How do we test whether AppLovin sales are incremental?
Our answer
Run a geo holdout. Switch AppLovin off in a matched set of US states for 2 to 3 weeks at your planned spend, with budget and creative frozen. Compare total sales, Amazon and stores included, through about a week after.
- Confidence
- Medium
- Strongest proof
- 2 controlled tests
- 1 AppLovin rule, 4 datasets, 4 practitioners, 11 opinions or claims
- Sources
- 12 sources
- 22 statements
A
R
New here? Take the 1-minute tour
Why this is the answer
4 reasons and our judgmentAppLovin lets you hold out whole states. Its campaigns target US states, and it sizes its own geo holdouts at 20% to 50% of the footprint. WorkMagic, an incrementality testing firm, uses that targeting to switch off control states by hand.
Two to three weeks is the usual flight, with a short read after it. Jetfuel Agency, a consumer marketing agency, advises 14 to 21 day geo holdouts with spend and creative frozen. Ridge's two and three week reads can carry about a week after treatment.
Most of the return can land outside your site. Ridge's three-week test with Haus, an incrementality testing platform, found most of its return on Amazon.
Your own test matters more than published averages. Haus's AppLovin results swung during 2026. Its tests beat the quarter's median about half the time in the first half, then 79% beat the same brand's other channels from July to November 2026.
Our judgment
- Hold AppLovin out of a matched set of US states for 2 to 3 weeks, then read about a week past the flight for late sales. The week follows Ridge's one-week window after treatment. The flight combines the holdout lengths Jetfuel Agency and Common Thread Collective advise, Ridge's three-week test with Haus and WorkMagic's manual state exclusion.
- Size the held-out states at 20% to 50% of your US footprint. That takes AppLovin's own sizing and Ridge's 20% of US regions.
- Measure total sales across every channel and read new-customer orders separately. Ridge's Amazon lift carried most of its return, Haus found a halo beyond the website, and Jetfuel Agency notes existing customers can't be excluded from prospecting.
- Until a test is affordable, judge AppLovin on blended new-customer results, total sales divided by total ad spend (MER) and a post-purchase survey answer. This is consistent with Trusting AppLovin's reported results and follows AppLovin's use of MER between tests, Jetfuel Agency's survey option, Ridge's survey check and Pilothouse's new-customer cost per purchase check.
Our judgment, built from Jetfuel Agency, Common Thread Collective, AppLovin, WorkMagic, Haus, Ridge and Pilothouse Digital: our reading of these sources, not a finding from any one of them.
What the sources say
22 statements from 12 sourcesHaus, a testing vendor, says 79% of AppLovin tests since July beat the same brand's other channels, against under half in spring 2026. Since October, its CPMs ran about a quarter above summer, Meta's about 11%.
“Since July, 79% of AppLovin tests beat the same brand's other channels.”
AppLovin's web campaign API targets by country, plus US states or US metro areas; no app, age or audience field appears. Budgets and goals can be set per country.
“The country targets for the campaign”
AppLovin's own help pages say the gap between its dashboard ROAS and geo holdout results has more than doubled in a year, and tell advertisers to triangulate rather than trust one number.
“Click-based attribution misses non-click conversions”
Common Thread Collective, an ad agency, says about 20 AppLovin incrementality tests for its clients averaged about 150%, with only one or two below 100%. AppLovin sponsored the episode.
“We've had, I think, one or two tests out of the 20 we've run below 100%.”
Ridge CEO Sean Frank says AppLovin was Ridge's best channel for scale and results last Q4, measured as incremental, and Ridge plans to run it again.
“Northbeam, haus, post purchase, applovin was incremental. We will do it again this year.”
CTC, which manages AppLovin ads, says that for 7 to 9 figure brands already on Meta and Google, a 2 to 3 week geo test of AppLovin-active versus holdout regions is the most reliable read before scaling.
“AppLovin supports geo holdout tests and third-party MMP integrations for incrementality measurement.”
HexClad's Connor Rolain says AppLovin reaches the most new customers for HexClad: holdouts plus Northbeam showed 90% new customers, 36% higher ROAS and 27% lower CAC. The post carries an AppLovin credit offer.
“Which channel will help me reach the most new customers?”
Ridge's Connor MacDonald says it mostly runs two and three week tests on Meta with about a one week post-treatment window, and is starting a six week test on upper-funnel channels.
“we mostly run two and three week tests”
WorkMagic, which sells incrementality testing, explains a manual geo lift test on AppLovin: pair control and test regions, exclude states in campaign targeting, and pause to measure. AppLovin supports only manual setup.
“AppLovin only supports manual setup”
AdExchanger reports that AS Beauty's holdout tests found AppLovin traffic real and its campaigns working, while one CFO doubted the results and one marketer alleged bigger clickable ad areas. No holdout figures given.
“But the traffic was legit and the campaigns were working”
CTC's 18 AppLovin incrementality tests averaged 171%, against a 113.4% average for its Meta acquisition campaigns.
“Out of 18 AppLovin incrementality tests, only one came back below 100%. Five returned 225%+ incrementality. The average was 171%.”
Pilothouse's head of Meta suggests checking AppLovin with multi-touch attribution such as Triple Whale and your actual new customer CPA, since the platform may take credit for sales paid for elsewhere.
“What is my actual new customer CPA?”
Causality Engine, which sells attribution, says to check AppLovin cheapest first: survey and audience overlap, then session quality, then a 2 to 4 week holdout or geo split, stopping at the first failed check.
“Discount weeks contaminate holdouts; run clean windows.”
The Haus data comes from big spenders: tests averaged about $4.8K a day, at brands spending about $3M a month on digital ads. In 2026 AppLovin beat the quarter's typical test only 53% (Q1) and 50% (Q2) of the time.
“If AppLovin is under 5% of your spend, you should still be validating whether it is worth continuing to invest in.”
AppLovin says brands' holdout-measured iROAS relative to its click-only reported ROAS has more than doubled in a year, giving no test count. Its own holdouts cover 20 to 50% of geos at 90% power.
“MER, total revenue divided by total marketing spend, is the simplest of all of them”
Jetfuel, an agency that audits AppLovin accounts, tests AppLovin with a 14 to 21 day geo holdout against a synthetic control, holding spend steady, and says an inconclusive result is not a negative one.
“an inconclusive result is not the same as a negative one”
Jetfuel, an agency, would keep brands with a buy cycle over a week, no product-market fit on Meta, or no measurement tool like Triple Whale or Northbeam off AppLovin, and wants $1,000 a day while learning.
“a 14 to 21 day geo holdout to confirm incrementality at your real spend level.”
Jetfuel, which audits AppLovin accounts, says the dashboard, an MMP and a geo holdout often give three different ROAS figures, and a channel can look incremental on dashboards yet test as zero in a holdout.
“at lower spend levels a channel can look incremental on every dashboard and still test as zero under an actual holdout.”
Elevarus, an agency, says that above about $80K a month per channel you should run 4 to 6 week geo holdouts for incremental ROAS; below about $40K, use Meta's lift tool or skip formal testing.
“Use a geo holdout when you need to test cross-channel effects, when offline conversions are part of the outcome, or when the platform-native option doesn’t cover the campaign type.”
Sets a lower floor than Sellforte's 30% to 50%: Elevarus says a holdout under 15% of regions is useless, and below $40K a month on a channel, use platform lift tests or skip testing.
“A 5% holdout on a national account produces a control group too small to compare against anything.”
CTC's Connor McLaughlin says AppLovin results vary widely, so each brand should run its own test, suggesting $5K of spend plus $5K of AppLovin credit. CTC is an AppLovin partner and offers a $5K spend match.
In AppLovin's case study, Ridge ran a three-week Haus geo holdout of 20% of US regions. Amazon revenue rose 14.61%, an incremental ROAS of $1.24 against a $0.80 target, and Ridge then nearly quadrupled spend.
“The lift was visible across both sales channels, with Amazon driving the majority of incremental value.”
In the full answer
For readers who sign in- What to do this week3 steps
- The numbers behind it3 figures from cited sources
- Where each source standsA chart placing each source on the question
- The strongest disagreementThe best case against the answer, from a cited source
- How sure we areWhat its confidence rests on, and what would change the answer