According to AppsFlyer's 2026 State of App Marketing report, global app install ad spend reached $65 billion in 2025, a 12% year-over-year increase driven primarily by Android growth in emerging markets and a resurgence of iOS spending as advertisers adapted to SKAdNetwork 4.0 and Apple's AdAttributionKit.
The report reveals that remarketing now drives 30% of all conversions for non-gaming apps (up from 23% in 2023, per AppsFlyer’s earlier benchmarks), signaling a structural shift toward retention-first strategies. According to AppsFlyer’s 2025 retargeting benchmarks, apps running retargeting campaigns see median 30% lift in conversion rates versus non-retargeted cohorts.
Day 30 retention rates across all categories averaged just 4.1% according to AppsFlyer, but apps investing in remarketing saw Day 30 retention of 8.7%, more than double the baseline.
Perhaps most critically for budget allocation, the cost-per-install gap between iOS and Android widened to 3.8x in North America (up from 3.2x in 2024 per AppsFlyer's prior year data), forcing growth teams to rethink platform allocation entirely.
Page Contents
What is the average cost per install (CPI) by app category and platform in 2025?
| App Category | iOS CPI (North America) | Android CPI (North America) | iOS CPI (Global) | Android CPI (Global) |
|---|---|---|---|---|
| Gaming (Casual) | $2.40 | $0.75 | $1.60 | $0.42 |
| Gaming (Midcore/Hardcore) | $5.80 | $1.90 | $3.90 | $0.95 |
| Shopping / eCommerce | $3.50 | $1.10 | $2.30 | $0.55 |
| Finance / Fintech | $6.20 | $2.40 | $4.10 | $1.20 |
| Food & Drink | $4.10 | $1.30 | $2.70 | $0.65 |
| Health & Fitness | $3.80 | $1.00 | $2.50 | $0.50 |
| Entertainment / Streaming | $3.20 | $0.90 | $2.10 | $0.48 |
| Travel | $4.50 | $1.60 | $3.00 | $0.80 |
| Social / Dating | $3.90 | $1.20 | $2.60 | $0.60 |
| Utilities / Productivity | $2.10 | $0.55 | $1.40 | $0.30 |
What are the average retention rates by app category in 2025?
| App Category | Day 1 Retention | Day 7 Retention | Day 14 Retention | Day 30 Retention |
|---|---|---|---|---|
| Gaming (Casual) | 28% | 10% | 6.2% | 3.5% |
| Gaming (Midcore) | 22% | 8% | 5.0% | 3.0% |
| Shopping / eCommerce | 24% | 11% | 7.5% | 5.2% |
| Finance / Fintech | 26% | 14% | 10.1% | 7.8% |
| Food & Drink | 20% | 9% | 5.8% | 3.9% |
| Health & Fitness | 21% | 9.5% | 5.5% | 3.4% |
| Entertainment / Streaming | 25% | 12% | 8.0% | 5.5% |
| Travel | 16% | 6% | 3.5% | 2.1% |
| Social / Dating | 23% | 10% | 6.5% | 4.0% |
| All Categories Average | 23% | 10% | 6.4% | 4.1% |
How does remarketing impact retention and ROAS by vertical?
Need help scaling your mobile app growth? Talk to RocketShip HQ about how we apply these strategies for apps spending $50K+/month on UA.
| App Category | Day 30 Retention (No Remarketing) | Day 30 Retention (With Remarketing) | Lift from Remarketing | Remarketing Share of Conversions |
|---|---|---|---|---|
| Shopping / eCommerce | 5.2% | 12.8% | +146% | 42% |
| Food & Drink | 3.9% | 10.1% | +159% | 38% |
| Travel | 2.1% | 6.5% | +210% | 35% |
| Finance / Fintech | 7.8% | 14.2% | +82% | 28% |
| Entertainment / Streaming | 5.5% | 11.0% | +100% | 25% |
| Health & Fitness | 3.4% | 7.8% | +129% | 20% |
| Gaming (Casual) | 3.5% | 5.8% | +66% | 12% |
| Gaming (Midcore) | 3.0% | 5.2% | +73% | 10% |
| Non-Gaming Average | 4.7% | 10.4% | +121% | 30% |
How has iOS vs Android ad spend share shifted from 2023 to 2025?
| Region | iOS Share 2023 | iOS Share 2024 | iOS Share 2025 | Android Share 2025 | YoY iOS Spend Change |
|---|---|---|---|---|---|
| North America | 58% | 54% | 51% | 49% | +6% |
| Western Europe | 48% | 45% | 43% | 57% | +4% |
| APAC (Developed) | 42% | 40% | 39% | 61% | +3% |
| APAC (Emerging) | 12% | 11% | 10% | 90% | +8% |
| Latin America | 18% | 16% | 15% | 85% | +12% |
| Middle East & Africa | 15% | 14% | 13% | 87% | +10% |
| Eastern Europe | 22% | 20% | 18% | 82% | +5% |
| Global Average | 35% | 33% | 31% | 69% | +7% |
What percentage of installs have attribution data available post-ATT?
| Attribution Signal | 2022 Rate | 2023 Rate | 2024 Rate | 2025 Rate | Trend |
|---|---|---|---|---|---|
| ATT Opt-In Rate (All iOS) | 25% | 27% | 30% | 33% | Gradually rising |
| ATT Opt-In Rate (Gaming) | 18% | 20% | 22% | 25% | Slowly rising |
| ATT Opt-In Rate (Non-Gaming) | 30% | 33% | 36% | 39% | Steadily rising |
| SKAdNetwork Coverage (installs) | 62% | 70% | 75% | 78% | Nearing ceiling |
| Modeled Attribution (AppsFlyer SSOT) | N/A | 40% | 55% | 68% | Rapidly expanding |
| Web-to-App Flows (Measurable) | 10% | 18% | 28% | 35% | Growing fast |
| Android Privacy Sandbox Adoption | 0% | 2% | 8% | 15% | Early but accelerating |
| CTV/OTT Attribution Coverage | 5% | 8% | 12% | 18% | Emerging channel |
Analysis
Android emerging-market growth vs.
iOS maturation. The 12% growth in global install ad spend to $65 billion, according to AppsFlyer’s 2026 report, masks a significant divergence: Android spend in emerging markets (APAC, Latin America, MENA) grew 18% year-over-year per AppsFlyer, while iOS spend in North America grew only 6%. The AppsFlyer Performance Index rankings confirm that Google Ads and Meta remain dominant self-attributing networks for both gaming and non-gaming apps across these regions in 2025/2026.
iOS buying has matured. Growth teams now operate profitably within SKAdNetwork’s measurement constraints and compete fiercely for a smaller, higher-quality user pool, driving CPIs upward.
The widening iOS-to-Android CPI gap (3.8x in North America per AppsFlyer, up from 3.2x the prior year) is the defining budget allocation challenge for 2026. Finance apps pay $6.20 per iOS install in North America versus $2.40 on Android according to the report. These Meta CPI benchmarks for mobile app installs consistently show iOS commanding a meaningful premium over Android across categories, with high-intent verticals like fintech among the most expensive on the platform.
For apps where LTV justifies it (fintech, subscription media), iOS remains the higher-ROAS platform. For apps with lower monetization ceilings, the math increasingly favors Android-first channel strategies.
Remarketing ROI superiority. Remarketing’s surge to 30% of non-gaming conversions (per AppsFlyer) is the most actionable finding. Shopping and food delivery apps generate 38-42% of their measured conversions from remarketing, according to the report. Understanding how mobile app retargeting works is critical: in our experience, re-engaging lapsed users is consistently more cost-efficient than acquiring net-new ones.
Industry patterns suggest the marginal dollar spent on re-engaging a lapsed user tends to deliver stronger ROAS than acquiring a net-new one — a dynamic we observe most clearly in eCommerce app campaigns.
The AppsFlyer eCommerce app marketing benchmarks corroborate this with vertical-specific data. Meanwhile, retention remains the industry's open wound. A 4.1% average Day 30 retention rate per AppsFlyer means fewer than 5 in every 100 acquired users are active a month later.
Travel apps are worst at 2.1%, while fintech leads at 7.8%, a gap explained by transaction frequency (daily utility vs. episodic need). The implication: UA budget allocation must account for category-specific retention curves, not industry averages.
Attribution recovery via SKAN 4.0 and SSOT. The ATT opt-in rate reaching 33% globally per AppsFlyer is higher than most predicted when iOS 14.5 launched. Non-gaming apps now see 39% opt-in rates according to the report.
Combined with SKAN 4.0’s finer postback windows and AppsFlyer’s SSOT modeled attribution covering 68% of installs, growth teams have significantly more signal than the 2022 post-ATT nadir. Building a comprehensive mobile measurement framework after ATT helps teams implementing this layered approach meaningfully recover installs that would otherwise go unattributed.
Web-to-app flows now cover 35% of measurable attribution paths per AppsFlyer (up from 10% in 2022), as advertisers route users through mobile web landing pages to recover deterministic signals. This tactic ties directly into holistic UA strategy and is especially prevalent in eCommerce and fintech.
Android's Privacy Sandbox adoption at 15% per AppsFlyer is still early but accelerating. Growth teams should prepare for an Android measurement landscape that, within 18-24 months, will resemble today's iOS environment.
What This Means For You
- Shift 15-25% of remarketing budget toward lapsed-user re-engagement campaigns, particularly for eCommerce, food delivery, and travel apps, where AppsFlyer data shows remarketing lifts Day 30 retention by 100-210%. In our experience managing remarketing campaigns, monitoring metric changes in proportion to spend ensures that significant ROAS drops on high-spend campaigns trigger action faster than larger percentage swings on small tests, helping teams cut through noise and focus on what matters.
- For apps with sub-$5 average revenue per user, conduct a rigorous iOS vs Android ROAS analysis using the CPI benchmarks above. With North American iOS CPIs running 3.8x higher than Android per AppsFlyer, the breakeven LTV threshold on iOS is now $15+ for most non-gaming categories. If your Day 90 LTV does not clear that bar, reallocate iOS budget to Android and reinvest savings into creative testing (see our guide on high-performing mobile ad elements).
- Implement web-to-app attribution flows immediately if you have not already. With 35% measurability according to AppsFlyer (up from 10% in 2022), this is the fastest-growing attribution signal for iOS campaigns. Route paid traffic through mobile web landing pages before the App Store redirect to recover deterministic click-level data that SKAdNetwork alone cannot provide.
- Prepare for Android Privacy Sandbox by auditing your current Android measurement stack now. According to AppsFlyer, adoption is at 15% and accelerating. Teams that built measurement resilience for iOS post-ATT are better positioned. Those relying on Android's open attribution as a crutch have 12-18 months to build redundancy before signal loss begins at scale.
- Tie UA spend to category-specific retention benchmarks rather than blended averages. According to AppsFlyer, the difference between fintech Day 30 retention (7.8%) and travel Day 30 retention (2.1%) is nearly 4x, which means a $4.50 travel CPI requires fundamentally different LTV modeling than a $6.20 fintech CPI. Use the Adjust State of App Growth benchmarks alongside AppsFlyer's data to triangulate your category's true payback window.
Frequently Asked Questions
Why is the iOS-to-Android CPI gap widening, and will it continue?
The gap is widening because iOS user quality (measured by downstream LTV) remains significantly higher for most categories, so advertisers with strong unit economics keep bidding up a shrinking deterministic pool. According to AppsFlyer's 2026 report, the North America gap moved from 3.2x to 3.8x in a single year.
As Android Privacy Sandbox matures and Android attribution becomes less open, expect a partial convergence, but iOS will likely maintain a 2-3x premium due to structurally higher purchasing power among iOS users.
How should I decide between investing in remarketing versus new user acquisition?
Start with your marginal ROAS comparison. In our experience, remarketing dollars frequently deliver meaningfully higher ROAS than net-new acquisition dollars for apps with an established existing user base.
If your Day 30 retention is below your category average per AppsFlyer (e.g., below 5.2% for shopping apps), you likely have a large lapsed audience to re-engage before incrementally scaling new UA spend.
What does a 33% ATT opt-in rate actually mean for my iOS campaign measurement?
It means roughly one-third of your iOS users grant tracking permission, providing full deterministic attribution. But combined with SKAN 4.0 postbacks (78% coverage per AppsFlyer) and modeled attribution via SSOT (68% coverage), you can piece together a reasonably complete picture for most installs.
The practical gap is in real-time optimization: deterministic data arrives instantly, while SKAN postbacks are delayed 24-48 hours, requiring more patience in campaign management cycles.
Are web-to-app flows worth the added funnel complexity for smaller apps?
Yes, if your monthly iOS ad spend exceeds $50K. According to AppsFlyer, web-to-app attribution coverage reached 35% in 2025, making it the single most effective way to recover deterministic iOS signals.
The tradeoff is a measurable drop-off at the web-to-store redirect, so you need sufficient volume for the recovered signal to offset the conversion rate loss.
How should growth teams prepare for Android Privacy Sandbox at only 15% adoption?
Treat the next 12-18 months as a preparation window, not a waiting period. According to AppsFlyer, Android Privacy Sandbox adoption is accelerating from 8% to 15% in one year. Start by running parallel attribution using both traditional Android attribution and Privacy Sandbox APIs for Android marketers, which include Topics (~470 topics) and Attribution Reporting as GAID deprecation begins rolling out to 2.5 billion+ active Android 13+ devices.
Build a growth team with at least one member dedicated to measurement infrastructure who can manage the transition before it becomes mandatory.
Why do gaming apps see much lower remarketing lift than non-gaming apps?
Gaming churn is driven primarily by content exhaustion and difficulty walls, which remarketing ads cannot solve. A push notification or retargeting ad can remind a lapsed shopping user to complete a purchase, but it cannot restore a lapsed gamer's interest in levels they have already beaten.
Per AppsFlyer's data, gaming remarketing lifts Day 30 retention by 66-73% versus 82-210% for non-gaming verticals. Gaming apps are better served investing that budget into creative-driven new user acquisition.
Is CTV/OTT a viable mobile UA channel given 18% attribution coverage?
It is viable as a top-of-funnel awareness play, but not yet as a direct-response performance channel. According to AppsFlyer, CTV/OTT attribution coverage grew from 5% to 18% between 2022 and 2025, but the measurement gap makes it difficult to optimize toward CPI or ROAS targets.
In our experience, CTV works best for apps that have already saturated major performance channels like Meta, Google, and TikTok and are looking for incremental upper-funnel reach.
What is the single biggest mistake growth teams make when using these benchmarks?
Comparing their own metrics to cross-category averages instead of their specific vertical. According to AppsFlyer, Day 30 retention ranges from 2.1% (travel) to 7.8% (fintech), meaning a travel app at 3% retention is outperforming its category while a fintech app at 5% is underperforming.
Always benchmark against your vertical, then against your own trailing 90-day performance, before comparing to industry-wide figures. See our guide on growth marketing vs. UA for a framework on contextualizing these metrics.
Looking to scale your mobile app growth with performance creative that delivers results? Talk to RocketShip HQ to learn how our frameworks can work for your app.
Not ready yet? Get strategies and tips from the leading edge of mobile growth in a generative AI world: subscribe to our newsletter.
Related Reading
- The complete guide to mobile user acquisition (comprehensive guide)
- Adjust State of App Growth Report: Global Trends and Benchmarks (2026)
- AppsFlyer State of eCommerce App Marketing Report: UA and Retention Benchmarks (2026)
- What Are the Best Paid Channels for Mobile App User Acquisition?
- What Are Fail Ads and Why Do They Work for Mobile Games?
