CPI Benchmarks by App Category and Platform: 2026 Data
Cost per install is one of the most-watched numbers in paid app marketing — and one of the most misread. Teams panic when their CPI climbs above a category average they found in a two-year-old blog post, or they celebrate a low CPI that's actually a sign of weak targeting. Neither reaction is grounded.
This post lays out 2026 CPI benchmarks by app category and platform, explains what's moved and why, and gives you a framework for deciding whether your number is a problem or not.
How These Benchmarks Are Constructed
No single dataset covers every app, every channel, and every geography. What follows draws on publicly available benchmark reports from mobile measurement partners (MMPs) like AppsFlyer, Adjust, and Sensor Tower, combined with directional observations from our own client engagements across iOS and Android. Where ranges are wide, that's intentional — a CPI benchmark is not a price tag. It's a signal band.
A few things that affect where you land within any range:
- Geo targeting. Tier-1 markets (US, UK, AU, CA) consistently run 2–4× higher CPIs than Tier-2 or Tier-3 markets.
- Channel mix. Apple Search Ads typically yields higher CPIs than broad-audience Meta campaigns, but with meaningfully better downstream quality.
- Creative quality. A well-optimized creative on TikTok can undercut the category average by 30–40%. A stale creative on the same platform can double it.
- Bidding model. Target CPA bidding versus manual CPC produces different CPIs even on identical placements.
Use the table below as a starting point, not a ceiling.
2026 CPI Benchmarks by App Category
The figures below represent blended averages across US-targeted campaigns (iOS and Android combined), with platform splits noted where the delta is significant. Ranges reflect the 25th–75th percentile band observed in published 2025–2026 industry reports.
| App Category | iOS CPI (USD) | Android CPI (USD) | Notes |
|---|---|---|---|
| Gaming — Casual | $0.90 – $2.50 | $0.40 – $1.20 | High volume, competitive creative landscape |
| Gaming — Midcore / Strategy | $2.50 – $6.00 | $1.20 – $3.50 | Longer LTV windows justify higher CPIs |
| Gaming — Hypercasual | $0.30 – $0.90 | $0.15 – $0.55 | Volume play; watch D1 retention closely |
| Finance & Fintech | $3.50 – $9.00 | $1.80 – $5.50 | Regulated category; Apple Search Ads dominant |
| Health & Fitness | $2.00 – $5.50 | $1.00 – $3.00 | Strong subscription attach rate |
| Shopping & Retail | $1.50 – $4.00 | $0.80 – $2.50 | Seasonal spikes during Q4 |
| Food & Delivery | $1.80 – $4.50 | $0.90 – $2.80 | Promo-driven installs common; check retention |
| Utilities | $1.20 – $3.50 | $0.60 – $2.00 | Wide variance by feature set |
| Travel | $2.50 – $7.00 | $1.50 – $4.50 | Post-pandemic demand still elevated |
| Education & EdTech | $2.00 – $6.00 | $1.00 – $3.50 | Parental targeting raises CPIs |
| Healthcare | $3.00 – $8.00 | $1.50 – $5.00 | HIPAA constraints limit channel options |
| B2B / Productivity | $4.00 – $12.00 | $2.00 – $7.00 | Small audiences, high LTV — CPI alone is misleading |
A few pattern observations:
iOS CPIs run roughly 1.5–2.5× Android across every category. This gap narrowed slightly in 2024 after ATT stabilization but has widened again in 2025–2026 as advertisers competing for Apple Search Ads inventory intensified.
Finance and B2B are the highest-CPI verticals, and correctly so. An acquired user in a fintech app or enterprise productivity tool can be worth hundreds of dollars in LTV. Benchmarking a $7 CPI against a casual gaming average is a category error.
Hypercasual gaming is effectively a volume arbitrage play. CPIs are low, but so are retention rates. If you're not seeing D7 retention above approximately 10–12%, a $0.50 CPI is still expensive relative to what you're building.
What's Shifted in 2026
A few specific movements are worth flagging:
Apple Search Ads volume pressure. Competition for ASA placements has increased substantially. Categories like finance, health, and productivity — where app store intent signals are strongest — have seen CPIs on ASA climb approximately 15–25% year-over-year based on published MMP trend data. Teams that relied heavily on ASA as a "cheap quality channel" need to revisit their channel mix.
TikTok as a viable install channel. TikTok's App Profile Pages and dedicated app install formats have matured. In our engagements, we're seeing it perform competitively against Meta for consumer apps targeting 18–34, particularly in fitness, food, and entertainment categories. CPIs are often lower, but creative production cadence needs to be higher — UGC-style content degrades faster than produced ads.
Google App Campaigns (UAC) remain the volume workhorse. For Android-heavy strategies or global rollouts, GAC still delivers scale that other channels can't match. The tradeoff is visibility — the black-box nature of GAC makes attribution and creative learning harder. Pair it with a strong mobile measurement partner to understand what's actually working.
Meta's advantage for behavioral targeting is narrowing. Post-ATT, Meta's lookalike modeling has been slower to recover than many anticipated. CPMs have climbed, and we're seeing teams shift budget toward contextual-signal channels (ASA, keyword-based Google) and toward creative-heavy strategies where the ad itself does the qualification work.
If you're rethinking your channel mix or want a second opinion on whether your CPIs reflect a strategy problem or a creative problem, the Semnexus mobile app marketing team runs paid UA across iOS and Android.
CPI Is Not the Metric You Should Optimize For
This deserves a direct statement: optimizing for lowest CPI is one of the most common and expensive mistakes in app marketing.
CPI measures the cost to get a user through the install door. It says nothing about:
- Whether that user activates
- Whether they return on Day 7 or Day 30
- Whether they convert to a paid tier, make a purchase, or complete a core action
- Whether they refer others
The right optimization target depends on your monetization model. For subscription apps, cost per trial or cost per subscriber is the number that matters. For transactional apps, cost per first order. For B2B tools, cost per qualified activation or cost per account.
If you're running campaigns without a clear downstream conversion event instrumented — meaning you can't trace from an install back to a revenue action — you're flying blind regardless of what your CPI is.
For a deeper look at how to structure your acquisition funnel to optimize beyond the install, see our post on 2026 mobile user acquisition strategy.
Platform Split: When It Actually Matters
There are situations where the iOS vs. Android CPI gap should drive a strategic decision:
If you're pre-revenue and budget-constrained, starting Android-first to prove out your acquisition funnel at lower CPIs is a legitimate strategy. The data you gather is transferable to iOS — you'll have creative performance signals, retention curves, and channel mix data before you pay the iOS premium.
If your LTV model is subscription-heavy, iOS is almost always worth the higher CPI. iOS subscribers have historically shown higher conversion rates from free to paid and lower churn. The CPI premium frequently pays for itself.
If you're targeting global markets outside Tier-1, Android dominance in those geos means iOS CPIs may be unrepresentatively low (smaller audience, less competition) while Android gives you real scale. Understand the market share split before assuming your platform CPIs will mirror US benchmarks.
The creative testing frameworks differ too. We covered building a systematic creative testing approach in detail in our post on building a 12-week creative testing pipeline for app install ads.
FAQ
What's a good CPI for a fitness app in 2026?
For a US-targeted iOS campaign, approximately $2.00–$5.50 is the typical range. Android runs closer to $1.00–$3.00. Whether that's "good" depends on your subscription price point and trial-to-paid conversion rate. A $5 CPI with a 25% trial conversion and a $14.99/month subscription is excellent. The same CPI with a 5% trial conversion is a problem.
Why is iOS CPI higher than Android?
Several reasons compound: the ATT framework limits third-party audience targeting on iOS, which increases reliance on contextual and keyword-based channels that carry higher CPMs. Apple Search Ads, the dominant quality channel for iOS, has seen sustained competition. And iOS users in Tier-1 markets represent a premium audience that advertisers actively compete for.
Should I use CPI or ROAS as my primary UA metric?
Use CPI for operational monitoring — it's the most immediate signal for campaign health. Use ROAS or LTV:CAC as your primary optimization target. If you don't yet have enough downstream data to calculate ROAS reliably, cost per activation or cost per day-7 retention event is a reasonable intermediate proxy.
How do CPI benchmarks differ for a B2B productivity app?
B2B and productivity apps typically run the highest CPIs in the market — approximately $4–$12 on iOS, $2–$7 on Android in the US. Audience size is smaller and competition for professional segments is intense. The key is that B2B LTV can easily justify a $10+ CPI if your annual contract value is meaningful. Don't benchmark against consumer app averages.
Which channels typically produce the lowest CPI?
Hypercasual and casual gaming often see the lowest absolute CPIs through broad-audience Meta and TikTok campaigns. For non-gaming apps, Google App Campaigns on Android typically deliver high volume at lower CPIs than ASA. That said, low-CPI channels don't always produce the highest-quality installs — evaluate by retention and conversion, not install cost alone.
How often should I update my CPI targets?
Quarterly is a reasonable cadence for reviewing benchmarks. Channel dynamics shift, seasonal pressure hits Q4 hard across nearly every category, and creative fatigue can gradually inflate your CPI without any platform change. Build a quarterly review of CPI trends by channel and creative into your growth ops rhythm.
If your CPIs are outside these ranges and you're not sure whether that's a channel problem, a creative problem, or a funnel instrumentation problem, that's a diagnostic conversation worth having. The Semnexus mobile app marketing team works with iOS and Android apps across paid UA, ASO, and growth strategy. Book a 30-minute call and we'll tell you what we're seeing.
