A loyalty program only works for a small online store once the second order is in good shape. Start with a good flow after the first purchase or the sample, measure how many customers come back, and only then choose between reward points, a subscription or a fixed perk for loyal customers.
Stores rent their customers. You pay for every new visitor through ads, and when you stop, nothing comes in anymore. A customer who comes back without you paying again is therefore worth the most. That's what customer retention is about. A loyalty program can help with that. But usually not as the first step.
Does a loyalty program work for a small online store?
Sometimes, but only if you already know how many customers come back on their own.
A loyalty program is a fixed way to reward returning customers, for example with points, discounts or perks. It works like an amplifier. If a reasonable share of your customers already comes back, you push that up. If almost nobody comes back, there's nothing to amplify.
With the founders I talk to, I often see the same thing. A points app is running, but nobody knows how many customers order a second time. Then that's where I start. Measure first, then make the second order easy, and only after that a program.
How much is a loyal customer worth?
What a customer brings in on average over the whole time they buy from you is called the customer lifetime value.
A simple way to estimate it:
- Take your average order value.
- Multiply by the average number of orders per customer per year.
- Multiply by the number of years a customer stays on average.
- Subtract your cost of goods, shipping and payment costs. Then you have the value in margin, and that's what counts.
A worked example, as a rough estimate with assumptions. A customer orders €45 on average, does that three times a year and stays two years. That's €270 in revenue. If your margin after costs is half, that customer is worth about €135. That number tells you how much you can spend to win and keep a customer.
The most important number here is the number of orders per customer. That's where you have the most influence. In Shopify Analytics you see how many of your customers are returning. How to calculate customer lifetime value exactly is in customer lifetime value for your online store.
How do you get customers to come back after a sample?
Send your message when the sample is almost finished, with an explanation and a reason to order the full size now.
In skincare, samples are a strong entry point. But the step from sample to full size often goes wrong, because nothing happens after delivery. This is how I set it up:
- Know how long your sample lasts. If it's a week, plan the first message a few days after delivery.
- Explain first, sell after. How do you use it, where in your routine, what should you watch out for.
- Credit the sample. For example a code worth the value of the sample on the full size. Then the step doesn't feel like paying again.
- Ask for a review. A short question, at a moment when the customer has really used the product.
- Send one last reminder. Then you stop. Anyone who doesn't respond gets a regular newsletter later.
Points, subscription or discount: what works?
That depends on how often your product runs out.
| Type | How it works | Fits |
|---|---|---|
| Reward points | Points per order, redeemable for a discount | Stores with many different products and frequent small orders |
| Subscription | Regular delivery, often with a perk | Products that run out: coffee, matcha, supplements, refills |
| Discount on the second order | One-time code after the first purchase | Every store, as a first step |
| Fixed perk without discount | Free shipping, early access, a sample with every order | Brands that don't want to erode their price |
For coffee and matcha, a subscription is the obvious choice. The product runs out, and the customer doesn't want to reorder every time. More on that on the page for coffee, tea and matcha stores.
For skincare, a fixed perk often works better than a points program. Think of free shipping for loyal customers, or a new sample with every order. That also fits well with your free shipping threshold.
Discounts work fast, but don't get customers used to them. Anyone who always gets a code waits for the next one.
Work out what a little more conversion is worth to your store.
Open the revenue calculator →Which emails belong with a loyalty program?
A program without emails is a button nobody finds.
At minimum these messages belong with it:
- Thank you after the first order, with an explanation of the product and what the customer can expect.
- The sample or refill message, at the moment the product is almost finished.
- A review request, once the customer has used the product.
- A winback email for customers who haven't ordered in a while.
- A program update, for example how many points someone has.
How to set up those flows in Klaviyo is on the page about Klaviyo email flows.
How do you measure repeat purchases?
Look at three things: the share of returning customers, the time to the second order and the revenue from loyal customers.
- Share of returning customers. How many of your customers ordered more than once? You see this in Shopify Analytics.
- Time to the second order. After how many days does a customer come back on average? Time your emails to match.
- Revenue from loyal customers. What share of your revenue comes from people who bought before?
Measure this per month and compare with the same month a year earlier. How to set that up with GA4 and Shopify is in tracking and measuring. In my dashboard these numbers sit next to your traffic from Google, so you see whether new and loyal customers are growing.
Where do you start this week?
Start small, and measure before you build anything.
- Look up in Shopify Analytics how many of your customers ordered a second time.
- Check how many days there are on average between the first and second order.
- Turn on one email that goes out just before that moment.
Only once that email works is it time for a program with points or a subscription.
Do you include samples? Then read how to turn a sample into a full-size order.
