--- title: How much organic does paid UA create? Measuring the paid-to-organic uplift slug: paid-to-organic-uplift summary: The routes by which paid spend lifts organic installs (rankings, browse/similar-apps, brand search), Playtika's regression-then-scale method for estimating organic installs per paid install (Brazil rewarded-video case 0.1 organic per paid via browse; France native-ads case ~0.4 via brand search), and why the answer depends on your organic mix. domain: mobile-apps tags: [aso, user-acquisition, organic, k-factor, regression, measurement, browse-traffic] related: [paid-vs-organic-cannibalization, aso-roi-measurement, aso-ranking-factors, aso-as-acquisition-channel, aso-download-burst-after-metadata-update, incrementality-test-design] sources: [yt:yi5GhHJ8eEo, yt:JtvOFVzxVLM, yt:3nas91cJjDE] updated: 2026-10-24 --- # How much organic does paid UA create? Measuring the paid-to-organic uplift ## TL;DR - Paid UA lifts organic through **(1) the store algorithm** (more first-time downloads → better search ranks and browse placements such as "similar apps"), **(2) brand awareness → brand search**, and **(3) virality**; brand/TV activity drives brand installs directly.[^1][^2][^3] - Tilting Point's 2021 correlation study across all store parameters found **UA had the strongest correlation with organic**, on both stores.[^4] - Playtika's method: pick a geo/OS with high traffic, a strong existing paid→organic K-factor and good user quality → **regress organic on each paid source** → scale the winning source **≥50%** and check the prediction → scale further.[^5][^6][^7] - Case (Bingo Blitz, Brazil, iOS): rewarded-video ads in other games were the winning source; the forecast of **0.1 organic per paid install** (10 paid → 1 organic) held over a ~2-week test and again at higher scale, arriving via browse/similar apps.[^8][^9] - Case (House of Fun, France, Android): native ads, forecast **0.5** organic per paid, actual **~0.4**, arriving as brand search.[^23][^25] - **It depends on your mix:** brand-heavy organic falls when you cut paid; Zynga apps that are ~85% organic from top generic ranks don't move when UA is switched on or off.[^10][^21] - Ramp paid **down gradually** — a sudden stop after heavy spend can drag organic down with it.[^12] ## Details **Check the data first.** Talia Hermesh (Playtika, ASO Conference 2023) first confirms that the paid–organic relationship seen in the MMP (AppsFlyer) also shows in the store consoles, and cross-checks with AppTweak, data.ai and MobileAction.[^13][^14] **Reading the regression.** The coefficient (beta) for a source is the number of organic installs per paid install: beta 0.5 means two paid installs bring one organic.[^15] Choose what you want to predict up front, and treat the regression output as a hypothesis to test with real spend.[^16] **Where the Brazil uplift came from.** App Store Connect showed the extra organic arriving as **browse**, and AppTweak's similar-apps tracking showed Bingo Blitz gaining similar-apps placements in the same two spikes as the paid phases — the rewarded-video traffic "pinged the algorithm".[^17][^18] **Second case (House of Fun, France, Android).** Chosen again for a strong existing K-factor; the regression winner was **native ads** (ads beside article content, e.g. a solitaire game under an article on the benefits of solitaire).[^19][^22] Playtika bid higher on the strength of a forecast of **0.5 organic per paid install**; the actual uplift was **~0.4** — short of the prediction but still large.[^23] This time Google Play Console showed the extra organic as **search**, and specifically **brand** search for "House of Fun" spiking on the test dates — native ads worked through brand awareness rather than the algorithm.[^24][^25] So the route differs by source: rewarded video → browse/similar apps; native ads → brand search. **Running it as a programme.** Put these tests into your KPIs and budget plan rather than testing ad hoc; scale down gradually; and a small budget can still do it if spent on the right geo and source.[^12][^26] Talia's closing caveat: a source that works in one geo, platform or season may not work in another.[^20] **Why channel mix matters (Shoham Lashkar, Zynga, 2023).** YouTube pre-roll viewers (e.g. via Google UAC) rarely install on the spot but may search for you later, outside the attribution window — cutting that spend shows up as an organic drop.[^11] ## Disagreements / caveats - Correlation and regression on observational data can't fully separate causation from seasonality and other launches; the scale-up phase is what validates it. For holdout-style proof see [[paid-vs-organic-cannibalization]] and [[incrementality-test-design]]. - The paid→organic halo is the mirror image of cannibalization: branded Apple Ads may buy users who would have come anyway (see [[aso-as-acquisition-channel]]). - Gaming examples (Playtika, Zynga, Tilting Point); subscription apps with smaller budgets may see much weaker browse effects. ## See also - [[aso-ranking-factors]] — download volume as a ranking input - [[aso-roi-measurement]] [^1]: yt:yi5GhHJ8eEo @ 09:40 — Talia Hermesh [^2]: yt:yi5GhHJ8eEo @ 10:25 — Talia Hermesh [^3]: yt:3nas91cJjDE @ 30:14 — Jonathan Fishman [^4]: yt:3nas91cJjDE @ 30:14 — Jevgen Tarasenko [^5]: yt:yi5GhHJ8eEo @ 16:22 — Talia Hermesh [^6]: yt:yi5GhHJ8eEo @ 17:13 — Talia Hermesh [^7]: yt:yi5GhHJ8eEo @ 17:58 — Talia Hermesh [^8]: yt:yi5GhHJ8eEo @ 18:45 — Talia Hermesh [^9]: yt:yi5GhHJ8eEo @ 19:30 — Talia Hermesh [^10]: yt:JtvOFVzxVLM @ 58:48 — Shoham Lashkar [^11]: yt:JtvOFVzxVLM @ 59:34 — Shoham Lashkar [^12]: yt:yi5GhHJ8eEo @ 25:32 — Talia Hermesh [^13]: yt:yi5GhHJ8eEo @ 08:48 — Talia Hermesh [^14]: yt:yi5GhHJ8eEo @ 07:14 — Talia Hermesh [^15]: yt:yi5GhHJ8eEo @ 15:36 — Talia Hermesh [^16]: yt:yi5GhHJ8eEo @ 26:28 — Talia Hermesh [^17]: yt:yi5GhHJ8eEo @ 20:20 — Talia Hermesh [^18]: yt:yi5GhHJ8eEo @ 21:05 — Talia Hermesh [^19]: yt:yi5GhHJ8eEo @ 21:57 — Talia Hermesh [^20]: yt:yi5GhHJ8eEo @ 27:14 — Talia Hermesh [^21]: yt:JtvOFVzxVLM @ 1:00:25 — Shoham Lashkar [^22]: yt:yi5GhHJ8eEo @ 22:44 — Talia Hermesh [^23]: yt:yi5GhHJ8eEo @ 23:34 — Talia Hermesh [^24]: yt:yi5GhHJ8eEo @ 23:34 — Talia Hermesh [^25]: yt:yi5GhHJ8eEo @ 24:27 — Talia Hermesh [^26]: yt:yi5GhHJ8eEo @ 26:04 — Talia Hermesh