscribers read by cohort at fixed ages, against similar customers who stayed on reminders |
| Typical misuse | Launching points to fix an experience customers leave because of | Copying a competitor's discount or pre-selecting the subscription at checkout |
The rows that decide most cases are the third and the fourth. A subscription needs a product the customer will genuinely want again on a cycle; without that, it produces skips, pauses and cancellations rather than orders. Both schemes then carry the same hidden cost in different forms: the loyalty programme rewards regulars for buying as they would have, and the subscription discounts deliveries that a well-timed reminder would have brought at full price. Neither cost shows up in sign-up or membership counts.
When to choose a loyalty programme
Choose a programme when customers buy across a range, on their own timing, in a category where they genuinely shop around, and when there is a specific behaviour the business wants to change: a shorter gap between orders, a first purchase in a second category, spending consolidated from a competitor. The loyalty guide's rule is to reward behaviour the business wants more of and would not otherwise get, and its list of what not to reward begins with purchases that would happen anyway. A programme can also do a data job, giving customers a reason to identify themselves at every purchase, which matters most where transactions would otherwise be anonymous.
Before committing, work out how much extra behaviour the programme has to create to cover what it gives away. The loyalty guide frames that question with the break-even calculator: rewards actually redeemed plus technology, staff and communication, set against the margin from behaviour the programme changed. The repeat purchase guide adds the measure that keeps the answer honest: a holdout group or a like-for-like comparison, because members are usually the customers who were already buying most.
A programme is the weaker choice when the category is low-involvement and customers already buy out of habit, when margin is too thin to give any back, or when customers leave because of the experience. The loyalty guide lists those signs, and in each case a reward adds cost without removing the reason customers go. For a single consumable with a steady cycle, a programme also rewards each reorder after it happens, where a subscription would have secured it in advance.
When to choose subscribe-and-save
Choose a subscription when the store's own order data shows a line whose reorder gaps cluster tightly, which is the test the subscribe-and-save guide sets: look at customers who bought the product more than once, how long they waited, how much that gap varies and what share reordered at all. Where gaps cluster, a standing order matches the need, takes away the moment when a competitor's advert might catch the customer and gives the store a forecast of demand. Where gaps are scattered, the guide sends the line to a replenishment reminder instead, built in the e-commerce email and SMS flows guide.
Set the offer from the store's own margin, not from a competitor's. The standing discount is paid on every delivery for as long as the customer stays, so the guide works out what each delivery leaves after product, packing, delivery and the discount, as contribution marginRevenue left after every cost that varies with each sale., and compares that with what the same customers bring in through reminders. The comparison of gross margin and contribution margin explains why the second figure is the one that should set the limit. Free delivery, first access to stock or simply never running out may carry much of the value at a lower cost than a percentage.
A subscription also carries obligations a programme does not. It takes payment repeatedly without a fresh decision, which is why the subscribe-and-save guide records that regulators such as the Competition and Markets Authority and the Federal Trade Commission treat clear terms, consent to charges and easy cancellation of subscriptions as consumer protection matters. That guide treats their published guidance as general direction rather than legal advice, notes that rules differ by country, and asks for the design to be checked by someone qualified before launch. The guide to designing the cancellation experience covers the leaving route, from the confirmation and final charge to one optional exit question.
Decision table
| Situation | Choose | Why |
|---|---|---|
| One consumable is reordered on a steady, predictable cycle | Subscribe-and-save for that line | A standing order removes the moment at which customers lapse or switch |
| Customers buy across a wide range at irregular times | Loyalty programme | There is no cycle to agree, and a reward can target frequency or a new category |
| Reorder gaps vary widely from customer to customer | A replenishment reminder, not a subscription | A fixed cycle would produce skips and cancellations rather than orders |
| Customers leave because of delivery, product or support problems | Neither yet | Rewards and standing orders add cost without removing the reason to leave |
| Margin is too thin to fund a standing discount or rewards | A perk such as free delivery, or neither | Both schemes give margin back on orders that would have come anyway |
| The business needs purchases identified across channels | Loyalty programme | The data job is one a subscription does not do outside the subscribed lines |
| A store sells one fast-moving consumable and a slower wider range | Both, with a boundary between them | The subscription secures the consumable; the programme targets the range |
Running both: subscribers inside a loyalty programme
Many stores that sell consumables alongside a wider range end up running both, and the design questions sit at the boundary. The first is whether subscription deliveries earn rewards. A subscriber is already paid, through the discount or perk, for orders that now arrive by default; points on top return margin twice on the same order. Treating the subscription as the reward for that line, and letting the programme apply to purchases outside it, keeps each scheme doing one job. Where the store does let deliveries earn, the terms of both schemes should say so.
The second is who gets which invitation. RFM segmentation sorts existing customers by recency, frequency and spend, and the frequent buyers of a single consumable are the natural candidates for a subscription offer, while customers who buy widely but irregularly are the ones a programme can still move. The subscribe-and-save guide's own rule points the same way: invite customers whose reorders already follow a steady rhythm, because their behaviour shows the cycle is real.
The third is consistency. A loyalty promotion that offers the subscribed product at a lower one-off price than the subscription gives subscribers a reason to cancel, which the subscribe-and-save guide lists among its common mistakes. Rules on rewards, expiry and changes to the scheme need the same clarity as subscription terms: the loyalty guide records that UK consumer protection law, which the Competition and Markets Authority enforces, expects terms to be fair and transparent and marketing not to mislead, and it too treats that as general direction rather than legal advice. When a subscriber cancels, the leaving route should say what happens to any points balance.
Measure the two schemes separately and on the same basis. Subscribers are read by cohort at fixed ages, as the retention hub sets out for any customer group, with purchase frequencyAverage number of orders each customer placed in the period. and repeat purchaseShare of customers who placed more than one order in the period. compared with similar customers left on reminders. Programme members are compared with a holdout, and the loyalty programme pilot test designs that comparison for a programme not yet launched, with the invitation randomised. The joins between orders, billing and customer records that this needs are covered in the subscription measurement guide, and a save offer at the cancel step can be tested on its own with the cancellation save offer test, which keeps easy cancelling as a guardrail. Whether money spent on either scheme would do more in acquisition is the question the comparison of customer acquisition and retention answers.
A skincare brand chooses between points and a refill subscription
Kevruax Skincare, an invented online skincare brand, sells one fast-moving cleanser at £18, which leaves £7.20 of contribution before any discount, and a wider range of serums and masks bought a few times a year, whose orders average £30 and leave £13. Its order data shows that customers who buy the cleanser more than once reorder every seven to nine weeks, with little variation; the wider range shows no pattern at all.
It tests both schemes on 3,000 regular cleanser buyers, split at random into three groups of 1,000 for six months. One group is invited to a refill subscription for the cleanser at 10% off, or £1.80 a delivery. One is invited to a points programme returning 5% of spend across the range. The third keeps only the existing replenishment reminder and places 2,400 cleanser orders, worth £17,280 of contribution.
The subscription group places 2,900 cleanser orders: 1,200 subscription deliveries and 1,700 one-off orders, worth £18,720, or £1,440 more than the holdout after £2,160 of discount. In the points group, 520 customers join and spend £31,000, earning £1,550 of rewards, of which £1,100 is redeemed; with £700 of platform and email costs, the programme costs £1,800. It adds 60 cleanser orders and 90 range orders, worth £1,602, and leaves £450 of rewards still owed.
Kevruax launches the subscription for the cleanser, with subscription deliveries outside the points scheme, and drops points on spend. It plans a narrower test of a reward for a first purchase in the wider range, the only place the points group showed behaviour that the holdout did not.
Where the points route is chosen, the break-even calculator set up for a loyalty programme values rewards owed in full and shows the extra orders the scheme must cause.
Common mistakes
- Choosing by what competitors run. Start from how customers already buy: a steady cycle on one product, or a range on their own schedule.
- Counting members and sign-ups as success. Both schemes are judged on the orders and margin they added against a group that was not offered them.
- Forgetting the cost on orders that would have come anyway. Rewards to regulars and discounts on deliveries a reminder would have brought are both real costs.
- Subscribing products without a steady cycle. Scattered reorder gaps produce skips and cancellations; a reminder serves them better.
- Paying twice on the same order. Decide whether subscription deliveries earn rewards, and say so in both sets of terms.
- Designing either scheme to depend on customers failing to act. A programme that needs high breakage, or a subscription that relies on hard cancellation, is built to disappoint the people it is meant to keep.
- Promoting a cheaper one-off price to subscribers. A loyalty offer that undercuts the subscription gives subscribers a reason to leave.
- Treating terms as an afterthought. Earning, expiry, repeated charges and cancellation all need to be plain before customers join.
The comparison library lists the other pairs that are easy to confuse.
