Which channels break even fastest?
№ 138 · Payback period by channel
Definition
MetricDays from install until cumulative cohort revenue covers acquisition cost
Unitdays to payback
Bar of days-to-CAC-payback per channel. Below 90 days typically healthy. Build by tracking cumulative contribution margin until it crosses CAC per channel. Channels above 180 days usually unsustainable. Example: Search 42 days, Meta 78, Affiliate 56, Display 124 (Display the marginal one).
- CAC
- Customer acquisition cost. Fully loaded sales and marketing spend in a period divided by new customers won in that period.
Benchmarks
| Bottom 30% | Median | Top 30% |
|---|---|---|
| over 270 days, or no crossing point inside the modelled horizon | 90 to 180 days | under 30 days |
Paid user acquisition cohorts in mobile free-to-play, 2024 to 2026. Payback is the first day on which cumulative cohort revenue equals acquisition cost. State whether the cost term is CPI, meaning media only, or CAC, meaning media plus creative production, tools, agency fees and team, because the two differ materially and CAC is the honest denominator. Revenue net of store fees unless stated. · Genre sets the band far more strongly than channel does, and no Tier 1 provider publishes payback by media source, so the supportable split here is by genre and platform rather than by channel. The genre ranges come from AppAgent's LTV > CPI Playbook and are corroborated in shape by Liftoff's platform ROAS split, but they are practitioner ranges rather than a measured distribution. Payback is also routinely quoted on CPI when it should be quoted on CAC, which understates the real window by weeks. The working capital consequence is the part most teams skip: at a 90-day effective payback, daily spend of $3,000 keeps roughly $270,000 permanently locked in cohorts that have not yet broken even.
| Category | Bottom 30% | Median | Top 30% |
|---|---|---|---|
| Hypercasual | beyond 7 days, at which point the model does not work | 2 to 7 days | 2 days |
| Hybrid-casual | beyond 60 days | 7 to 60 days | under 14 days |
| Casual | beyond 180 days | 60 to 180 days | around 60 days |
| Midcore and hardcore | beyond 730 days | 180 to 730 days | around 180 days |
| Casual on iOS | not published | D30 ROAS of 47%, so under half of cost is recovered by day 30 | not published |
| Casual on Android | not published | D30 ROAS of 15%, so under a sixth of cost is recovered by day 30 despite a much lower CPI | not published |
When it looks bad
Payback bars lengthen month over month across every channel at once while the cost bars sit flat, which means the revenue side of the ratio is eroding rather than the cost side rising.
Blended payback moving from 96 to 118 to 141 days across three consecutive cohorts with CPI steady at $1.60, and one channel showing no crossing point at all inside the 180-day window.
What to do about it
- Quote payback on CAC, not CPI. Adding creative production, tools, agency fees and team typically moves the window out by weeks, and every capital plan built on the CPI version understates how much cash is locked inside cohorts.
- Set the payback gate by genre rather than running one company-wide number. Hypercasual pays back in 2 to 7 days and midcore in 6 months to 2 years (AppAgent Playbook), so a single 90-day rule either starves the midcore title or over-funds the casual one.
- Check the genre's spend-to-revenue gap before blaming campaign execution. In the US in 2025, lifestyle and puzzle took 56.2% of gaming ad spend for 41.3% of IAP revenue while casino took 10.4% for 21.6% (Sensor Tower), so part of a long payback is auction structure rather than anything the UA team controls.
- Size the UA budget from the working capital the window actually locks up. At a 90-day effective payback, $3,000 of daily spend keeps roughly $270,000 permanently tied up in cohorts that have not broken even, and that money has to sit in the business alongside operating costs.
Sources
- AppAgent Typical payback times by genre: hypercasual 2 to 7 days, hybrid-casual 7 to 60 days, casual 2 to 6 months, midcore and hardcore 6 months to 2 years. At 105% D120 ROAS it takes $200,000 of ad spend to generate $10,000 of margin, against $16,700 at 160% ROAS. appagent.com ↗
- Liftoff and Singular Average D30 ROAS of 47% on iOS and 15% on Android for casual games. Sports highest on iOS at 80%, kids 68%, strategy 60%; RPG best on Android at 39%; tabletop lowest at 7%. Average casual CPI of $1.41 on iOS and $0.14 on Android. liftoff.ai ↗
- Liftoff Advertising-driven games generally need to show returns within 30 days, while IAP-driven games may take 90 to 180 days to hit their payback window, which can be acceptable if LTV projections support the spend. liftoff.ai ↗
- Adjust Global CPI rose 30% year on year to $0.56. Most expensive installs are slots at $4.47, idle RPG at $3.19 and strategy at $1.03. North America CPI $1.68, Europe $0.53, LATAM $0.14. Average CPM rose 20% to $4.34, with North America at $15.98. adjust.com ↗
- Liftoff and GameRefinery (via GameDev Reports) Average casual CPI of $2.17, with $0.65 on Android and $4.83 on iOS. Regional CPI of $6.78 in North America, $1.11 EMEA, $1.05 APAC, $0.44 LATAM. Average D7 ROAS of 5.7%. gamedevreports.substack.com ↗
- AppsFlyer IAP revenue reaches 60% of its D60 total by day 7 while advertising reaches 89%, which is the mechanical reason advertising-led titles pay back in weeks and IAP-led titles do not. appsflyer.com ↗
- Moloco The top 1% of iOS winning bid prices surged 140% in 2024 while median bid prices held steady, so payback on the highest-value user segment is lengthening even where the median cost line is stable. moloco.com ↗
- AppAgent A healthy LTV-to-CAC ratio is roughly 3:1. A 1:1 ratio is breakeven and defensible only as a short-term land grab, ratios above 5:1 suggest underinvestment in growth, and hypercasual frequently fails to clear 2:1. appagent.com ↗
- Sensor Tower (via AppAgent) In the US in 2025, lifestyle and puzzle drew 56.2% of all gaming ad spend while generating 41.3% of IAP revenue. Casino was the mirror image at 10.4% of spend for 21.6% of revenue. Action and strategy were roughly balanced at 30.7% against 34.8%. sensortower.com ↗
- Metricus A 120-day payback assumption has become a common planning default for iOS user acquisition under aggregated attribution, with incrementality testing used to validate it. metricusapp.com ↗