A Shopify store needs a loyalty program once its repeat purchase rate passes roughly 20%, it ships at least 300 orders a month, and reward costs fit inside 3% to 5% of gross margin per order. A referral program makes sense earlier if customers already recommend you and the product is easy to gift. Below those numbers, fix product, email, and post-purchase experience first.
Loyalty apps are easy to install and hard to judge. The dashboard shows points earned and members joined, which feels like progress. What it rarely shows is whether those members would have bought again anyway.
We have set up loyalty and referral programs on our own demo stores and helped merchants remove a few that were costing more than they returned. The difference between the two groups came down to three numbers, all of which you can pull from Shopify analytics in ten minutes.
This guide walks through that math, a 60-day test you can run before paying for an annual plan, and how three common apps compare.
- When repeat rate, order volume, and margin say yes or no
- Loyalty points vs referral rewards: which to try first
- Cost math on a real-looking $60 AOV store
- How ReferralCandy, Growave, and BON Loyalty differ
Which three numbers tell you if a program will pay?
Open Shopify Analytics and look at the last 90 days. Then answer three questions.
What share of customers ordered more than once? Shopify reports this as returning customer rate, and its customer reports documentation explains how first-time and returning customers are counted. Under 15% usually means the problem is upstream: product fit, shipping speed, or no follow-up email. A points program will not fix a product people do not want twice. Over 20% means people already come back, and a reward can nudge them to come back sooner or spend a little more.
How many orders a month? Below about 100, rewards are too thin for anyone to notice. A customer who earns 60 points a quarter does not feel loyal, they feel ignored. Above 300, you have enough volume to measure a change within two months.
What is your gross margin per order? If a $60 order leaves $24 after product and shipping, a 5% reward costs $3. That is fine. If margin is $9, the same reward eats a third of your profit.
Research summarized by Harvard Business Review on the value of keeping the right customers is often quoted for the idea that small retention gains raise profit a lot. That holds only when retained customers buy at healthy margins. A loyalty program that trains people to wait for rewards can do the opposite.
Should you start with points or with referrals?
They solve different problems.
Points and VIP tiers aim at existing customers. They work best for consumables and categories people restock: coffee, skincare, pet food, supplements. The goal is to shorten time between orders.
Referral rewards aim at new customers, using existing ones as the channel. They work best when the product is visible or giftable, when customers are already happy (check your review scores), and when the referral reward is meaningful compared to price. A $10 give, $10 get offer on a $25 product is generous. On a $300 product it barely registers, and a percentage or free accessory may land better.
Our rule of thumb: high repeat rate and low advocacy, start with points. High advocacy and low repeat (furniture, mattresses, durable goods), start with referrals. Both high, run both, but launch one at a time so you can measure each.
What does the math look like on a typical store?
Here is a worked example. The store does 500 orders a month at $60 AOV with 40% gross margin, so $24 margin per order. Baseline returning customer rate is 18%.
Scenario one, points at 5% back. Suppose repeat rate climbs to 22% over two months. That adds roughly 20 extra orders a month, about $480 in extra margin. Reward cost on all redeemed points runs around $150 at a typical 40% redemption rate. Net gain about $330, before the app fee.
Scenario two, referral at $10 give, $10 get. If 12 new customers arrive through referrals monthly, that is $288 in first-order margin. Subtract $120 in rewards, and you net about $168. These customers also have a chance to come back, which the first month does not show.
Scenario three, VIP tiers on top of points. Our client data suggests VIP tiers lift repeat rate more than flat points, because top spenders see a status to protect. Say 24% repeat rate, about 30 extra orders, roughly $470 net.
Now subtract the app. A $49 a month plan on scenario two leaves $119. That is a thin win. On scenario one it leaves $281, which is comfortable.
These numbers are illustrations, not promises. Your actual lift depends on product, email, and how visible you make the program. The point is to run the math before you install anything.
How do ReferralCandy, Growave, and BON Loyalty compare?
We checked entry plans in October 2026. Always confirm current pricing on each app's Shopify listing.
ReferralCandy is referral only. It does one job and does it well: post-purchase referral emails, a share page, fraud checks, and reward payouts. Paid plans start around $59 a month, and some plans add a commission on referred sales. Good fit for stores with strong word of mouth that do not want points.
Growave bundles loyalty, referrals, reviews, and wishlists in one app. There is a free tier with limits and paid tiers above it. The appeal is one dashboard and one set of scripts instead of three apps. The catch: you load more features, so check speed and only enable modules you use. We compared Growave's review side in our Loox vs Stamped vs Growave test.
BON Loyalty focuses on points, VIP tiers, and referrals at a lower price. It has a free plan and paid plans from around $49 a month. The interface is simple, which helps small teams. Fewer extras than Growave, but fewer scripts too.
Short version: referral only, pick ReferralCandy. Loyalty plus reviews in one place, try Growave. Points and VIP on a budget, try BON.
What should the reward itself look like?
Small details in reward design change redemption more than the app you pick.
Fixed-amount discounts ($5 off) are easier to understand than percentages, and shoppers redeem them more often in our tests. Free shipping rewards work well when your shipping fee is a known pain point. Free products work for consumables with low unit cost, such as a sample size.
Keep the first reward reachable. If a shopper needs four orders before earning anything, most never see it. Aim for a first reward after one or two orders at your AOV. Then make later tiers harder.
Set points to expire, but give long windows, such as 12 months. Expiry emails ("Your 400 points expire in 14 days") are among the best-performing messages in any loyalty program we have run, because they create a deadline the shopper set up themselves.
How do you run a 60-day test before committing?
Never buy an annual plan first. Run a test that tells you if the program earns its keep.
Week 0: export baseline. Pull 90 days of returning customer rate, AOV, and average days between first and second order. Write them down.
Week 1: launch small. One earn rule (points per dollar spent), one reward (a fixed discount), one launcher. Turn off signup bonuses and birthday rewards for now. You want to measure one thing.
Weeks 2 to 8: promote it. Add a line to your order confirmation email, a section on the account page, and a mention in one flow in your email tool. Our Klaviyo vs Omnisend breakdown covers setting up those post-purchase flows. A program no one knows about cannot change behavior.
Week 9: decide. Compare repeat rate and days-to-second-order against baseline. Calculate net margin gain minus reward cost minus app fee. Keep it if the net gain beats the app fee by at least two times. Otherwise uninstall, and remove leftover scripts and theme snippets so the launcher does not keep loading.
Where should the loyalty launcher sit on mobile?
Placement gets ignored and causes real problems.
Most loyalty apps add a floating button in a bottom corner. On phones, that corner is crowded: chat widgets, sticky add-to-cart bars, cookie banners, and back-to-top buttons all want it. We have seen launchers cover the checkout button on product pages, which is the worst possible spot.
Test at 390px width on every key template. Move the launcher to the opposite corner from chat, or hide it on product pages and show it only on the home page and account page. Google's Cumulative Layout Shift guidance is worth reading here too, because launchers that pop in late can shift content as the shopper is about to tap.
When should you not run a program at all?
Some stores should skip loyalty for now.
If you sell a one-time purchase with no natural repeat (wedding dresses, a single piece of equipment), points do little. A referral program might still work, but loyalty will not.
If your margin is already thin from heavy discounting, rewards stack on top of discounts and erode profit. Fix pricing first.
If your repeat rate is low because of slow shipping or poor reviews, rewards paper over the cause. Look at post-purchase experience. For subscription products, a subscription app often does more than points. We covered that in our Appstle subscriptions pricing math piece.
Questions merchants ask us about this
What is a good repeat purchase rate for a Shopify store?
It varies by category. Consumables often see 25% to 40%, while apparel and gifts commonly sit between 15% and 25%. Compare against your own baseline more than against averages.
Do loyalty programs slow down my Shopify store?
They add scripts. A single loyalty app typically adds a small amount of JavaScript. Stacking loyalty, reviews, and wishlist apps from different vendors adds more, so check speed after each install.
Is a referral program better than a loyalty program?
Neither is better in general. Referrals bring new customers through existing ones, loyalty brings existing customers back sooner. Pick based on whether your growth problem is acquisition or retention.
How long before a loyalty program shows results?
Expect six to eight weeks for a measurable change in repeat rate, longer if your average time between orders is more than a month.