The best user acquisition agency for a mobile app depends on which half of the job you are hiring for. If you need breadth across paid channels and the plumbing that measures them, REPLUG publishes the widest channel list in this group. If you need an agency that will both buy the media and show that the spend caused the revenue, Admiral Media is the only one here that publishes a working incrementality method and offers to run it. If your app is a subscription product and the real constraint is Meta creative rather than the number of channels you are live on, that is the narrow job we do, and it is exactly why we rank sixth on this page.
Every agency in this category publishes a list of platforms it can buy on. Almost none of them publish anything meaningful about how they would tell you whether the buying worked. That gap is the most decision-relevant thing in the category and no competing list for this query leads with it, so it is the axis this comparison is ordered on.
Where our interest lies. The Social Outline is one of the eight companies on this page, so treat this as a comparison written by a competitor rather than a neutral audit. Every claim below comes from the company's own live website, checked on 18 September 2026, and we say when something is not published rather than filling the gap. We place sixth, because this page ranks on channel breadth and measurement depth and we deliberately sell neither. Two of the ranked lists competing for this query were published by agencies that put themselves first without disclosing it. We would rather tell you where we are the wrong call.
The shortlist at a glance
Eight companies, compared on what each one is, the channels it names on its own site, the measurement it publishes, whether it maintains a subscription practice, and whether it publishes a price. The measurement column is the one worth reading twice.
| Company | What it is | Channels it names | Measurement it publishes | Subscription practice | Published pricing |
|---|---|---|---|---|---|
| Admiral Media | Paid media and creative agency | Meta, Google, TikTok, Snapchat, Apple Ads, YouTube, Reddit, Telegram, ChatGPT Ads | Incrementality framework published, offers to run it | Dedicated subscription app practice | €363 per AI video ad; media fees not disclosed |
| REPLUG | Multi-channel UA specialist | Apple and Google search, Meta, TikTok, Snap, OEM, DSPs, SDK networks, rewarded, affiliate | MMP audit and setup, SKAN 4.0 audit | None published | Not disclosed |
| Miri Growth | Games and apps UA agency | Meta, TikTok, Google UAC, Apple Ads, Snap, CTV, Moloco, Liftoff | MMP management, signal optimisation | One named subscription case study | Not disclosed |
| Phiture by Precis | Growth consultancy | Not specified on the site | Data and tech practice, named MMP and analytics partners | Published subscription framework | Not disclosed |
| SplitMetrics | Software first, managed services second | Apple Ads depth, claims 15+ UA channels via services | Combines Apple Ads and MMP post-install data in one platform | None published | Not disclosed |
| The Social Outline | Meta and performance creative agency | Meta, with TikTok and YouTube where the economics justify it | MMP and SKAdNetwork setup, cohort ROAS reporting | Consumer subscription apps are the core business | From £1,500/month, published |
| Moburst | Full-service group | Social, search, networks and RTBs, influencer, email, OTT | None published | None published | Not disclosed |
| Yodel Mobile | Store-led app agency | User acquisition offered, platforms not specified | None published | None published | Not disclosed |
How this list is ordered, and what it deliberately ignores
Ordering is by the buying function and the measurement behind it. That means four things: the range of inventory an agency can genuinely buy, the depth of measurement it publishes, whether it demonstrates that it understands subscription economics rather than install volume, and whether its business type matches what a consumer app actually needs. Pricing transparency is used as a tiebreaker, not a ranking factor.
Two capabilities are deliberately excluded. App store optimisation and lifecycle messaging are real disciplines and they decide a lot of app outcomes, but they are not user acquisition, and ranking an app store consultancy against a media buyer produces a list where nothing is comparable. Those sit in our wider comparison of app marketing agencies, where several of the companies below place very differently.
This also means the order here is not the order elsewhere. Ranked on Meta specifically, the list looks like our Meta agency comparison. Ranked on creative production capability, it looks like our performance creative comparison. A company near the bottom of this page can be the right hire on a different axis, and we flag that where it applies.
What a user acquisition agency actually does day to day
User acquisition is the buying function: channel mix, campaign structure, the events campaigns optimise toward, bids and budgets through the week, creative fed into tests, and the reporting that says whether any of it worked. It is a narrower job than most agency websites in this category imply.
The part that separates competent from expensive is the choice of optimisation event, which is the highest-leverage decision in a subscription app's ad account because it determines which users the platform goes and finds. Optimise toward installs and you get people who install. Optimise toward trials or subscriptions and the signal is scarcer, slower and far more valuable. We have written about which of those numbers to steer on and when and how to compute subscriber acquisition cost correctly, and an agency that cannot hold a conversation about that trade-off is selling account admin.
The measurement ladder: three rungs that all get called measurement
Ask five agencies in this category about measurement and you will get five answers that sound similar and describe three different services. Separating them is the most useful thing a buyer can do in a pitch, because the word covers everything from configuring a tracking SDK to proving that a channel produced revenue that would not otherwise have existed.
| Rung | Question it answers | What the work is | Who publishes it |
|---|---|---|---|
| 1. Plumbing | Is the data arriving and is it labelled correctly? | MMP selection, setup and audit. SKAdNetwork or AdAttributionKit conversion value mapping. Event taxonomy. Deduplication between sources. | REPLUG names this as an explicit service. Miri Growth publishes MMP management and signal optimisation. Phiture names its measurement and analytics partners. |
| 2. Cohort reporting | What did this cohort of users actually do, in our numbers? | Reporting cost per trial, cost per subscriber and cohort return from your own revenue data rather than platform-reported conversions, with the measurement window stated. | SplitMetrics combines Apple Ads and MMP post-install data in one view. The Social Outline publishes cohort-level return reporting. Most agencies in this category publish nothing specific here. |
| 3. Causality | Would this revenue have happened anyway? | Baseline holdouts, geo lift tests, platform lift studies, or a permanent small holdout that keeps recalibrating the rest of the reporting. | Admiral Media is the only agency in this group that publishes a named method for each of these and offers to run them for clients. |
Rung one is table stakes and it is genuinely necessary. If conversion values are mapped badly, everything above it is decorative. Our piece on how long postbacks take to arrive covers why this layer is harder than it looks on iOS.
Rung two is the real floor, and it is where most of this category quietly stops. Reading outcomes from your own revenue data rather than from platform-reported conversions is not exotic. It is the difference between an agency reporting what Meta says it did and an agency reporting what your subscribers did.
Rung three is where the honest caveat belongs, because it is the rung most often used as a stick. Incrementality testing requires enough volume for a holdout to produce a readable result. An app spending a few thousand pounds a month cannot run a geo lift test and learn anything, and an agency that promises one at that spend is either overselling or planning to show you a result with no statistical power. Rung three becomes a fair expectation as budgets scale into the range where a few percent of wasted spend is a real number. Below that, insisting on it is theatre, and rung two done properly is worth far more.
Why an agency that reports on installs is telling you the least
There is a sourced test you can apply to any agency's reporting and it takes one question: what window do you report on, and what share of the outcome has landed by then?
RevenueCat's State of Subscription Apps 2026, which covers more than 115,000 apps, finds that 50.6% of paid conversions happen on install day and 19.2% arrive in week six or later. Both figures are a share of the conversions that eventually happen, not a share of installs, which matters because the median download-to-paid rate within 35 days of install is 2.0%. Read the 50.6% as a share of installs and you would conclude something twenty-five times too optimistic.
The consequence for hiring an agency is direct. A subscription app judged on a seven day window is being judged on roughly half of what the cohort will eventually do. An agency reporting installs and cost per install is reporting the fastest and least informative number in the funnel, and it will look competent for about two quarters before the revenue fails to arrive. An agency reporting cohort outcomes at 30 and 60 days looks worse early and is telling you considerably more. Our guide to reading day 7 return on ad spend works through what an early number can and cannot support.
The eight agencies
1. Admiral Media
Founded in 2019, remote across more than 15 countries, and publishing €500M+ in managed ad spend across 150+ brands. It names nine buying platforms including Meta, Google, TikTok, Apple Ads, Snapchat, YouTube, Reddit, Telegram and ChatGPT Ads, which is the broadest named platform list here after REPLUG.
It ranks first because it is the only agency in this group that publishes a practical incrementality method, naming baseline holdouts, geo lift testing, platform lift studies and always-on holdouts, and then offers to run them. It also maintains a dedicated subscription app practice that treats trial and subscription events as the bidding target and works to a 90 day lifetime value model rather than an install target. That combination, rung three measurement plus subscription economics, is not published by anyone else here.
One caveat worth stating, since this page is partly about how these lists are built: Admiral also publishes its own ranked list of user acquisition agencies and places itself first on it. That is not why it places first here, and its own ranking should not be read as corroboration of this one. It ranks first because its published capability leads on the criteria set out above, and declining to rank a company highly because it also rates itself highly would be its own kind of bias.
Best for: subscription apps with enough spend to make incrementality testing meaningful, that want buying and measurement from the same team.
Wrong choice for: teams whose constraint is a deep single-channel creative engine, or small accounts where nine channels means attention spread thin across campaigns that cannot each carry enough budget to learn.
2. REPLUG
The broadest genuine inventory coverage on this page. Alongside Apple and Google search and the major paid social platforms, it names OEM advertising, demand-side platforms, SDK networks, rewarded traffic and affiliate, which are the channels apps reach for once Meta and Google stop absorbing budget efficiently. Named clients include Bolt, FREE NOW, Ecosia, Satispay, Bild and Le Figaro.
On measurement it sits firmly at rung one, and unusually it says so plainly: mobile measurement partner audit and setup, and a SKAN 4.0 audit, are listed as explicit services rather than implied competence. That is more than most of this category publishes. It does not publish incrementality work, and it publishes nothing specific to subscription economics.
Best for: apps that have outgrown the two big platforms and need someone who can buy networks, OEM and programmatic without learning on your budget.
Wrong choice for: a subscription app whose main problem is that trials are not converting, which is a creative, paywall and product question rather than a channel question.
3. Miri Growth
Games heritage applied to apps. It names Meta, TikTok, Google UAC, Apple Ads, Snap, CTV, Moloco and Liftoff, which is a media plan that looks like a mobile games UA team rather than a paid social shop, and that is the point of hiring them. Client logos span Supercell, Zynga, King and Rovio alongside Duolingo, Capital One and Runna.
Measurement is published as MMP management and signal optimisation, which is rung one with a nod toward rung two, without detail on what is reported. It publishes one named subscription case study. It is backed by MSQ, which it states openly on its own site, describing a group of 1,850 people across 22 offices.
Best for: apps that need network and programmatic buying run by people who have done it at games volume, where bidding discipline across many sources is the skill being bought.
Wrong choice for: buyers who specifically want an independent boutique, and for anyone whose category experience question is about consumer subscription rather than games.
4. Phiture by Precis
A growth consultancy rather than a buying desk, and the distinction is the whole point. Performance marketing is one of five services alongside app store optimisation, retention, creative and a data, tech and AI practice. It does not specify which platforms it buys on, which for a consultancy is consistent rather than evasive. Client logos include Spotify, Headspace, Skyscanner, Adobe, NordVPN and Clue, and it names Adjust, AppsFlyer, Amplitude, Mixpanel and mParticle as measurement and analytics partners.
It publishes a subscription framework, which puts it ahead of most of this list on subscription thinking. It was acquired by the Swedish group Precis in February 2026 and now trades as Phiture by Precis.
Best for: companies with an in-house team that can execute, where the gap is strategy, measurement design and knowing what to build rather than someone to run the auction.
Wrong choice for: a small team that needs someone in the ad accounts every day. Buying consultancy when you needed execution is one of the more expensive mismatches in this category.
5. SplitMetrics
Software first, services second, and ranking it flat against agencies is a category error most lists in this space make. Its products cover Apple Ads automation, market intelligence and store testing, with a Growth Services arm offering fully managed Apple Ads and multichannel work, claiming more than 15 user acquisition channels. It states $2 billion in ad spend managed through the platform, which is a platform throughput figure rather than an agency service figure, and the two should not be compared.
On measurement it reaches rung two in a specific and useful way, combining Apple Ads data with mobile measurement partner post-install data in one view, with AppsFlyer, Adjust and Singular integrations. Named clients include Bolt, Skyscanner, NordVPN, SumUp and Babbel.
Best for: apps where Apple Ads is a serious line item and the team wants tooling plus managed execution on that channel.
Wrong choice for: anyone expecting a pure services relationship with a named team, or an app whose growth is creative-constrained on paid social.
6. The Social Outline
We are sixth on this page and it would be dishonest to be higher. This list ranks on channel breadth and measurement depth. We buy one channel properly. Meta is the core of what we do, with TikTok and YouTube run alongside where the economics justify it, and we do not sell app store optimisation, lifecycle, Apple Search Ads or ad network buying. If your definition of a user acquisition agency is a team that can buy across nine platforms, we are not one, and three companies above are.
What we do is the other half of the job. We are a performance creative agency that also runs the media, built around consumer subscription apps, with more than $100 million in Meta spend managed. On the measurement ladder we sit at rung two: mobile measurement partner and SKAdNetwork setup, and cohort-level return reporting taken from revenue data rather than platform-reported numbers. We do not publish an incrementality practice and we are not going to claim one.
The case for hiring us is narrow and specific. Most subscription apps that come to us are not short of channels. They are short of creative that produces winners, which is a different constraint with a different fix. Steps & Beasts is the clearest example on the site, where that work produced a 145% revenue increase and 118% more active subscriptions. Our pricing is published: from £1,500 per month as a flat fee rather than a percentage of spend, with a £2,000 minimum monthly ad spend.
Best for: consumer subscription apps already spending on Meta, where creative has stopped producing winners and the channel is not the bottleneck.
Wrong choice for: anyone who needs multi-channel user acquisition, app store optimisation, lifecycle or network buying. Take REPLUG, Miri Growth or Admiral Media instead, and we will say so on a call rather than after a contract.
7. Moburst
The largest brand roster here by some distance: Samsung, Reddit, Calm, Uber, eBay, Microsoft and Sony among others, with offices across the US, Tel Aviv, London and Sofia. Its service list spans social, search, networks and RTBs, influencer, email and OTT advertising, alongside organic, PR, creative and product work.
It places seventh on this page for one reason: on the axis this list ranks on, it publishes almost nothing. Its own site does not describe its measurement or attribution approach, and its channel list is stated as service categories rather than named platforms. That is a fair criticism of the published marketing, not a claim about the work, and Moburst ranks considerably higher in our broader app marketing comparison where breadth of service is the point.
Best for: brand-scale apps buying media, organic, creative and communications together under one roof.
Wrong choice for: a buyer whose decision hinges on attribution rigour, who will need to establish in the pitch what is not on the website.
8. Yodel Mobile
A London app agency operating since 2007, now part of NP Digital following an agreement announced in April 2025, and stating it on its own site. User acquisition is offered, but the centre of gravity is app store optimisation and store-led growth, and its published case studies lead with store outcomes.
It places last here because user acquisition is the secondary discipline and no measurement capability is published, which on this list's criteria leaves little to assess. On a store-led list it would rank near the top. Note also that headline statistics on its site are scroll-triggered counters that render as zero to anything that does not execute the page, so any figure quoted from that page in a competing list is worth checking.
Best for: apps where organic store visibility is the constraint and paid is support rather than the engine.
Wrong choice for: a subscription app that needs paid acquisition run as the primary growth channel.
Who was left out, and why
Three names appear on competing lists for this query and were considered and excluded, which is worth stating because omissions are usually invisible.
- Upptic describes itself as a growth marketing platform and services team for games growth, and its named clients are games studios. Genuine capability, wrong category for a consumer subscription app.
- AppAgent is mobile games focused, with no published consumer app work. It is a strong agency inside its category and that category is not this one.
- Appsumer is a user acquisition reporting product rather than an agency, so ranking it alongside media buyers is a category error. It is also no longer reachable: as of 18 September 2026 its website returns a 404 and the domain serves a certificate for an unrelated company. It still appears on at least one agency's current ranked list.
What the competing lists for this query actually are
It is worth knowing how the pages you are comparing were assembled, because this query is unusually dominated by interested parties.
Two of the ranked lists currently visible for this search were published by agencies that placed themselves first. Admiral Media's list ranks Admiral Media first of eight. SEM Nexus's list ranks SEM Nexus first of seven. Neither page discloses that the publisher is one of the ranked companies. Admiral's does at least publish five named criteria, which is more than most, though it does not explain how they were weighted to produce the order. This page has the identical conflict of interest, which is why the disclosure is at the top and why we are sixth rather than first.
A smaller signal, but a telling one about how carefully these lists are maintained. Two of the eight companies here have an obvious domain that is not theirs. REPLUG's real site is rplg.io, while replug.co is a parked domain offered for sale. Admiral Media's real site is admiral.media, while admiralmedia.com is also a parked domain offered for sale, confirmed on 18 September 2026. Several published lists link the wrong one. If a list cannot reach the right website, it is unlikely anyone verified what the agency does.
Six questions that separate the rungs
Take these into a pitch. They are designed so that a good answer is uncomfortable to fake.
- What event will you optimise toward in month one, and when would you move deeper down the funnel?A good answer names a starting event, a signal volume condition for moving, and the trade-off in learning speed. A bad answer is the name of a campaign type.
- What window do you report on, and what share of the outcome has landed by then?A good answer names a window and admits what is still missing at that point. A bad answer is a seven day return figure with no caveat.
- Are your numbers platform-reported or from our revenue data?This is the rung one to rung two test and it takes ten seconds. If the answer is the ad platform's own conversion count, you are buying rung one.
- At what spend would you recommend a holdout test, and what would you expect it to show?A good answer includes a spend threshold and an admission that below it the test cannot resolve. An agency that promises incrementality testing at any budget is selling you an unreadable result.
- Which channel would you turn off first if the budget halved?Tests whether they hold a view on relative efficiency or simply run everything they sell.
- Who owns the ad accounts, the measurement partner and the raw creative files?All three should be yours. Free to agree at the start and expensive to unpick later. Our list of questions to ask before hiring goes further on the commercial terms.
Frequently asked questions
What is the best user acquisition agency for a mobile app in 2026?
It depends on whether your constraint is channel coverage or measurement. Admiral Media is the only agency in this comparison that publishes a working incrementality method and offers to run it, and it maintains a dedicated subscription app practice that bids toward trial and subscription events rather than installs. REPLUG publishes the broadest channel list, including OEM, programmatic, SDK networks, rewarded and affiliate alongside the major platforms, plus an explicit mobile measurement partner and SKAN audit service. Miri Growth brings games-grade network buying to apps and is backed by MSQ. If your app is a subscription product and the bottleneck is Meta creative rather than the number of channels you are live on, that is the narrow job The Social Outline does, and it is why we place sixth on this particular list.
What is the difference between a user acquisition agency and an app marketing agency?
A user acquisition agency runs the buying function. Its scope is the auction: channel mix, campaign structure, bids and budgets, the events campaigns optimise toward, the feed of creative into tests, and the measurement that says whether any of it worked. An app marketing agency is a broader label that usually also covers app store optimisation, lifecycle messaging and sometimes product analytics. The distinction matters at the point of failure. If your store listing converts at half the category norm, a user acquisition agency will buy more expensive installs into the same leak rather than fix it, because every paid install still lands on the same product page.
What measurement should a user acquisition agency provide?
There are three rungs and they are not the same service. The first is plumbing: setting up and auditing your mobile measurement partner and your SKAdNetwork or AdAttributionKit conversion values, which tells you the data is arriving. The second is cohort reporting: reading outcomes from your own revenue data rather than from platform-reported numbers, which tells you what a cohort of users actually did. The third is causality: holdout and geo lift testing that shows what your spend caused rather than what it was credited with. Rung one is table stakes. Rung two is the real floor and plenty of agencies never reach it. Rung three only becomes meaningful above a spend level where a holdout has enough volume to produce a readable result.
Why is reporting on installs a problem for a subscription app?
Because the install is the fastest and least informative number in the funnel. RevenueCat's State of Subscription Apps 2026 finds that 50.6% of paid conversions happen on install day and 19.2% arrive in week six or later, both measured as a share of conversions rather than a share of installs. An agency reporting a seven day window on a subscription app is therefore showing you roughly half of what that cohort will eventually do. That is not dishonest, but it makes early cost per subscriber look far worse than it will settle at, and it makes any agency judged on a short window look worse than one judged on a long one.
How much does a user acquisition agency cost?
Almost nobody in this category publishes a rate. Of the eight companies compared here, one publishes fees for its management service: The Social Outline lists from £1,500 per month for accounts spending up to £20k and from £3,250 per month up to £75k, as a flat fee rather than a percentage of spend. Admiral Media publishes a rate from €363 per finished video ad for its AI creative output, which is a creative production price rather than a media management fee. The other six quote on request. Expect one of three structures: a flat monthly fee, a percentage of ad spend, or a hybrid with creative billed separately.
Should I trust a ranked list of user acquisition agencies?
Check who published it and whether they appear in it. Two of the ranked lists currently visible for this query were published by agencies that placed themselves first: Admiral Media's list ranks Admiral Media first of eight, and SEM Nexus's list ranks SEM Nexus first of seven. Neither discloses that the publisher is among the ranked companies, and neither explains how its criteria were weighted. This page has the same conflict, which is why the disclosure sits at the top and why we are sixth. Also check the list is maintained: one agency list still ranks Appsumer, a reporting product rather than an agency, whose website now returns a 404.
Decide which half you are buying
The choice on this page is not really between eight companies. It is between two jobs that share a name. One is coverage: more channels, more inventory, measured well enough to know which of them is worth keeping. The other is depth: one channel, worked hard, with the creative engine that moves the number. Buying the first when you needed the second is how an app ends up live on nine platforms and no better off.
Work out which constraint is real before you shortlist. If installs are cheap and subscribers are not, more channels will not help you, and our subscription app marketing guide walks the funnel from install to profitable scale. If your Meta account has plateaued because nothing new is beating the control, that is the specific problem we are built for, and you can apply to work with us. If your constraint sits anywhere else on this page, one of the other seven is a better call, and we would rather you found that out here than a quarter into a retainer.