Most subscription products don't die in Year 1. They die in Year 2.
Year 1 is about acquisition. The numbers are climbing. Everyone's celebrating.
Year 2 is when you realize that getting people in the door and keeping them are completely different problems.
Your early adopters start churning. Acquisition costs climb because you've burned through the easy audiences. And you start to see two distinct problems that you never had to think about when growth was covering for everything.
The first is involuntary churn. These are members who didn't choose to leave. Their credit card expired. A payment failed silently. A billing update didn't go through. They wanted to stay, but your system let them fall out the back door. This is pure waste, and it's often the first thing worth fixing because the ROI is immediate. Smart retry logic, dunning sequences, card updaters. None of it is exciting. All of it matters.
The second problem is harder. It's the members who technically stayed but quietly disengaged. They signed up, tried one or two things, and never went deeper. They're still paying for now, but they're one price increase or one competitor away from canceling.
Solving both requires the same shift in mindset: stop treating the member base as one group.
In my experience, the companies that survive Year 2 start segmenting early. Different members use the product differently. Different members need different onboarding. Different members find value in different parts of the benefits ecosystem.
When you treat a power user and a casual browser the same way, you lose both. The power user feels under-served. The casual browser never gets deep enough to stick.
What actually works in practice:
🔸 Segmented onboarding: Instead of one generic welcome flow, build paths that match why someone joined. A member who signed up for one specific benefit needs a different first week than someone exploring the full ecosystem.
🔸 Engagement tied to the benefits ecosystem: The goal isn't just "get them to log in more." It's getting members to discover and actually use more of what's available. The more benefits a member actively uses, the harder it is for them to leave.
🔸 Involuntary churn recovery: Failed payments are not a billing problem. They're a retention problem. The best teams treat payment recovery as a product surface, not an afterthought.
🔸 Personalized re-engagement: When a member goes quiet, the worst thing you can do is send a generic "we miss you" email. The best retention teams know which benefit that member used most and lead with that.
Spotify is a good example of getting this right. Discover Weekly isn't just a playlist. It's a personalized reason to come back every Monday. Wrapped turns listening data into something people actually want to share. These aren't generic features. They're experiences built on understanding individual behavior.
Strava figured out that their most retained users weren't the fastest runners. They were the ones connected to other people. So they invested in segments, leaderboards, and clubs. The product experience changed based on how you used it.
The New York Times went from one subscription to a portfolio of daily habits: Cooking, Games, Wirecutter, The Athletic. They tested which combinations drove the highest retention and built around those patterns.
The real work of Year 2 is experimentation. Test, learn what drives habit formation, and double down on what works.
None of this is glamorous. It's cohort analysis. It's A/B testing onboarding flows. It's fixing payment retry logic. It's tracking which benefit combinations correlate with 12-month retention. It's figuring out what makes one member stick and another leave, and then building around those signals.
The foundational work is what compounds.