Pre-PMF growth: the free version
If nobody comes back to your product yet, you should not hire me. Here is everything I would do in your position, at no cost.
the short answer: if your product does not yet have people returning on their own, growth spending buys you an expensive version of the same problem. this is the segment i turn away, so here is the whole answer for free. none of it requires a budget or a hire.
First, establish whether you are actually pre-PMF
not by intuition. by one number.
take a cohort of users who arrived in a given month, without paid acquisition. what share are still doing the core action a month later? if that is low and flat, you are pre-PMF regardless of how the topline looks.
for a subscription product the sharper version is the first renewal. among 3,519 consumer AI apps covering 50 million paid subscriptions, the apps keeping 13.9% of subscribers at twelve months renewed 57.9% of monthly subscribers at the first renewal, while the ones keeping 1.4% renewed 30.2%. that single window separates the two groups.
Then talk to the people who left
twenty conversations, not a survey. specifically people who used it and stopped, not people who never started.
the question that works is not “why did you stop”, because people rationalise. it is “walk me through the last time you used it”. the moment they lost interest shows up in the narration, usually somewhere they expected something to happen and it did not.
this is the highest-value hour available to any pre-PMF founder and it costs nothing but discomfort.
Fix the moment before value, not the funnel
almost every pre-PMF product has a gap between signing up and getting anything.
find the specific step where people stop, then remove things until it is shorter. not redesign, remove. every field, every screen, every decision between arriving and getting something is a place people leave.
the reason this beats acquisition work is that it improves every future user, including the ones you eventually pay for.
Go where your users already are, as a person
not to promote. to learn the language.
pick three to five communities where people discuss the problem you solve. participate for weeks with nothing to sell. the exact words they use become your copy later, and the recurring objections become your positioning.
this is the same first step i take on paid engagements, and at ZuAI the reddit participation phase produced the language that the paid creative was eventually built from. it is the cheapest research available and almost nobody does it because it is slow.
the rules that keep you from getting banned doing this are in the SAFE reddit playbook.
Ship narrower
the most common pre-PMF pattern i see is a product doing four things adequately for a broad audience.
pick the one thing a specific group needs badly, and do only that for only them. retention arrives when a small group finds something indispensable, not when a large group finds something acceptable. you can widen afterwards; nobody has ever narrowed successfully after scaling.
What not to do
do not run paid acquisition. you will learn what you already know at a higher price.
do not hire a growth marketer. including me. we optimise what exists, and what exists does not yet hold people.
do not launch on Product Hunt hoping for a reset. featured status now goes to roughly 10% of launches and outside the top ten a launch produces very little. it is not a fix for retention.
do not raise on the growth story. investors at the next stage are buying a repeatable mechanism, and pre-PMF you do not have one to describe.
When to come back
when a cohort you did not pay for is still active a month later. that is the whole test.
at that point acquisition work compounds instead of leaking, and hiring someone makes sense. before that, everything above is free and it is genuinely what i would do in your position.
if you want a second opinion on which side of that line you are on, let’s talk. the call is free and “not yet” is a common and useful answer.
sources: revenuecat AI app retention study, 3,519 apps and 50M+ paid subscriptions · published Product Hunt launch statistics 2026.