Thinking · Newsletter
App marketing agency: what it covers, when to hire one, and how to judge it on your own MMP numbers
What an app marketing agency covers (UA, creative, ASO, measurement), when to hire one vs in-house, and how to judge it on your own MMP numbers.
TL;DR. An app marketing agency runs some or all of four jobs for a mobile app: user acquisition (buying installs on networks such as Meta, AppLovin and TikTok), the ad creative that feeds it, app store optimisation (the store page that turns a visit into an install) and the measurement that ties them together. Judge one in your own MMP, on cost per install and on what each install cohort does afterwards (retention and revenue), not in the agency's slides. Hire one when you need creative supply and channel experience faster than you can hire them; keep it in-house when spend is steady and you can staff both buying and creative. This is a buyer's guide written by an agency that does part of this work, and we say which part.
What an app marketing agency covers, and who is writing this
"App marketing agency" covers very different shops. Some are user acquisition agencies that buy media, some are creative studios, some specialise in the store page, and some try to do all of it. Know which job you are hiring for before you compare.
| Job | What it means | What to judge it on |
|---|---|---|
| User acquisition | Buying installs on paid networks, moving budget to what works | CPI and cohort value in your MMP |
| Creative | The videos, UGC, playables and end cards the networks run | Which new ads take spend, and the cohorts they bring |
| App store optimisation (ASO) | Testing the icon, screenshots and preview video on the store page | Store conversion rate, read in Apple's and Google's own tools |
| Measurement | MMP setup, iOS privacy frameworks, incrementality tests | Whether you trust the numbers enough to move budget on them |
| Re-engagement | Ads aimed at lapsed users | Reactivated users, separated from new installs |
Flixr is a performance marketing agency for apps, games and commercial brands that makes its own creative. We run and read paid media on AppLovin, Meta and TikTok, and make the UGC and creator video, playables and end cards. So we cover user acquisition and creative (the showcase has examples such as Pull the Pin, a playable for Wildlife Studios); for store page work you may want an ASO specialist alongside whoever runs your media. We do not name or rank other agencies on this page (our list of user acquisition agencies for mobile games does that, with its method).
The store page is part of the funnel
Paid installs land on a store page, and both stores let you test it. Apple's product page optimization lets you "test different elements of your App Store product page", and "A test can include up to three treatments with alternate app icons, screenshots, and app previews." Apple shows "the number of impressions, conversion rate, percent improvement, and confidence level for each treatment." Google Play's store listing experiments let you "run experiments to find the most effective graphics and text for your app", with "up to 2 variants" against the current listing.
Our read: a mobile app marketing agency that buys installs but never asks about the store page is optimising half the funnel. If the agency does not do ASO itself, ask how it will work with whoever does, so the ad and the store page make the same promise.
Measurement: what your MMP can and cannot see
An app growth agency should start from your MMP, the independent tool that attributes installs across networks. On iOS, Apple's frameworks set limits. Apple says AdAttributionKit "helps advertisers measure the success of ad campaigns while helping maintain user privacy", and that "Apple determines a postback data tier for the conversion, and the device uses the tier later to determine the level of detail the postback can contain to help ensure crowd anonymity." Apple also says AdAttributionKit and SKAdNetwork "are frameworks that enable ad attribution and user engagement measurement for conversions."
In practice that means iOS results can can arrive with less detail than you may be used to. An agency that reports iOS performance at the same detail as Android without explaining how should be asked to show its working.
Acquisition and win-back: a read from Flixr Studio
App marketing is not only new installs. Flixr Studio, our own tool, reads public game ads from the Meta Ad Library. Of the 12,331 game ads it had read for who the ad is aimed at (23 game brands, ads launched between October 2024 and November 2026, nearly all since July 2026, watched since October 2026):
| Who the ad is aimed at | Share of read ads |
|---|---|
| Acquisition (new players) | 85.1% |
| Win-back (lapsed players) | 12.6% |
| Engagement or update | 2.2% |
| Unclear | 0.1% |
So roughly one in eight of these game ads speaks to players who have already left. This counts what game brands chose to run, not what worked. Our read: when you brief a user acquisition agency, ask whether win-back is in scope, and make sure those installs are reported separately from new users so they do not flatter your CPI.
When to hire an app marketing agency vs build in-house
| Situation | Our lean |
|---|---|
| New channel, no one in-house has run it | Agency, to borrow the experience |
| Creative is the bottleneck and hiring is slow | Agency or studio for creative supply |
| Steady spend, a stable team, one or two channels | In-house, with specialists for spikes |
| Store page converts poorly | An ASO specialist, before more paid spend |
| You cannot trust your numbers | Fix measurement first, whoever does it |
These are our rules of thumb, not industry data. Many teams end up with a mix: in-house ownership of the numbers and the budget, outside help for creative volume or a new channel.
How to judge an app marketing agency on your own MMP numbers
- Give the agency read access to your MMP and agree to read one dashboard. If you and the agency look at different numbers, every review becomes a debate about whose are right. Flixr starts every engagement with read-only access to your results and ad accounts for this reason.
- Set the baseline before day one. Record the last 90 days of CPI, early retention and revenue per cohort, by network and platform.
- Agree the cohort metric, not just CPI. Cheap installs that churn are not a win. Pick a retention rate day and a revenue or LTV checkpoint that matters for your app.
- Read iOS and Android separately. iOS postbacks carry less detail, so the comparison should be like for like.
- Track which new ads take spend. An agency's creative is working when the networks keep funding its new concepts; see how often to refresh ad creative.
- Test whether the spend added installs. A holdout or a bake-off tells you whether a channel brings new users or re-counts ones you had; see incrementality testing for ad creative.
Questions to ask, and red flags
Ask:
- Which of the five jobs above will you do, and who does the rest?
- Will campaigns run in our ad accounts, with you added as a partner?
- Which MMP dashboard will you be judged on, and which cohort metric?
- How many new creative concepts per month, in which formats (video, UGC, playable ads, end cards)?
- How will you report iOS?
Red flags:
- Reporting installs with no cohort data behind them.
- Ad accounts or MMP set up in the agency's name.
- Win-back or reactivated users mixed into new-install numbers.
- No plan for creative, or creative described as your job.
- Promised CPIs before anyone has seen your data.
How app marketing agencies are paid
The common models are a fee for the team's time, a fixed fee for defined deliverables, a percentage of media spend, and performance incentives on top of a fee. A percentage of spend rises as spend rises, and an incentive tied to installs can reward cheap users who leave, so tie any incentive to a cohort metric in your MMP. Flixr does not publish pricing; it is a conversation after we have read your account.
Questions people ask
What does an app marketing agency do? It runs some or all of user acquisition, ad creative, app store optimisation and measurement for a mobile app. Agencies differ a lot in which of those they actually do, so ask.
What is the difference between an app marketing agency and a user acquisition agency? A user acquisition agency focuses on buying installs on paid networks. An app marketing agency may also cover creative, the store page and re-engagement.
How do I measure an app marketing agency's results? In your own MMP, on CPI plus what each install cohort does afterwards (retention and revenue), set against a baseline you recorded before the agency started. Read iOS and Android separately.
Should a mobile app marketing agency also do ASO? It should at least work with whoever does, because paid installs land on the store page. Apple and Google both let you test store page elements with their own tools.
When should I hire an app growth agency instead of building in-house? When you need creative supply or experience on a new channel faster than you can hire for it. Build in-house when spend and channels are steady and you can staff both buying and creative.
Sources: Apple Developer, Product page optimization; Google Play Console Help, store listing experiments; Apple Developer, AdAttributionKit; Apple Developer, AdAttributionKit and SKAdNetwork interoperability; Flixr Studio, public Meta Ad Library game ads, read 9 Nov 2026; Flixr, About.