The 95/5 Growth Inversion
In fast-moving AI markets, spend 95% inventing new growth loops and 5% optimizing existing ones
- Difficulty
- Advanced
- Time to result
- ~ongoing to results
- Steps
- 4
- Confidence
- 90%
A reallocation rule for growth teams operating in fast-moving, competitive categories. Traditional growth spends most of its time optimizing existing user journeys (fixing drop-offs, tuning funnels); this framework flips it — most effort goes to standing up entirely new growth loops and features because the market moves faster than optimization can compound. Use it when you are in an exploding category with heavy competition, not a mature saturated market.
Origin
Elena Verna, Head of Growth at Lovable, describes throwing out most of her 15-20 year growth playbook. She found only 30-40% of her prior patterns transferred, and that optimization was 'not worth our time' because being ahead of competitors requires reinvention, not tweaking dials.
Core principles
- 01Being ahead is not optimization of the problem, it is reinvention of the solution
- 02Market opportunity is perishable — capture it by standing up many new loops fast
- 03Optimization compounds too slowly to outpace a category doubling every few months
- 04Even 'core product' work (integrations, agent behavior) can be a growth-team job
How to run it
- 1
Audit where your time goes
Measure what share of growth effort is optimization (funnel tweaks, A/B tests on existing journeys) versus innovation (new loops, new features). Most teams are 90%+ optimization.
Pro tip Optimization frameworks are abundant precisely because they are codifiable; innovation resists frameworks by definition.
- 2
Stop optimizing existing journeys
Deprioritize micro-optimizing acquisition-to-activation drop-offs. In a fast-moving category the returns are marginal versus the cost of missed loops.
Watch out More growth loops does not automatically mean more growth — but in a perishable, competitive market you must stand up many bets to capture demand.
- 3
Stand up new growth loops back-to-back
Have the growth team ship new features and loops one after another — even ones that traditionally belong to core product, like a Shopify integration or voice mode.
Pro tip Growth launching a core product integration is a signal you are doing this right, not a scope violation.
- 4
Go deeper into core product, not just the outer layers
Move growth work from 'smoothing surfaces' into actual product functionality — e.g. codifying agent instructions so customers activate better.
In the wild
Lovable's growth team launched a Shopify integration to enable the e-commerce use case, letting people vibe-code storefronts — work that 'usually would never come out of growth.'
→ Captured an emerging use case the market was already trying to force.
Instead of optimizing an activation funnel, the growth team began writing agent instructions and agentic workflows so customers reach the aha moment faster inside the agent itself.
→ Growth work moved 'deeper into product' rather than staying a smoothing layer.
Common mistakes
Copy-pasting old playbooks
Assuming the patterns that worked at prior scale-ups transfer. Elena found only ~30-40% did in an AI-native, fast-moving category.
Optimizing to the moon
Pouring effort into funnel optimization while competitors reinvent the solution and overtake you.
Is it for you?
Best for
Growth leaders at AI-native or emerging-category startups facing heavy competition and rapid market shifts
Not ideal for
Mature products in saturated, slow-moving markets where marginal optimization still yields the best ROI
From the transcript
“is not optimization of the problem. It's reinvention of the solution.”
“Right now I'm spending 95% innovating on growth and only 5% on optimization.”
“optimizations are just not worth our time. So, a lot of the times my growth team actually ends up working on new features or just…”
From the episode
The new AI growth playbook for 2026: How Lovable hit $200M ARR in one year
Elena Verna (Head of Growth)