Thomas Petit
Independent consultant (since ~2014) specialising in subscription apps — from indie developers to very large apps — on acquisition, monetisation and the data behind them (attribution, product analytics). Active in the Mobile Dev Memo Slack, X and LinkedIn; doesn’t take new clients easily.
Strong on: practical UA audits, conversion-signal setup for Meta/Google, combining last-click with incrementality, web2app funnel metrics, chargebacks.
Bias to keep in mind: a hands-on practitioner’s anecdotes (his client portfolio), mostly subscription apps rather than games; numbers are rules of thumb, not studies.
Sources: Mobile Dev Memo podcast (sources/youtube/mobile-dev-memo/).
Notes citing Thomas Petit
- Apple Ads account structure — the two starter campaignsThe two-campaign Apple Ads setup (Max Conversion discovery + single-keyword exact campaigns) App Masters runs at $7K–$85K/month, and how they work together.
- Apple Ads audience settings — customer types, demographics, location and ad schedulingWhat Apple Ads audience options actually do — "new users" vs the ambiguous "returning users", why age/gender targeting reaches only users with personalised ads on (Thomas Petit estimated ~5% effective reach in 2023), when location targeting and ad scheduling to fill organic-rank dips make sense, and where practitioners disagree.
- Apple Ads brand vs non-brand — split campaigns by intent and never blend the resultsWhy 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.
- Apple Ads match types, search match and negative keywords — check your search termsWhy Apple Ads broad match and search match spend on irrelevant queries (Shazam → a TV guide; "diet app" → DIY and digit apps; "fitness" → Apple Fitness+), why new keywords default to broad, the hours-long delay when pausing, and the exact-match + negative-keyword + search-terms-report routine to prevent it.
- Apple Ads measurement pitfalls — AdServices isolation, match rate, storefront country, CPP comparisonsThomas Petit's list of ways Apple Ads numbers mislead — AdServices attribution can't be compared with SKAN-measured channels, MMP-vs-Apple install "match rate" (80–90% is good), App Store storefront ≠ user location (up to ~30% non-US installs on US campaigns), and custom product page results that look 10× better only because they serve a different audience.
- Apple Ads single keyword ad groups (SKAG)Why and how to run Apple Ads SKAG campaigns — one exact-match keyword per ad group, search match off, bid near target CPA, 5–10 keywords.
- Do you need an MMP? AppsFlyer, Adjust and the state of SKAdNetwork (2026)Why mobile measurement partners (AppsFlyer, Adjust, Branch…) remain mandatory for app advertisers buying Meta at scale, what the 2026 AppsFlyer consortium investment means for neutrality, and why SKAdNetwork/AdAttributionKit is not worth heavy investment.
- Evaluating a marketing mix model (and whether Robyn/Meridian are biased)Assume any MMM is wrong until it passes three checks — parameter recovery on simulated data, out-of-sample forecasting of a period with real budget changes, and agreement with lift tests; open-source Robyn and Meridian are not rigged for Meta/Google, but most MMM vendors are thin wrappers around them.
- How far does Apple Ads scale? Channel share, ceilings and what limits youRealistic expectations for Apple Ads volume — Thomas Petit's view that it usually ends up 5–20% of a scaled app's marketing mix (some accounts still reach seven figures/month), that return targets rather than search volume usually cap spend, and the opposing practitioner views on whether Apple Ads is a first channel or a "nice to have".
- How last-click attribution biases channel resultsLast-click flatters low-funnel, intent-capturing channels (search, Apple Ads, Google UAC standard) and under-credits view-based ones (YouTube, CTV, chatbot ads); incrementality tests have reversed channel rankings — and platform-reported numbers are biased toward the platform.
- How mobile UA changed (2015 → 2026) — the remaining levers, team structure and channel mix by budgetFrom 100-person manual media-buying teams (Machine Zone era) to algorithmic buying on 4–5 consolidated networks — what's left to control (bid, campaign type, budget, creative, signal), how a modern UA team is structured, and how channels and process change from six- to eight-figure monthly budgets (gaming lens).
- How the Apple Ads auction behaves — bid probing, bidding down and avoiding a competitor's "crown"Practical Apple Ads bidding lessons — the auction is not a clean second-price auction (raising bids can add cost but no impressions), high tap-through rate lets you cut bids up to 90% and hold position, probing bids reveals a competitor's ceiling, and keywords where one rival owns ~33% share of impressions are best avoided by small budgets.
- How to set up an Apple Ads Max Conversion campaignStep-by-step settings for the Apple Ads Max Conversion (target CPA) discovery campaign — CPA goal, daily budget, extra ad group, audience, CPP.
- Measuring Apple Ads revenue without an MMPLightweight ways to attribute Apple Ads revenue — custom product pages per ad group, RevenueCat attribution, SKAG, or simple weekly spend-vs-revenue — and their blind spots.
- Optimising Apple Ads keywords for revenue, not installsWhy install- or CPA-optimised Apple Ads quietly funds keywords that never pay, how to judge each keyword on trials, cost per paid and predicted LTV (≥20 installs for a first signal, ~100 for conversion rates), what MMP fees cost per payer (~$1.40–3.50), automation rules tied to trials/subscriptions, and Adapty's claim of accounts scaled to ~$60K/month at ≥90% D31 ROAS.
- Should we add a new ad channel? Scale, return, effort — and small platform syndromeHow to decide whether a new or smaller ad channel (CTV, AppLovin for e-commerce, chatbot ads, retail media) is worth testing — judge scale, achievable ROAS and team effort, require it to beat (not match) incumbents, and budget for a real test plus data integration.
- Signal engineering — auditing and shaping the conversion events ad platforms optimise onFirst check what conversion data actually reaches Meta/Google (duplicates, wrong optimisation event), keep "what the platform needs" separate from "what the business wants", and engineer enough purchase events (e.g. ~20/day per campaign in gaming) for campaigns to exit learning.
- The measurement stack — last-click attribution, incrementality and MMM togetherWhy app advertisers need both deterministic last-click attribution (for platform optimisation and daily decisions) and incrementality testing (for budget allocation), where MMM fits, and how to reconcile them against top-line results.
- Which countries to target with app ads — high-RPD bundles and cheap-but-paying geosCountry choices for subscription-app ads in 2025–26 — Samet Durgun's single Meta ad set covering high revenue-per-download countries with the language set to English, Adapty's case for testing cheaper geos (South Africa, Brazil, Romania, Poland) on Apple Ads, and the App Store storefront vs IP-country mismatch.
- Why start paid acquisition with Apple AdsWhy App Masters starts most iOS apps on Apple Ads (plus ASO) before Meta or TikTok, and when that advice does not apply.