---
title: Apple Ads brand vs non-brand — split campaigns by intent and never blend the results
slug: apple-ads-brand-vs-non-brand
summary: Why Apple Ads should be read as two channels (brand and non-brand) — Thomas Petit's 500% vs 80% ROAS example hidden inside a "200% blended" account — plus Adapty's four intent campaigns (brand, competitor, category, discovery), defending your brand term, and pause tests for brand cannibalisation.
domain: mobile-apps
tags: [user-acquisition, apple-ads, brand-keywords, campaign-structure, cannibalization, benchmarks]
related: [apple-ads-campaign-structure, apple-ads-auction-and-bid-tactics, paid-vs-organic-cannibalization, apple-ads-scale-ceiling, apple-ads-keyword-revenue-optimization, last-click-attribution-biases, aso-as-acquisition-channel, incrementality-test-design]
sources: [yt:EYfeV7fkCeI, yt:ZRVB9O3vOX4]
updated: 2026-10-24
---

# Apple Ads brand vs non-brand

## TL;DR
- Thomas Petit calls this his most important Apple Ads point: treat **brand (B)** and **non-brand (NB)** as two channels, and always report them separately.[^1]
- A "200% ROAS, on target" account was, split out, **brand ~500% and non-brand ~80%**. Blending hides that non-brand is losing money. Petit calls blending "an agency trick".[^1][^2]
- Separate campaigns by intent: **brand**, **competitor**, **category/generic** and **discovery**. Then you can see which actually generates revenue (Adil Bilgi, Adapty).[^3]
- Search your own brand name in the App Store today. If a competitor holds position #1, set up a brand campaign now. It's usually the cheapest CPA in the account.[^4]
- Measure cannibalisation with **on/off pause tests**, run separately for brand and non-brand. Brand cannibalises far more.[^5]
- Ignore benchmarks that don't split brand/non-brand. That's nearly all of them.[^6]

## Details
**Why brand behaves differently.** On your own brand term you usually hold the top organic slot and people know what they want. Ads there buy installs that would largely have come anyway. ROAS looks spectacular but is mostly not incremental.[^5] There are still good reasons to bid on brand: competitors conquesting it, and the data you get.[^2][^7]

**Meta and Apple Ads feed each other (Adapty, June 2026).** Mykola Martynovets frames Meta as creating demand and Apple Ads as capturing it: Meta-driven users go to the App Store and type your brand. That is why Adil Bilgi defends the brand term. A competitor bidding on it captures awareness you paid for. Brand taps are cheap because your TTR and relevance on your own name are the highest and competition is low.[^11][^12] Mykola adds that in his experience **competitor keywords** can perform even better than generic/category ones.[^13]

**Cheap brand defence.** On a brand campaign with ~90% share of voice, Petit cut bids by 90% and kept the same volume (see [[apple-ads-auction-and-bid-tactics]]).[^8]

**Pause test method (Petit).** Periodically pause a campaign and watch total installs, not just paid. The drop is the incremental part. You may accept some cannibalisation, but you should know whether it's ~10% or ~50%.[^5]

**Brand traffic can carry higher prices.** Adapty clients tested higher-priced paywalls for brand-intent users who already know the app (see [[paywall-by-acquisition-source]]).[^9]

**Benchmarks.** Petit sees most published Apple Ads benchmarks as clickbait. They lack segmentation by vertical, country and size, and blend brand with non-brand. He credits SplitMetrics with doing it as well as anyone, but still says to use them with caution.[^6]

## Disagreements / caveats
- Steve P. Young's starter structure (Max Conversion + SKAG) doesn't split by intent ([[apple-ads-campaign-structure]]). Adding a separate brand campaign is compatible with it and makes ROAS readable.
- Adapty's four-campaign split comes from a vendor webinar; Petit is independent ("nothing to sell") but speaks from 2023 accounts.[^10]

## See also
- [[paid-vs-organic-cannibalization]] — geo holdouts for the same question across channels
- [[last-click-attribution-biases]] — why search channels look better than they are
- [[people/thomas-petit]], [[people/adil-bilgi]], [[people/mykola-martynovets]]

[^1]: yt:EYfeV7fkCeI @ 21:26 — Thomas Petit
[^2]: yt:EYfeV7fkCeI @ 22:48 — Thomas Petit
[^3]: yt:ZRVB9O3vOX4 @ 07:15 — Adil Bilgi
[^4]: yt:ZRVB9O3vOX4 @ 45:30 — Adil Bilgi
[^5]: yt:EYfeV7fkCeI @ 31:53 — Thomas Petit
[^6]: yt:EYfeV7fkCeI @ 22:48 — Thomas Petit
[^7]: yt:EYfeV7fkCeI @ 33:23 — Thomas Petit
[^8]: yt:EYfeV7fkCeI @ 24:19 — Thomas Petit
[^9]: yt:ZRVB9O3vOX4 @ 42:14 — Mykola Martynovets
[^10]: yt:EYfeV7fkCeI @ 00:00 — Thomas Petit
[^11]: yt:ZRVB9O3vOX4 @ 11:20 — Mykola Martynovets
[^12]: yt:ZRVB9O3vOX4 @ 12:58 — Adil Bilgi
[^13]: yt:ZRVB9O3vOX4 @ 13:54 — Mykola Martynovets