Why Sustainable App Growth Takes More than AI-Driven Acquisition
Rakuten Viber’s Kiril Filipov, Polarsteps’ Laurens Jansen, MyPie’s Kirill Pospelov and REPLUG’s Lorenzo Rossi share why sustainable app growth requires more than AI-driven acquisition, focusing on product value, retention and CX.
Topics
The promise of AI in app marketing is straightforward: let machines handle repetitive decisions, process more signals and improve campaign performance at a scale human teams cannot match.
The difficulty begins when an efficient decision is not necessarily the right one.
App marketing has become exceptionally good at bringing people through the door. Campaigns can be optimised in real time, creatives tested at scale and acquisition measured down to the cost of an install. Yet, an install is only the beginning of the relationship, and increasingly, an expensive one.
The harder question is what happens after someone arrives.
- Does the app fulfil the promise that brought them there?
- Does the experience give them a reason to return?
- And if growth depends on buying the next wave of users, how sustainable is the model?
As artificial intelligence takes on more of the work behind targeting, bidding and optimisation, these questions are becoming harder to ignore.
Kiril Filipov of Rakuten Viber, Laurens Jansen of Polarsteps, Kirill Pospelov of MyPie and Lorenzo Rossi of REPLUG approach the problem from different angles, but their views converge on one point: automating acquisition is not the same as building a growth engine.
The next advantage may belong to marketers who understand what makes that attention worth keeping.
The Limits of Automation & Where Human Judgement Matters
Kirill Filipov, Marketing Lead Europe at Rakuten Viber, draws a distinction between tasks that can be automated and decisions that still need context.
“Hand over what’s repetitive, measurable and easy to reverse, like bid optimisation, analytics, A/B testing and creative variations at scale. Keep human anything that needs judgement or could cost us trust.”
His example is less about technology than timing. During summer wildfires in Bulgaria and Greece, an algorithm might continue serving an otherwise successful campaign. A marketer, aware of what is happening outside the dashboard, might recognise that the message is inappropriate and pause it.
“While the algorithm sees a campaign performing well, a human sees that the timing is wrong and pauses it,” Filipov says.
The distinction matters as automation becomes more capable. Performance data can tell a team whether a campaign is working against its chosen objective. It cannot always establish whether that objective remains appropriate.
Filipov sees AI as an amplifier of existing ideas and tactics, rather than a substitute for the person deciding where they should lead. That puts a different responsibility on marketers: automation can accelerate execution, but judgement must still determine the direction.
The Retention Problem Starts Before the First Session
A user who installs an app has responded to a promise. Whether that promise survives the first session depends on the product, the experience and how closely both match the expectations created by marketing.
“If I had to put a number on it, marketing owns maybe a third — who we bring in and what we promise them. The other two-thirds happen after the first session,” Filipov says.
He points to Viber Dating as an example of why reach alone is insufficient. An established audience does not automatically understand a new feature, and broad messaging cannot guarantee that the feature will be relevant to everyone who encounters it. Targeting, an honest promise and an experience that competes with dedicated dating apps all matter.
Laurens Jansen, Head of Growth at Polarsteps, reaches a similar conclusion from the product side. “The core is always the product, and that doesn’t change.”
Marketing can improve performance, he argues, but it cannot compensate indefinitely for a product that fails to meet expectations. That gap is especially costly when acquisition campaigns succeed in bringing in users who quickly discover that the experience is not what they anticipated.
- The promise: Does the advertising accurately represent what the app offers?
- The first session: Can users get started without unnecessary friction?
- The moment of value: Does the product solve a problem or meet a need?
- The reason to return: Is there a continuing benefit, rather than another message asking for attention?
For Kirill Pospelov, Head of Growth at MyPie, the relationship between product value and retention is even more direct. “If users are not satisfied with the product, you’ll never convince them otherwise.”
That is a difficult reality for marketers accustomed to treating engagement as a communications challenge. More notifications may create more activity, but activity is not proof that the product has become useful. Retention begins when the user has a reason to stay.
When Acquiring More Users Stops Being the Answer
Lorenzo Rossi, Co-Founder of REPLUG, looks for the warning signs in the distance between acquisition performance and business results. An app can report a healthy cost per install while the cost of acquiring a paying customer rises.
Users may register without completing onboarding, start a trial without subscribing, or make a first purchase and never return.
“Sometimes a company genuinely needs more demand. There are also situations where product-market fit, pricing, or audience quality are the real issues,” Rossi says. “But before recommending additional acquisition investment, I want to understand how much value the company is getting from the demand it’s already creating.”
The distinction changes where growth teams look for improvement. Instead of immediately increasing media budgets, they examine where the customer journey breaks down and which missed conversions represent lost revenue.
Rossi cites a delivery app where improvements to onboarding and customer relationship management helped increase first purchases by 24.7% and repeat purchases by 18.5% within three months.
The work focused on users who had registered but never purchased, alongside existing customers who could be encouraged to return. The lesson is not that acquisition has become less important. It is that the economics of growth cannot be understood through acquisition metrics alone.
When performance, product and retention teams work towards different objectives, a campaign can look successful even while the wider business struggles to convert that success into revenue.
The Next Growth Advantage May Come From Beyond Paid Media
For years, app growth has often been organised around a familiar set of acquisition channels. But rising costs and changing discovery habits are forcing marketers to reconsider how users find products in the first place.
“We are at the start of a distribution shift,” says Jansen. “I’m a big believer in creating value for users rather than only ‘selling’ to them.”
For him, organic content should offer genuine value rather than rely on short-lived tricks designed to exploit a platform. Community building, word of mouth and real-world experiences also offer ways to establish relationships beyond paid impressions.
Rossi agrees that diversification means more than adding another advertising platform. His work with mobile apps has included expanding campaigns across TikTok, Reddit, Snapchat and other channels, but he stresses that measurement must accompany that expansion.
Otherwise, businesses risk spreading spending without understanding which audiences produce valuable customers.
His broader test is revealing: “If your biggest advertising channel stopped performing tomorrow, how much of your growth engine would still work?”
- Organic discovery: Create useful content that attracts people with a genuine interest in the product.
- Community and referrals: Give existing users reasons to recommend the app and bring others into the relationship.
- Channel diversification: Test new platforms while measuring downstream revenue, not just installs.
- Customer engagement: Build reasons for people to return without repeatedly paying to reach them.
Jansen believes apps dependent on paid or search traffic should already be experimenting with alternatives. The ecosystem is still developing, but waiting for established channels to weaken before exploring new ones could leave businesses with fewer options when they need them most.
The Human Advantage in an Automated Market
There is an irony in app marketing’s AI transition. As more campaign decisions become automated, the qualities that distinguish a lasting customer relationship may become harder to replicate through automation alone.
Pospelov believes brands need to connect with users emotionally as well as functionally. Stronger retention can increase lifetime value and help offset rising acquisition costs, but the relationship must offer something beyond another prompt to engage.
“‘Call to action’ — users are getting tired of constant calls to do something. They’re bombarded with thousands of them every day,” he says. His alternative is “set and forget”: products should adapt to existing habits rather than continually demand attention.
That idea challenges a long-standing assumption in app marketing: that more interventions necessarily produce more engagement. Sometimes, the better experience is the one that asks less of the user.
Filipov’s distinction between measurable tasks and human judgement reinforces the point. Automation can identify patterns, test variations and scale activity. It still needs a clear understanding of what users value, what a product can deliver and when a message should not be sent.
The next growth model, then, is unlikely to emerge from a single tool or channel. It will depend on how well businesses connect acquisition with product experience, retention with revenue, and automation with decisions that respect the customer.
The measure of progress will not simply be how efficiently an app finds its next user. It will be whether that user has a reason to stay — and whether the business can grow without having to buy the relationship all over again.
ALSO READ: The New Loyalty Equation: Value, Recognition and the Race for Relevance
