The Product-Market Fit Treadmill
In AI, you must recapture product-market fit every three months as both product and market shift under you
- Difficulty
- Expert
- Time to result
- ~ongoing to results
- Steps
- 4
- Confidence
- 90%
A model for why product-market fit is no longer a milestone you achieve once and then scale. In AI, the underlying LLM capabilities and consumer expectations both step-change roughly every three months, so even a large, fast-growing company must repeatedly re-find PMF rather than settling into pure scaling. Use it to set organizational expectations and team design in AI-native companies.
Origin
Elena reframes her earlier belief that PMF evolves over years (second/third horizons taking 5-10 years). She now sees a 3-month cycle driven by two forces changing simultaneously — LLM capability leaps and unprecedented speed of consumer expectation shifts.
Core principles
- 01PMF is not a done state — it is an endless fight to keep it
- 02Two variables shift every ~3 months: LLM capability and consumer expectation
- 03You must build ahead of the model, betting so functionality is ready when the model catches up
- 04The team that finds PMF differs from the team that scales — you now need both, continuously
How to run it
- 1
Accept the 3-month cycle
Assume both your underlying technology ceiling and consumer expectations will step-change roughly quarterly, forcing a PMF re-capture.
Watch out Nobody's future is bulletproof — even a $200M company growing 10% month-over-month can lose PMF in three months.
- 2
Build ahead of the model
Don't wait for the next LLM to ship then build on it. Make bets and build functionality beforehand so it's live the moment the new model releases.
Pro tip Have the functionality ready for the model to 'catch up' to, not the reverse.
Watch out Waiting for technology to mature before building means arriving late every cycle.
- 3
Track both product and market shifts
Watch capability leaps (what the LLM can now do) and expectation leaps (what consumers now demand) as separate inputs, since both move fast.
Pro tip Consumer perception now moves faster than in the past — sometimes faster than the technology can address.
- 4
Staff a team that both finds and scales PMF
Build an organization that can repeatedly re-find PMF while also scaling, because you can no longer hand off cleanly from a discovery team to a scaling team.
Watch out You must throttle scaling efforts periodically to reinvent — short blitzes of growth, not year-long commitments.
In the wild
When Gemini 3 launched, OpenAI — with nearly a billion monthly active users — reportedly lost several percent of market share in a week, triggering an internal code red.
→ Illustrates that even the category leader must keep recapturing PMF.
Marketing positioning and messaging that used to hold for years now holds for ~3 months before the product changes underneath it, forcing constant re-work.
→ Short narrative cycles that must be delegated partly to product/engineering.
Common mistakes
Treating PMF as achieved
Assuming that once you have PMF you just scale, hire salespeople, and go up-and-to-the-right — a model Elena says no longer holds in AI.
Over-focusing on pioneers
Constantly recapturing early-adopter pioneers leaves no time to reach the latent majority (adjacent users), risking alienating the broader market.
Is it for you?
Best for
Founders and growth leaders at AI-native companies building on rapidly evolving foundation models
Not ideal for
Stable, non-AI products in slow-moving markets where PMF genuinely does persist for years
From the transcript
“every 3 months I feel like we have to recapture our product market fit and not just recapture on the same technology and with same…”
“you have to build beforehand to like make a bet and then it's the LLM to catch up because when that model releases you already…”
“the team that finds your product market fit is very different than the team that usually scales your company. Yet we have to find the…”
From the episode
The new AI growth playbook for 2026: How Lovable hit $200M ARR in one year
Elena Verna (Head of Growth)