- Subscription Prescription
- Posts
- Dose 223: The Only 3 Things That Make Subscriptions Work
Dose 223: The Only 3 Things That Make Subscriptions Work
Product, offer, and usage. Everything else is noise.
This week’s dose is also a podcast episode. Tune in on your favorite platform:
Quick Note on ESPs
Here's how I'm thinking about ecommerce tech stacks right now, especially email. The brands pulling ahead are getting better ROI from more affordable systems, testing faster with AI, and working with software partners that actually show up.
Our partner Omnisend checks all three: up to 35% less than Klaviyo, AI built in at no extra cost (you can manage the whole platform from Claude or ChatGPT), and real humans answering support in 4 minutes. They'll even migrate you free in 5 days.
Get an additional 30% off when you signup with code SUBDOC30 at signup.
Matt here with your weekly Subscription Prescription 💊
This week I want to go back to fundamentals.
I'm on three or four discovery calls with brands every week, plus client work and the questions that come in online. And the same thing keeps nagging at me: brands are chasing CAC, benchmarks, whatever metric showed up in their feed that morning.
So I've been trying to boil everything I've seen work over the years down to core principles. I came out with three.
There are only three things that make subscriptions work: your product, your offer, and how people use your product. That's it. Everything else is noise. I’ll die on that hill.
This issue is the strategy. Next week I'll get into the tactics for moving each lever. But the foundation comes first, because most of the problems I audit trace back to one of these three.
1. Product: does it fit a subscription, and does it actually work?
Two questions here, and most brands only ask the first one.
The first is fit. The cleanest subscription products are things people consume every day. Coffee. Protein powder. Pet food. A nightly sleep supplement. Daily consumption means predictable reorders, and predictable reorders are the whole game.
Quarterly air filters or toothbrush heads can work too, but the longer the gap, the more likely someone just reorders when they need it.
Then there's consumption control. A jar of 30 multivitamins, one per day, is easy to prescribe. Toothpaste and shampoo are harder. You and I might both brush twice a day, but you might use twice the paste I do. That variance makes cadence messy.
And don't force consumption that isn't real. If your model assumes two scoops of protein a day so everyone buys a 60-serving bag monthly, you'll find a niche that does that. But as you scale past that niche, you hit people taking one scoop or less, and your numbers fall apart.
The second question is quality, and it's the one that matters most.
I can look at the retention cohorts of a brand we're auditing and tell you how good their product is. No email sequence or free gift covers for it. If the product fails to deliver on the promise, subscriptions won't work.
We've worked with brands that do nothing on retention and hold strong numbers anyway, purely on product quality. We've also worked with brands where we threw every tactic at the back end, saw real improvement, and still couldn't reach what a better product would get on its own.
Here's my own example. I'm bald and shave my head regularly, and I was on a razor subscription for years. As the blade quality slipped, I noticed more cuts and started avoiding the shave without knowing why. A friend handed me a different brand. Now I'm on that subscription instead.
People keep coming back to products that work. That's the whole story.
Takeaway: Be product obsessed. Check whether your consumption cadence is real, not assumed, and keep testing formats, flavors, and ingredients. If your retention cohorts look weak, look at the product before you look at the flows.
2. Offer: make the subscription the obvious choice
An offer is whatever you give someone to get them to buy. It covers price, but also the bundle, the gift, the whole package.
Most brands stop at the product page default: subscribe and save 10%. You'll get some subscribers that way, but it won't scale. Your CAC stays high and the subscription never feels like a no-brainer.
Here's the math. Say you sell a $50 bag of protein. You discount 20% for subscribers, so $40. Landed cost is $10, so you have $30 of margin. But CAC is $60. You're $30 underwater on the first order.
Now add a free gift, like a second bag. Landed cost goes to $20, but that extra value might drop CAC by $15 to $20. Now you're roughly $10 underwater instead of $30.
A few structures work consistently.
Starter kits. Dollar Shave Club has run one for years: a handle, a set of blades, some shaving cream, "a $25 value" for a few dollars plus shipping. Everyday Dose does this on their main PDP right now, and they frame the starter kit as $140 in gifts. The point is packing in enough value that someone actually starts using the product.
Bundling. Buy two bags, get the third free. Or pair a core product with a related one at a steep subscriber discount, like creatine at 50% off with your protein.
Gifts as incentives. ButcherBox has done free wings for life, free turkeys around Thanksgiving, free steak cuts. HelloFresh adds free breakfasts or desserts. These lower acquisition cost, and when you extend gifts to later orders, they become a retention play too.
Takeaway: Build an offer framework and test it. Starter kits, bundles, gifts, merch, whatever makes the subscription feel obvious. A percentage off is a starting point, not a strategy.
3. Usage: nobody renews a product they never used
I feel a little silly saying this on calls, because it sounds obvious. But if someone isn't using your product, they're not renewing. You could sell the miracle cure of the century, and if nobody opens the jar, you don't have a subscription program.
I've run audits on brands large and small where I ask simple questions. Do new subscribers know how this should be mixed? Do they know it might make them a little jittery at first? Do they know it tastes different from what they're used to? More often than you'd think, the answer is "I'm not sure."
That first product experience is everything. If you sell mushroom coffee, which tastes and behaves differently from regular coffee, guide people through a perfect first cup. FAQs, how-tos, and a best-practices email triggered off the delivery event in Shopify, not a date on a calendar.
Brands are often too scared to engage here. Don't be. Those first 10 to 15 days are your ripest window to get someone activated. An email on day 25 is late.
Takeaway: Map your first two weeks around adoption. Trigger education off delivery, answer the questions nobody thinks to ask, and treat first use as a retention metric.
Bottom line: Product, offer, usage. Everything else is a detail. If retention is weak, check the product. If CAC is high, check the offer. If people cancel after order one, check whether they ever used what they bought. Get obsessed with those three and the rest unlocks.
Until next Tuesday, that's your Subscription Prescription. 💊
- Matt Holman 🩺