Retention comes up in a planning meeting, and within about a minute someone says it: we should do points. It’s the reflex answer, and it’s so automatic that the actual decision usually goes unmade.

Points are a mechanism, not a strategy. So are memberships. Both are attempts to make the next purchase more likely, but they work on opposite ends of the transaction, and they suit different catalogs, different margins, and different customers.

The merchants who get the most out of retention programs are the ones who picked the mechanism on purpose.

This isn’t an argument that one type of program “wins”. Instead, we’ll walk through the questions that tell you which type of program you should build — whether a Shopify loyalty program or membership — even if the outcome is “go install a points app, this article isn’t for you.”

What a points program does

A points program converts a purchase into a small deferred claim on a future purchase. You spend, a balance goes up, and at some threshold the balance turns into something. The reward is retrospective: it acknowledges what already happened.

The mechanism it leans on is progress. People work harder as they get closer to a goal — Kivetz, Urminsky, and Zheng demonstrated this in live loyalty programs, where customers purchased more frequently the nearer they got to a reward. Starbucks stars, airline miles, hotel nights, and a coffee shop’s paper punch card are all the same idea wearing different clothes: today’s purchase makes tomorrow’s reward a little closer. (This is also why a card with two squares already stamped feels so different from an empty one, even if both need the same number of visits!) Tiers stack a second progress bar on top of the first, which is why status ladders are such a durable feature of these programs.

What’s genuinely good about points is how little they ask: there’s no fee to evaluate, no decision about whether joining is worthwhile, and no change to your product model. Customers simply begin earning, often without opting into anything at all. A points program overlays onto any catalog, and adoption friction is close to zero. That is a wonderful advantage.

What a membership does instead

A membership reverses the sequence. The customer joins first, and from that moment the relationship is different — they have access, recognition, or standing before the next purchase happens rather than after it. Nike’s program is the clearest large-scale version: membership isn’t a tier you climb into, it’s the default way to shop.

That reordering matters more than it sounds. Joining is a decision about identity, and it gets made once instead of being re-litigated at every checkout. We’ve written the full argument for why people join memberships rather than re-running it here, but the short version is that people join for access, recognition, and relief from deciding, and none of those three is a price.

The behavioral evidence points the same way. McKinsey found that members of paid loyalty programs are 60 percent more likely to spend more on a brand after subscribing, against 30 percent for free programs — and, more usefully for this comparison, that hard-value benefits like discounts drive sign-ups, while experiential benefits drive whether anyone renews.

How each one changes behavior

Set side by side, the two instruments differ on 5 areas:

• When the reward lands. Points pay out after the purchase. A membership pays out before it, and keeps paying out for as long as someone stays.
• What the customer is buying. Points buy a discount on some future order; a membership buys access and standing, which a competitor can’t reproduce by changing a number.
• What it does to price perception. Points keep the conversation on price, because the reward is denominated in it. Customers judge a price against a reference point rather than in absolute terms (St. Louis Fed), and a standing member discount quickly becomes that reference; we unpack that in why your membership should be more than a discount.
• What leaving costs the customer. Walking away from points costs an unredeemed balance, which is real but usually small. Walking away from a membership costs access and status, which is larger and ongoing — Amazon Prime is the extreme case, where leaving means rebuilding a habit rather than forfeiting a balance. We’ve covered the difference in shopper perception this creates in member discounts vs access.
• What it does to your P&L. Points accrue a liability that grows with volume; a membership books revenue that funds its own perks.

Points make the next purchase cheaper. A membership makes leaving expensive.

The money works in opposite directions

Here’s where the two instruments stop resembling each other. These are illustrations rather than benchmarks — plug in your own numbers, because the ratios matter more than the totals.

Model the points program

Take a store with 1,000 customers, placing 1,000 orders a year at an $80 average order value — so $80,000 in revenue. Offer 5 percent back in points and you’ve accrued $4,000 in reward value over that year.

Two things follow:

• The first is that the $4,000 sits on your books until it’s redeemed or expires. When it’s redeemed, it comes off the margin of a future order.
• The second is that the cost scales with your success: double the orders and you’ve doubled the accrual. A points program is an expense that grows in exact proportion to how well it works.

Some share of that balance never gets redeemed, which improves the accounting. Unredeemed points are worth looking at squarely, though — a customer who accrued a balance and then didn’t come back for it isn’t a saving, so much as a retention failure that happened to be cheap.

Model the membership

Same store, same 1,000 customers. Say 12 percent of them join a $9 per month membership. That’s 120 members generating $12,960 a year in recurring revenue that didn’t exist before.

Now subtract the perks. If the headline benefit is free shipping and those members place about three orders a year at $6 of shipping cost each, that’s $2,160 — leaving $10,800 in contribution. The other perks in a well-built program (early access, a members-only product, gated content, and a community space) are mostly about access, which means they cost roughly the same whether you have 120 members or 1,200.

There’s one caveat if you’re modeling a member discount instead of shipping: stacked discounts compound, so a 15 percent member perk on a product already 20 percent off nets around 32 percent off, eating further into margins. (You can disable stacking, though!)

The comparison to draw isn’t that points cost money and memberships don’t — it’s the direction of travel. One is a liability that grows as volume grows; the other is revenue that pays for its own perks and whose main costs stay flat as it scales.

For a high-frequency catalog, the growing liability of a points program can be a perfectly good trade — you’re buying incremental purchases you wouldn’t otherwise get. For a low-frequency catalog, you’re funding a discount for people who were going to buy anyway.

When a points program is the better choice

A real chunk of the people reading this should go install a points app anyway. Points are the better instrument when:

• The basket is small and the frequency is high. A running balance only motivates if customers come back often enough to watch it move. Coffee, groceries, cosmetics, pet supplies, and quick-service — a punch card works there because the customer is deciding where to go tomorrow morning.
• The category is price-driven and customers came for the number. If price is the main reason people chose you, a rewards balance reinforces the thing that’s already working.
• You want a referral program. If refer-a-friend is central to your plan, many points apps already offer it.
• You want to reward non-purchase actions — reviews, social follows, birthdays, and profile completion. Membership triggers don’t cover any of that.
• You don’t want to ask anyone to pay or sign up for anything. Passive, zero-friction opt-in is a points program’s home ground.

If two or more of those describe your store, a points program is the right tool. You could also consider using Zendra to do rewards instead of points — for example, giving a $5 store credit for every third purchase.

What each job needs from your tooling

These are two classes of software, and the distinction is worth holding onto while you shop.

• A points ledger needs earning rules, a redemption catalog, balance tracking, expiry handling, and usually referrals.
• A membership platform needs plans and billing, tiers, gating and access control, and a lifecycle that handles trials, failed payments, and cancellation. Different problem, different app.

Plenty of stores end up running both, and they coexist fine. The bridge is Shopify customer tags: Zendra tags customers automatically based on their active membership and maintains customer segments per plan, so a points app can read membership status and treat members differently without the two systems needing to know about each other.

If you’re specifically weighing Zendra against a points app, our comparison page lays out where each one’s job starts and stops.

How to get loyalty outcomes without a points ledger

Most merchants asking for points don’t want a ledger. They want what the ledger produces: recognition, escalating status, rewards for spending, and a reason to come back. Those outcomes don’t require points, so map each want to the thing that delivers it.

• “I want customers to earn their way in.” Grant membership by condition through Shopify Flow — spend $500 lifetime, become a member. That’s earned membership, and it’s the closest analogue to climbing a tier.
• “I want escalating status.” Tiers with distinct perks, pricing, and access rules, plus a Flow action that moves someone up a tier when they qualify.
• “I want to reward spend.” Flow can add store credit or generate a member discount code at a threshold. That pattern is already documented in why milestone rewards outperform flat discounts.
• “I want a members-only sale.” Perk scheduling gives member discounts start and end dates, so a sale is a bounded event instead of a standing expectation.
• “I want members to see what they’ve got.” The customer portal’s perks overview lists everything a plan includes, so members can see the full value of what they’re paying for in one place.
• “I want recognition in the store and at the counter.” Automatic customer and order tagging, managed segments, and membership details surfaced on the POS customer screen so staff can see who they’re talking to.

Tiers are where the status ladder lives, and each one carries its own perks, pricing, and access rules.

This reproduces the outcomes, not the ledger. There are no points balances, no earning rules, and no redemption catalog, and there’s no referral program. If you need those specific things, you need a points app.

How to pick between them

The decision usually resolves against three or four statements about your own catalog.

• High frequency, small basket, price-led category. Build a points program.
• Low frequency, higher basket, brand people feel something about. Build a membership; a running balance won’t move often enough to matter.
• You need recurring revenue rather than another discount line. Build a membership. Points don’t produce revenue; they spend margin.
• You have something worth gating. Build a membership. Access is the perk points can’t imitate.

If you landed on membership, our guide to starting a membership program on Shopify covers the structural decisions in order, and Zendra, our Shopify memberships app, handles the plans, tiers, gating, and billing once you’ve made them.

And if you landed on points, go build a great one. The mistake was never picking the other instrument; it was picking neither on purpose.