For a consumable product, loyalty is repeat purchase behavior, and the campaigns that produce it are the ones that get a customer to finish their first bottle. Points, tiers, and referral rewards operate on customers who already came back. They’re a multiplier on retention you’ve earned, not a source of it.
That distinction decides where the money goes. A brand with a 12% ninety-day repeat rate installing a points program is putting a rewards layer on top of a product experience that isn’t working, then wondering why redemption is low.
I’ve watched this play out at brands from $1M to $60M+. The retention work that moved the number was almost never the loyalty platform.
Why supplements are different
Most loyalty thinking comes from retail and fashion, where purchases are discretionary, occasions are irregular, and the job is to be top of mind when the customer decides to buy something.
Consumables invert that. Your customer will need more of your product on a predictable schedule whether or not they think about you. The question is only whether they reorder from you, switch, or stop taking it entirely.
Stopping is the big one, and it gets ignored. In supplements, the largest source of churn is people who quit the habit, not people who chose a competitor. That means your biggest retention lever is whether the customer got through the first thirty days and felt something worth continuing.
Nothing in a points program addresses that.
The retention hierarchy
Work in this order. Each level compounds and the ones underneath make the ones above worth building.
1. The product does what you said within the window the customer expected. Not a marketing problem, but it caps everything above it. If your 90-day repeat rate is under 15% in a consumable category, no campaign fixes that. Go talk to the people who didn’t reorder.
2. The customer uses it correctly and consistently. Massive, underinvested, and cheap to improve.
3. Reordering is easier than deciding. Subscription mechanics, timing, and friction.
4. There’s a reason to consolidate spend with you. Bundles, cross-sell into the routine, category expansion.
5. There’s a reason to advocate. Referral, community, status.
Most brands start at level five because it’s the most visible and has the most vendors selling it. The returns are at levels two and three.
The campaigns that actually work
First 30 days: the usage sequence
The highest-return retention campaign in supplements is a well-built onboarding flow, and most brands run a version so generic it may as well not exist.
What it needs to do: set the timeline honestly, tell people exactly how and when to take the product, name the thing they’ll notice first, and check in before the bottle runs out.
The specifics that matter:
- Set expectations against a real timeline. If the effect shows up at week three, say week three. Customers who expect week one quit in week two.
- Send a usage reminder around day 5 and day 12. Non-compliance is invisible churn. People stop taking it, feel nothing, and don’t reorder. They’ll tell you the product didn’t work.
- Ask a question at day 21 that requires a real answer. “What have you noticed?” outperforms “How are you liking it?” and doubles as product research.
- Trigger the reorder prompt at 70-75% of the bottle’s supply, not on a calendar month. A 60-count bottle taken twice daily is a 30-day supply, and a prompt on day 30 arrives after they’ve run out.
I’ve seen this sequence alone move 90-day repeat rate by 5 to 10 points. It costs nothing but the writing.
Ongoing: the replenishment engine
Once someone reorders, your job is removing every reason to think about the decision again.
- Subscribe and save priced honestly. 15-20% is the working range. Under 10% and the discount doesn’t overcome the commitment. Over 25% and you’ve trained your buyers to never pay full price while giving away margin on your best customers.
- Skip and delay controls that are obvious. Brands hide these to protect churn numbers and it backfires. A customer who can’t skip a shipment cancels the subscription instead. Make skipping one click.
- Cadence that matches actual consumption. If people take two capsules a day and you ship a 30-day supply every 30 days, they’ll stack up and cancel. Ask about their usage and set the interval to match.
- Pre-shipment notice with a real window. Three days out, one click to change anything. This converts cancellations into delays, and delays into retained customers.
Whether your engine is working is a question with a numeric answer. The month-by-month ranges I use are in subscription retention benchmarks, and the month-one-to-month-two drop is where most supplement brands lose the argument.
Quarter two onward: consolidation
Now loyalty mechanics start earning their keep, because you have customers with enough history to reward.
- Routine bundles rather than category bundles. Group products by when they’re taken, not by what shelf they sit on. Morning stack and evening stack outperform “immunity collection.”
- Tiered benefits with non-discount value. Early access to new products, free shipping thresholds that drop, a direct line to a practitioner or the founder. Discounts train price sensitivity, while access builds the relationship.
- Win-back at 45 and 90 days post-lapse, with different messages. At 45 days assume they forgot. At 90 assume something went wrong and ask. The second one gets replies that tell you what to fix. The full build is in the win-back sequence that recovers lapsed subscribers, including the failed-payment fix that usually recovers more revenue than the emails do.
The referral program, in its right place
Referral is a level-five campaign and it works when levels one through three are solid.
Two-sided incentives outperform one-sided. Ask at the moment of demonstrated satisfaction, which for supplements is roughly day 45 to 60, not at purchase. A customer who has been taking your product for two months and reordered is the only person whose recommendation carries weight.
When points programs do make sense
They earn their place in a specific set of conditions:
- Multi-SKU catalogs where customers can meaningfully consolidate spend
- Repeat rate already above 30% at 90 days
- Enough purchase frequency for balances to accumulate to something desirable
- A brand with community identity, where status is worth something
Under those conditions, points raise the switching cost for customers already inclined to stay. Outside them, you’ve built a discount liability on your balance sheet and a dashboard nobody looks at.
If you install one, tie earning to behaviors you want rather than dollars spent. Points for reviews, for referrals, for staying subscribed through six shipments, for completing a usage check-in. Points per dollar is just a delayed discount.
What to build at your specific revenue is a longer answer. Loyalty program strategy by stage covers the build order at $1M, $5M, and $20M, including how to price point liability so the program doesn’t quietly eat your margin.
How to measure whether any of it worked
Loyalty platforms report enrollment, points issued, and redemption. None of that tells you whether retention improved.
Measure this instead:
- Cohort repeat rate at 30, 60, 90 days, before and after each campaign launch, by first-purchase month
- Subscription retention at months 1, 3, and 6 for each acquisition cohort
- Contribution margin per customer at 12 months, by cohort, which catches discount programs that lift repeat rate while destroying margin
- Reactivation rate from win-back flows, measured against the lapsed population rather than the send
The trap in loyalty measurement is comparing enrolled customers to non-enrolled. Enrolled customers were already your best customers, so the program looks fantastic. Compare cohorts across time instead. Full definitions are in the nine metrics that predict scale.
What I’d build first at $3M
If you’re at $3M with a repeat rate under 25% and want the fastest return:
- Rewrite the first-30-days sequence. One week of work. Biggest single lever available.
- Fix your reorder timing to match actual supply duration rather than the calendar.
- Make skip and delay obvious in the subscriber portal.
- Build the cohort repeat table so you can see whether any of it worked.
- Call twenty customers who didn’t reorder. Nothing on this list will teach you more.
Then look at points, tiers, and referral. In that order, once there’s retention worth multiplying.
Retention is where a wellness brand’s enterprise value actually lives. A brand with 40% ninety-day repeat and healthy subscription retention is worth substantially more than a brand with the same revenue and half the retention, because one of them owns its customers and the other rents them monthly from an ad platform. Ancestral Supplements grew from $28M to $60M+ on a business built to keep customers, not just to acquire them.
Common questions
What’s a good repeat purchase rate for supplements? 20-30% at 90 days is workable, above 35% is strong. Under 15% points to a product, expectation, or usage-compliance problem rather than a marketing one.
Should a supplement brand run a points program? Only above roughly 30% ninety-day repeat rate, with a multi-SKU catalog. Below that, the same effort spent on onboarding and reorder timing returns considerably more.
What discount should subscribe and save offer? 15-20%. Below 10% it doesn’t overcome the commitment. Above 25% you’re giving away margin on customers who would have stayed anyway.
How do I reduce subscription churn? Match shipping cadence to actual consumption, make skip and delay one click, send a pre-shipment notice three days out, and fix month-one churn with a better onboarding sequence. Month 1 churn above 25% is usually an over-sold checkout offer.
When should I ask for a referral? Day 45 to 60, after a customer has used the product long enough to have an opinion and ideally after they’ve reordered. Asking at purchase gets you volume with no conversion.
If your acquisition costs keep climbing and retention isn’t carrying its share, that’s a solvable problem and usually a fast one. Here’s how engagements work, and the button below books thirty minutes.