Asia-Pacific app marketers are increasingly confronting a problem that is easy to miss on a conventional growth dashboard: acquisition can improve while monetization or longer-term retention weakens.
AppsFlyer’s State of Marketing in APAC 2026, which draws on 30 billion installs across nearly 12,000 apps and US$6.7 billion in ad spend, found several examples of those signals moving in different directions. TNGlobal previously reported on the divergence between installs, retention and paying-user behavior across APAC.
In this TNGlobal Q&A, Joen van Driel, Vice President of Sales, APAC at AppsFlyer, discusses how marketers should interpret those mixed signals, where cost per install can mislead budget decisions, why Day 30 retention deserves more attention, and how regional averages can obscure important differences between markets, platforms and verticals.
AppsFlyer’s data shows several cases where install growth and monetization are moving in opposite directions. When those signals conflict, which measures should marketers prioritize when making budget decisions, and which commonly used growth metrics risk being overvalued?
The data in this report makes a straightforward case: installs show how many people arrived, but not necessarily whether your business grew. Finance shows the clearest example. Installs grew across APAC while the share of paying users fell in some of its strongest markets. If a team had sized its budget on install growth alone, it would have mistakenly concluded the strategy was working.
The metrics we would elevate are share of paying users, revenue per user and Day 30 retention. These are harder to move than install volume, which is precisely why they are more honest. Cost per install is the metric most at risk of being overvalued right now. It measures top-of-funnel efficiency with no view of what happens below it.
Day 30 retention declined across every vertical, APAC sub-region and platform measured in the report, even where early retention remained relatively strong. What does that widening gap between initial engagement and longer-term retention tell marketers about the quality of the users they are acquiring?
The fact that Day 30 retention declined across every vertical, every sub-region and both platforms is the finding that should prompt the most serious reflection. What makes it particularly telling is that it held even where Day 7 retention was strong.
Finance in Southeast Asia, for example, saw Day 7 retention on Android grow 45 percent year over year. Day 30 retention for the same vertical, on the same platform, fell 32 percent. That divergence points to an acquisition quality problem. For most verticals, month one is when a user becomes commercially meaningful, so losing them before then means the acquisition spend is not compounding.
Finance provides an interesting example: installs grew in parts of APAC while the share of paying users weakened in some markets. How should marketers distinguish healthy user growth from acquisition that is simply increasing the top of the funnel without improving commercial outcomes?
The practical distinction is whether new users are reaching the same commercial milestones as the cohorts that preceded them. Marketers should track what share of each new acquisition cohort converts to a paying user, and where in the journey that conversion happens.
If that share falls as install volume rises, the funnel widens at the top and narrows faster below it. What happens after acquisition should be the core focus, showing that growing installs and growing your business are not the same thing.
Gaming moved in almost the opposite direction, with Android installs declining across APAC sub-regions while the share of paying users increased. At what point can declining acquisition volumes coexist with a healthier business, and how should marketers determine whether they are attracting fewer but more valuable users?
Gaming’s numbers in this report are a useful corrective to the assumption that falling installs signal a market in trouble. Android installs fell between 17 percent and 26 percent across every APAC sub-region. Yet the share of paying users grew across every region and platform, including 29 percent in the Indian Subcontinent on iOS and 28 percent in Indonesia on Android.
Declining acquisition can coexist with a healthier business when the users being acquired are more likely to convert and spend than the ones being lost. That typically reflects tighter targeting or a product experience that has improved enough to convert a higher share of a smaller audience. Marketers should ask not just how many users they acquired, but what share paid, and whether that share is improving over time.
Indonesia saw finance remarketing conversions rise sharply, while paid installs grew only modestly. What does that suggest about the balance between acquiring new users and generating more value from existing ones? When does remarketing deserve a larger share of the growth budget?
For many APAC markets, the most valuable users are already in the database. The growth question is whether teams are investing accordingly. Indonesia is one of the report’s clearest illustrations of where the growth budget is finding better returns.
Remarketing deserves a larger share of the growth budget when reactivating an existing user costs less than acquiring a new one, and when the reactivated user is more likely to reach a commercial outcome. In a market like Indonesia, where digital payments are expanding and finance apps are already widely used, remarketing is increasingly likely to outperform new user acquisition. Teams still focusing too much of their growth budget on new user acquisition in that environment should examine that assumption closely.
Vietnam’s entertainment market recorded particularly strong growth in user-acquisition spending. When spending rises much faster than the wider regional market, what should marketers look at to determine whether that investment is producing sustainable growth rather than simply reflecting higher competition or acquisition costs?
Vietnam’s entertainment market recorded user-acquisition spend growth of 194 percent on Android, more than seven times the regional pace. That kind of acceleration warrants careful scrutiny, because rapid spend growth in a single market can reflect genuine demand or simply a more competitive auction driving up costs without a corresponding improvement in outcomes.
Three questions are worth asking. Are install volume and paying-user share growing in proportion to spend? Is retention for that market keeping pace with the regional benchmark? And is the cost of each new user rising or holding steady? If spend is outpacing both installs and commercial outcomes, growth likely reflects cost inflation rather than market expansion. The benchmark should be the same market’s prior performance at a lower spend level.
The report also shows how regional averages can conceal very different fraud trends by market, platform and vertical. How should marketing teams incorporate fraud quality into performance measurement so that apparently strong acquisition or monetization figures are not evaluated in isolation?
The fraud data in this report makes a case that regional averages are among the least useful numbers a marketing team can track. Shopping’s iOS fraud rate in Southeast Asia fell 90 percent. Finance’s iOS fraud rate in the same region fell 80 percent. Those are meaningful improvements. But gaming’s Android fraud rate in Japan rose 68 percent over the same year, and entertainment’s iOS fraud rate in the Indian Subcontinent rose 170 percent, both within a period when the regional headline was moving in a positive direction.
The practical implication is that a team tracking fraud as one blended regional number might assume it is under control everywhere and miss a spike in a specific market, platform or vertical. Fraud measurement needs to operate at the market, platform and vertical level. Teams should treat fraud rate as a dimension of every performance report and flag any market or channel moving opposite the regional trend, since that is exactly where teams get blindsided.
Looking toward 2027, if you could redesign the typical mobile-marketing dashboard used by an APAC growth team, which metrics would you give more weight to, which would you demote, and which indicators need to be considered together rather than separately?
Most APAC growth teams’ dashboards today are weighted toward the top of the funnel: installs, cost per install and early engagement. This report’s findings suggest that weighting is increasingly misaligned with where value is actually created and lost.
The metrics I would weight more are share of paying users tracked by cohort and market, Day 30 retention as a standard reporting metric rather than an occasional diagnostic, and fraud rate broken down by platform and vertical rather than reported as a single figure. I would demote cost per install as a primary optimization target. It is useful as a cost-control measure, but it has no direct link to commercial outcomes and can mislead budget decisions when install quality varies across channels.
I’d also stop scoping the dashboard to mobile alone. As measurement expands into web, CTV, AI chatbots and other channels, a team reading only mobile signals is working from an incomplete picture of the customer.
The indicators that need to be considered together are Day 7 and Day 30 retention as a pair, install volume and share of paying users in the same view, and remarketing conversion rates alongside new-user acquisition costs. None of those pairs tell a complete story in isolation. And every one of them should be visible by market, vertical and platform by default. The report’s clearest lesson is that regional averages routinely conceal the most important signals, and dashboards need to show that divergence.
Joen van Driel is Vice President of Sales, APAC at AppsFlyer and leads the company’s Asia-Pacific sales business. Based in Thailand, he has more than 25 years of technology industry experience across sales management, strategy and business development. He previously held senior roles at Elastic and Oracle and served in country management positions covering Google Cloud, Google Enterprise Search, Adobe and Salesforce.com. He earned a doctorate in financial economics from the University of Amsterdam.
Editor’s note: This Q&A has been lightly edited for clarity and TNGlobal house style. The substance of the interviewee’s responses has been preserved.
Share your perspective: TNGlobal welcomes contributed insights and expert commentary from across Asia’s technology and innovation ecosystem. Submit a contribution for editorial consideration, or explore more conversations in our TNGlobal INSIDER and TNGlobal Q&A and Interviews archive.