Most founders don’t have a product problem. They have a detection problem.
One pattern I keep noticing in early-stage products:
Founders spend months improving the product — but almost no time improving their ability to detect what actually matters.
So they optimize what’s easy to see:
– more features
– better UI
– faster performance
– more AI
But the real signals are harder to detect.
Things like:
– when users quietly stop trusting the system
– when the product becomes optional instead of necessary
– when people adapt their workflow to avoid using it
By the time these signals show up in dashboards or metrics, the damage is often already done.
The uncomfortable part is that detection requires different behavior:
talking to users, watching real workflows, noticing hesitation, confusion, workarounds.
Not just reading dashboards.
Execution matters.
But executing on the wrong signal is just efficient waste.
I'm curious how other founders here deal with this:
What’s the earliest signal that tells you something is wrong with your product?
The 'product becomes optional instead of necessary' signal is the most dangerous because it's invisible in dashboards. Login frequency drops, session depth drops, but nobody cancels — they just ghost.
One concrete detection signal that works: payment failure response time. When a subscription payment fails and the recovery email goes out, how quickly does the customer update their card? A customer who still values the product updates within hours. Someone who's already disengaged will let it sit for days or let the subscription die.
It sounds like a billing metric, but it's actually a product health signal — it measures how much friction customers are willing to tolerate to keep access. That behavior is usually correlated with NPS and future cancellation risk.
The earliest signal I watch: does the customer move toward the product when friction is introduced, or away from it? The ones who move toward it are your real users.
That’s a really interesting signal.
I like the framing of “movement toward the product vs. away from it when friction appears.” That’s a much more behavioral way to look at product health than most dashboards.
The payment recovery example is great because it reveals something most metrics hide: how much effort a user is willing to spend to keep access.
What’s tricky is that many teams only monitor success rates (recovered payments, churn rate), but not the response time distribution, which probably carries much more information.
Your point also connects to something I’ve been thinking about recently:
early product disengagement often shows up first as increasing tolerance for friction.
When people value a product, they resolve friction quickly.
When the product becomes optional, they simply stop reacting.
Out of curiosity — have you seen other signals like this outside billing flows?
For example in onboarding or feature adoption?