---
title: Is paid acquisition incremental? Organic ceilings, cannibalization and geo holdouts
slug: paid-vs-organic-cannibalization
summary: Organic-heavy apps (Replika, Solid Starts) eventually saturate their niche and find that new paid spend often cannibalizes organic; Simply measures this with geo holdouts (switch app or web campaigns off per country) and aims for "enough truth" in CAC models, not perfect attribution.
domain: mobile-apps
tags: [user-acquisition, attribution, incrementality, organic, paid-social]
related: [organic-ugc-vs-paid-ads-margins, apple-ads-attribution-without-mmp, why-web2app-funnels, built-with-science-meta-scaling-case, measurement-stack-attribution-incrementality-mmm, paid-to-organic-uplift]
sources: [yt:jzlPf100vy4, yt:Xv04J1cr16Q, yt:EyoYQU6JyV8]
updated: 2026-10-24
---

# Is paid acquisition incremental?

## TL;DR
- Big organic is a double-edged sword: once your niche's followers have all joined, organic flattens — and paid spend added later **cannibalizes organic about half the time** (Yuliya Lennox, Replika & Solid Starts).[^1]
- So **start performance marketing alongside the brand from day one**; ads also grow organic reach (Eva Learn English's ~7M Instagram grew partly from ad-driven followers).[^2]
- Measure incrementality with **geo holdouts**: Simply turned off app campaigns in one geo and web campaigns in another, then watched organic, cross-app and cross-platform effects.[^3]
- The goal is not perfect attribution but **"enough truth" in CAC modelling to know where the next dollar goes** (Yoav Sharon).[^4]
- With heavy organic, isolate paid conversions: Built With Science built a custom dashboard to validate each trial truly came from Meta before feeding high-intent users back.[^5]

## Details
**The organic trap (Yuliya).** Organic feels great — revenue without spend — until it stagnates. You then can't show growth to buyers/investors, and switching on paid largely buys users who would have come anyway. You can't steer organic day-to-day; performance is controllable. Organic only scales if you offer something genuinely unique (Replika early; a time tracker or cleaner app won't get it).[^1][^6][^7]

**Simply's test (mid-2025).** Multi-app portfolio, iOS + Android, app campaigns + web2app campaigns. By switching channel types off per geo they could "crystallize" cannibalization between apps, platforms and channels. They also check that their own apps don't compete for the same audiences.[^8][^3]

**Practical holdout recipe (derived):** pick comparable geos → turn one channel off in one geo for a fixed period → compare total installs/revenue (not attributed) vs control → repeat per channel.

## Disagreements / caveats
- Yuliya's "50%" is a rough anecdotal figure, not a measured lift.
- Geo holdouts need enough volume per geo and stable seasonality; small apps can use simpler on/off time tests (see [[apple-ads-attribution-without-mmp]]).

## See also
- [[organic-ugc-vs-paid-ads-margins]]
- [[built-with-science-meta-scaling-case]] — paid on top of a 7M-subscriber audience
- [[people/yuliya-lennox]], [[people/yoav-sharon]]

[^1]: yt:jzlPf100vy4 @ 47:52 — Yuliya Lennox
[^2]: yt:jzlPf100vy4 @ 49:23 — Yuliya Lennox
[^3]: yt:Xv04J1cr16Q @ 11:12 — Yoav Sharon
[^4]: yt:Xv04J1cr16Q @ 11:58 — Yoav Sharon
[^5]: yt:EyoYQU6JyV8 @ 23:06 — Ethan Ethier
[^6]: yt:jzlPf100vy4 @ 46:21 — Yuliya Lennox
[^7]: yt:jzlPf100vy4 @ 50:56 — Yuliya Lennox
[^8]: yt:Xv04J1cr16Q @ 10:23 — Yoav Sharon