Subscription & Retention Experiences
Anyone Can Launch Subscriptions. We Make Them Profitable.
63x subscriber count. 1,400% subscription growth. Churn down 32%.
Subscription program design, account portals, offer architecture, and churn reduction — plus contract migrations that don’t lose subscribers. The tooling is the easy part. What decides whether the program is profitable is how much control you give customers and how clearly you explain the value of your offer.
Subscription programs are easy to launch and hard to keep profitable
Installing a subscription app takes an afternoon. What follows is the actual work: an offer people understand, a portal that lets them stay on their own terms, and a churn rate that doesn’t quietly consume your acquisition spend.
Where programs lose money
Churn matching acquisition
When AG1 came to us, new subscribers were signing up every day and the same number or more were churning at the same pace — growth that costs money without accumulating anything.
Auto-enrollment used as a growth strategy
Bare Performance Nutrition grew subscriptions through education and visibility instead of auto-enrollment: subscribers grew roughly 90% in eight to nine months, compounding to 1,400% growth across four years.
A portal that can’t do what subscribers need
If someone can’t change their delivery date, skip a shipment, or update a preference without emailing you, their only lever is cancellation. AG1’s portal removes every reason to contact them or to leave.
Offers nobody can parse
Subscribe-and-save buried in a page, unclear frequency options, prepay terms that require arithmetic. BPN’s existing offer wasn’t the problem; finding and understanding it was.
Retention treated as a lifecycle email problem
Discount-led win-back campaigns aimed at people already leaving. That’s the last and least effective place to intervene.
Subscription data as the riskiest thing in a migration
Good Ranchers’ cutover ran in low-traffic hours with monitoring and support trained before launch, so post-migration friction never reached a subscriber. → System Replatform
Experience
18+
Years of Unlocking Growth
Scale
$3.14B+
in GMV Migrated to Shopify
Partnership
1 of 5
Founding NA Shopify Platinum Partners
Precision
94.7%
Implementation Predictability
Clients
8
Unicorns & Counting
A subscription platform we built accommodates hundreds of thousands of orders every month.
Five things we’ve concluded about subscription and retention
Seventeen years across subscription businesses at every scale.
Control reduces churn
People who can’t manage a subscription cancel it, because cancellation is the only control they have. Good Ranchers’ program was redesigned around subscriber control, and churn fell 32%.
Retention is designed at acquisition
The expectations set during the buying experience determine whether someone is still a subscriber in month three. → eCommerce Funnel Engineering
Non-transactional value is what keeps customers subscribed
Portals built only to process transactions have nothing to bring anyone back to. A consultation, educational content, or community gives customers a reason to stay engaged between orders.
Buy the platform, build the experience
Ordergroove, Recharge, StayAI, and their peers handle billing, contracts, and dunning better than a custom build will. AG1’s custom subscription platform is the exception that proves the rule. → Technology Stack
Frequency and prepay framing move more than discounts do
How you position a prepay option against month-to-month changes the mix more than another five percent off. → CRO & Experimentation
The work
Program design and offer architecture
Frequency options, prepay positioning, box construction, and how the offer is presented and explained.
Platform selection and implementation
Ordergroove, Recharge, StayAI, and their peers — selected against your model and integrated into the rest of the stack. → Integrations
Account portal design and build
Delivery management, skip and pause, preference centre, order history and reorder, self-service tools. → Visual Design
Loyalty and membership integration
Points, redemption, referrals, milestone recognition, and member pricing — integrated so subscribing reads as joining something.
Churn diagnosis and reduction
Where subscribers leave, why, and which of those causes is addressable through program design rather than discounting.
Contract migration
Moving subscription contracts, billing state, and payment tokens without lapsed billing. → System Replatform
What a subscription engagement commits to
| Commitment | Acceptance criterion |
|---|---|
| No dark patterns | Cancellation and management are as easy as subscribing. If you want otherwise, we’re not the right partner |
| Migration integrity | Zero lapsed billing events, verified — not estimated |
| Support readiness | Your support team trained on the new system before cutover, not after |
| Self-service | An agreed list of what a subscriber can do without contacting you |
| Churn attribution | Causes identified and separated into addressable and not, before any discount is proposed |
| Platform rationale | Documented reasoning for the platform chosen, including why not the alternatives |
| Downstream measurement | Retention and LTV tracked alongside subscriber count |
Retention built on friction shows up in reviews, chargebacks, and eventually in a regulator’s attention.
How subscription work runs
Diagnosis
Weeks 1–3
Churn analysis by cohort and cause, portal audit, offer comprehension review, platform assessment, LTV baseline.
Program design
Weeks 2–5
Offer architecture, frequency and prepay structure, portal design, loyalty integration model.
Build
Weeks 4–12
Platform implementation or migration, portal build, integration into the wider stack.
Optimize
Ongoing
Offer and frequency testing, churn intervention, portal iteration.
Timelines: Portal and offer work on an existing platform lands in 8 weeks end to end. A platform migration with complex box or bundle logic runs 12 to 16.
Where subscribers leave
The second delivery
The point where the trial ends and the actual commitment starts, which is why retention is an acquisition-side problem too.
The moment they need to change something
A delivery date that doesn’t work, a frequency that’s wrong, a pause they need for a month. If the portal can’t handle the request, cancellation will.
Frequency mismatch
People subscribed to a cadence that doesn’t match consumption end up with too much product and cancel.
The offer they didn’t understand
Subscribers who never grasped the terms churn at the first surprise.
Involuntary churn
Expired cards, failed payments, and address changes — usually the most recoverable share, because nobody in that group decided to leave.
Nothing to come back for
A portal that only processes orders gives a subscriber no reason to engage between deliveries.
Anatta’s Agentic Operating System
Every subscription engagement runs on one system.
Senior architects paired with an agentic framework that absorbs the commodity execution — portal components, integration scaffolding, migration verification — so senior time goes to program design and the migration protocol, where the consequences of being wrong are measured in lost contracts.
See how we workSubscription work we’ve run
Bare Performance Nutrition
Subscribers up ~90% in under a year, and 1,400% growth across four — with no auto-enrollment
Explore the Bare Performance Nutrition success storyAG1
A custom subscription platform that 63x-ed subscriber count and handles hundreds of thousands of orders a month
Explore the AG1 success storyGood Ranchers
Churn down 32% after redesigning the program around subscriber control
Explore the Good Ranchers success storyTrade Coffee
Hundreds of thousands of subscribers migrated with an AI taste-profile engine rebuilt intact
Explore the Trade Coffee success storyGrove Collaborative
A decade of proprietary subscription logic moved to Shopify and Ordergroove, development time cut 200%
Explore the Grove Collaborative success storyMack Weldon
Recurring revenue up 44% in the first year of the partnership
Explore the Mack Weldon success storyWe went from focusing on getting new customers to focusing on keeping them, and Anatta helped us build a more holistic retention roadmap.
Markus KarjalainenFour Sigmatic
Anatta addressed all of our challenges in a practical way.
Chris ClarkCo-Founder & CDO, Grove Collaborative
How to diagnose your churn
Many brands reading this know their churn rate and not its causes. The distinction is the whole job.
- Split voluntary from involuntary churn. Usually the most recoverable share.
- Find your cliff. Churn by delivery number, not by month.
- Read your cancellation reasons properly. “Too much product” is a frequency problem, not a price problem.
- Count what a subscriber can’t do alone. Every action requiring a support ticket is a cancellation risk.
- Test whether people understand the offer. Ask five customers to explain your terms back to you.
- Check what acquisition promised. Gaps show up as churn at delivery two.
- Ask what brings someone back between orders. If the answer is nothing, that’s the retention gap.
- Model LTV against your discount. If retention depends on a standing discount, you’ve bought it rather than earned it.
The Churn Diagnostic Framework
The worksheet version, with cohort and delivery-number analysis structures included. Name and email.
Frequently asked questions
Should we auto-enroll customers into subscriptions?
We’d argue against it. It produces subscribers who didn’t decide to subscribe, which arrives later as churn, chargebacks, and support volume. Bare Performance Nutrition grew subscriptions through visibility and education instead.
Which subscription platform should we use?
It depends on your model rather than a ranking. We’ve implemented Ordergroove for Grove Collaborative, Trade Coffee, and BPN, and Recharge for Good Ranchers.
Should we build a custom subscription platform?
Almost certainly not. Billing, contracts, and dunning are handled better by platforms maintained by someone else. AG1’s custom platform is the exception, justified by its scale and single-product model.
How do we reduce churn?
Start by splitting it into causes. In our experience the largest addressable cause is control — people who can’t manage their subscription cancel it.
Can we migrate subscriptions without losing subscribers?
Yes, with a protocol: cut over in low-traffic hours, monitor in real time, train support before launch, and verify billing continuity contract by contract.
What belongs in the customer portal?
More than most brands build. Treat it as your customer service desk. The test is whether a subscriber can handle their own account without contacting you.
How do we grow subscriptions without discounting deeper?
By making the offer easier to understand and the commitment easier to control, then testing framing rather than price.








