The Hidden Forces Reshaping App Growth: Lessons from Flo Health

Ahead of her appearance at Appsforum Lisbon, Flo Health’s Senior Product Manager, Ekaterina Elkina, shares her perspective on how privacy, AI, and retention are reshaping modern app growth.

What has fundamentally changed about app growth in the last two years that many teams are still underestimating?

Privacy restrictions and AI integrations have significantly reshaped app growth over the last several years, with persistent impacts on attribution, monetization flexibility, and measurement accuracy.Apple's ATT (launched 2021), combined with SKAdNetwork limits and third-party cookie declines, leads to restricted access to individual user data.

Pure subscription fatigue poses real risks: users get tired of endless monthly/yearly fees, driving higher churn and subscription resistance.AI now plays a bigger hidden role in app growth, powering key areas but often added too late or ignored.

As user acquisition costs increase, where should product teams focus first to drive sustainable growth?

In this case, it makes sense to prioritize retention and onboarding optimization first, along with user segmentation. These will help to deliver faster, cheaper sustainable growth than buying more installs because high-quality installs matter more than volume.Next, focus on activating personalized engagement through push notifications and lifecycle messaging.

Predictive models for churn risk and user lifetime value forecasting deliver reliable improvements with clean data and clear KPIs like D7 retention or ARPU for example.

Onboarding is the ideal first touchpoint for requesting push permissions, but always explain the value users will receive - it affects notifications acceptance rate significantly. Finally, set goals to improve D7/D30 retention and ARPU, rather than focusing solely on installs.

In your experience, where do growth initiatives fail most often - strategy, execution, or alignment with product?

In general, from my 18+ years in IT and technical background, the more entropy in a system (disorder, uncertainty, complexity), the higher the failure risk. More dependencies, variables, and unpredictability multiply failure points exponentially. Strategy can sound good on paper but fail due to poor product-market fit, for example, though this can be validated early.

Product disconnections happen but can be fixed with proper goals, metrics, and clear prioritization. Execution creates the most friction through details like technical debt, excessive cross-team coordination, unclear ownership, unrealistic timelines, insufficient testing, even scaling too fast without proper infrastructure - all of this adding complexity.

Where does AI genuinely move growth metrics today, and where is it still mostly hype?

AI genuinely moves growth metrics through personalized creatives and predictions. It generates ad variants, onboarding screens, and personalized content much faster, boosting retention and LTV when A/B tested properly. Predictive models for churn risk and user lifetime value forecasting deliver reliable improvements with clean data and clear KPIs like D7 retention or ARPU for example. AI excels at reviewing content against specific rules.

Mostly hype: fully autonomous agents replacing humans, creative breakthroughs, or "thinking outside the box."

AI still can't handle complex growth strategies, invent bold new ideas, or replace PM judgment. It needs human context, good prompts, and oversight. This is the current state, but AI will continue evolving.

What is the most common mistake teams make when trying to scale a growth win across markets or segments?

The most common mistake when scaling growth wins across markets or segments is assuming one-size-fits-all works. I am a strong advocate for behavioral psychology, and we create products for people. Different people across markets and segments have different "jobs to be done", payment habits, privacy norms, and cultural expectations vary widely. What works for one market or segment can be totally irrelevant for another. I'd suggest scaling carefully: avoid blind copy-pasting, validate in 1-2 similar markets/segments first, research local behaviors, and iterate through experiments.

At Flo, what product decision had a disproportionately positive impact on growth that wasn’t originally framed as a “growth” initiative?

While at Flo we always keep growth in mind, I'd highlight the Flo for Partners feature. Its goal: boost a partner's empathy and understanding of her cycle, making them feel more connected.

For her: Feel understood and supported by sharing what's happening with her body.

For him: Better understand her rhythms to provide real support. This sparked last year's "Flo pareja" TikTok trend, driving organic virality and couple retention without UA spend.

How does operating in a high-trust health category change the way you think about growth compared to mainstream consumer apps?

The high-trust health category emphasizes trust and health by its very name. This requires critical focus on privacy, user data protection, regulations, and medical credibility.Unlike fun consumer apps, users here want clear benefits first before sharing personal health info. Health data feels very private, so one mistake can ruin user trust. Users choose more trusted brands over just feature parity.

You’re on the Future of App Growth panel at Appsforum Lisbon - what are you most looking forward to discussing?

I'm excited to dive deeper at Appsforum Lisbon into onboarding hacks, retention strategies, monetization models (including hybrids), and AI's 2026+ impact - all while prioritizing privacy, which is crucial especially for high-trust apps in today's world.