Model 01
In-house ASO team
Wins on
Product knowledge, same-day small changes, and a seat next to LiveOps and legal.
Costs you
Headcount for every language, months of hiring, and a playbook that leaves when people do, unless it is written down.
ASO delivery models · App Store and Google Play
In-house wins for one core app, tools for data, a partner for 20+ markets.
Last verified 4 November 2026Published by Admiral Media, model five of fiveNo other agencies or tools named, no prices
| Live store-locales | |||
|---|---|---|---|
| Apps | Under 10 | 10 to 19 | 20 or more |
| 1 app | 1 app, under 10: 02In-house plus a tool | 1 app, 10 to 19: 02 + 03In-house plus a tool, with freelancers | 1 app, 20 or more: 01 or 05In-house team or a portfolio partner |
| 2 to 4 apps | 2 to 4 apps, under 10: 02 or 04In-house plus a tool, or a single-market agency | 2 to 4 apps, 10 to 19: 05Portfolio partner with your lead | 2 to 4 apps, 20 or more: 05Portfolio partner with your lead |
| 5 or more apps | 5 or more apps, under 10: 01In-house team | 5 or more apps, 10 to 19: 05Portfolio partner with your lead | 5 or more apps, 20 or more: 05Portfolio partner with your lead |
The short answer
Under 10 store-locales, in-house with a tool wins; a portfolio partner fits at 2 or more apps in 10 or more store markets, or one app in 20 or more. That is the rule Admiral Media applies when buyers ask which ASO model fits, even when the answer is in-house.
Updated
Original research, November 2026
271 of 1,107 language-supported app and storefront pairs were not fully localized in the App Store listings of 150 top apps we read on 2 November 2026.
Source: Admiral Media portfolio store benchmark, public App Store data collected 2 Nov 2026. 25 portfolios, 37 developer accounts, 150 top apps, 14 storefronts. A selected sample, not a random one. Custom product pages and tests are not visible in public data and are not counted. Public data does not show who runs each listing, so these gaps say nothing about any one delivery model. The full localization study Benchmark method
The comparison
In-house wins on product knowledge, a tool on data, freelancers on language range, a single-market agency on home-market depth, and a portfolio partner on 20+ store-locales.
No price in any cell; each chip is our one-line reading of it. Other agencies run model five too: compare them, us included, with the RFP below. Open a row for the full detail. How this page was written
| Criterion | Model 01In-house ASO team | Model 02In-house team plus a self-serve ASO tool | Model 03Freelancers plus a translation vendor | Model 04Generalist or single-market ASO agency | Model 05, how Admiral Media works. Other agencies offer it too.Portfolio ASO partner tied to Apple Ads and UA |
|---|---|---|---|---|---|
| Coverage of 20+ store-locales | Capped by headcount Limited by headcount and the languages the team speaks. Each new store-locale adds work for the same people. | Tracks every store Your team still researches, writes and designs every listing. | Many languages fast Someone in-house still briefs, checks and publishes each one. | Home market first Usually strong in its home market and a few large ones. Ask how many store-locales it runs today. | Built for it Every app and storefront is sorted into three tiers, so effort follows installs and revenue. Tier 3 markets get lighter, less frequent work. |
| Native keyword research per market | Own languages only Other markets often get translated keywords. | Search data only Picking the words locals actually type still takes a native speaker. | Depends on the brief Translation vendors translate keyword lists. Brief freelancers for research, not translation. | Often translated Often researched in one language, then translated. Ask who researches each language. | Native, every market Keyword research by native speakers in every market you run, kept in one keyword library per language and shared across apps. |
| Screenshot and creative localization throughput | Capped by designers Launches and in-app events compete for the same designers. | Your designers Mostly data. Screenshots and captions are still made by your designers. | Grows with briefing Grows with the number of freelancers, and so does the briefing and review work. | Often a separate scope Good craft per project. Localizing every screenshot set into 20+ store-locales is often a separate scope. | Tiered by market Localized screenshots and preview video in tier 1, localized captions in tier 2, reviewed captions in tier 3. Every listing is reviewed before release. |
| Custom product pages linked to keywords and kept aligned with Apple Ads | If one team owns both Works when ASO and Apple Ads sit in one team. In many companies they report to different people. | Manual linking Shows keyword data. Linking keywords to pages and ad groups stays manual work for your team. | Rarely in scope Freelancers seldom see your Apple Ads account. | If it runs Apple Ads If not, two teams plan from two lists. | One keyword map Each keyword cluster gets a custom product page and its own Apple Ads ad group that sends taps to that page. |
| In-app events and LiveOps cadence | Closest to releases When the ASO lead sits with the LiveOps team. | Monitors only Event cards and copy stay with your team. | Hard to keep weekly Every event needs a new brief and a new translation round. | Monthly rhythm Works to a monthly or project rhythm. Weekly LiveOps needs a tighter loop. | One calendar, all apps One release and event calendar across all apps, so metadata, in-app events and seasonal creative reach every store page on time. Event timing depends on your team sharing the release calendar early. |
| Compliance review for regulated apps (fintech, banking) | Closest to legal The review load lands on a few people. | Outside the tool Not a compliance tool. Approval steps run outside it. | You run approvals Each freelancer has to learn your rules. Approval and the audit trail stay with you. | Varies, ask Ask for the approval workflow and a sample change log. | Change log for auditors Every text change goes through your compliance review, with a written change log you can show an auditor. |
| Measurement and proof of incremental installs for finance | With analytics support Without a control-market test, results show rankings and conversion rate, not extra installs. | Rankings and estimates Rankings, visibility and estimates. Proving extra installs needs a test you design. | Out of scope Out of scope in most setups. | Often before and after Ask whether results are read against control markets. | Control-market tests Test markets against matched control markets, rules agreed before the change ships, and a readout with your revenue data. |
| Continuity when a key person leaves | Biggest risk Unless the work is written down. With a playbook and change log, a new hire can pick it up. | Data stays, judgment goes The data stays in the tool. The judgment that used it leaves. | Single points of failure Each freelancer is a single point of failure for their markets. | If documented The account survives staff changes if the work is documented. Ask who covers absences. | Named owner, playbooks A named senior owner, written playbooks and a change log. You now depend on an outside team, so write the handover file into the contract. |
| Console access and asset ownership | Yours throughout | Check exports Your consoles, plus a data subscription. Check what you can export if you stop. | Depends on setup Depends on your setup. Give each freelancer a named role and keep files in your shared library from day one. | Put it in the contract Should stay yours. Write it into the contract: your consoles, your files, your keyword sets. | Your consoles Work runs in your consoles, tools and calendar. Role-based access with the fewest rights first, edit rights only for the people who publish, approved by your Legal and InfoSec teams. |
| What drives the cost | Salaries, mostly fixed Plus hiring time and management. The same team costs the same in 5 store-locales and needs new hires to cover 50. | Subscription plus hours A subscription, plus your team's hours to act on the data. | Day rates or per word Plus your team's time to brief, check and publish. | Project fee or retainer Set by scope and review rounds. | Store footprint Apps, store markets, stores, custom product pages and how often you test. Separate from media. Plans run at least 6 months. |
| Time to first live change | Fast if staffed If you are hiring, add the time to hire and onboard. | Data on day one The first change ships when your team has time to make it. | Days per market For one market, once briefs and access are in place. | Ask for a date Depends on its onboarding. Ask for the date of the first live change in the proposal. | Weeks 3 to 6 Access, baseline and the tier map in weeks 1 to 2. First approved updates in tier 1 markets in weeks 3 to 6. Slower to start than an in-house team that is already in place. |
| Right choice when | 1 or 2 core apps One or two core apps, a few markets, and native speakers for each on the team. | A skilled marketer A skilled in-house marketer, one or two apps and a few markets to watch. | One-off launch A one-off market launch or a translation backlog, with someone in-house to run it. | One set project One app, one or a few large markets, and a set project such as a relaunch. | Portfolios or 20+ markets 2 or more apps in 10 or more store markets, or one app in 20 or more, where Apple Ads and organic search share one keyword plan and finance wants proof. |
Model by model
Every model has a cost: in-house pays in headcount per language, a tool in your team's hours, and a portfolio partner in a program of at least 6 months.
Product knowledge, same-day small changes, and a seat next to LiveOps and legal.
Headcount for every language, months of hiring, and a playbook that leaves when people do, unless it is written down.
Data for every storefront from one subscription. A skilled marketer goes far with it.
Hours. A tool gives data and suggestions; your team still picks the words per language, designs screenshots and links pages to ads.
Language range and speed for a set job, with little commitment.
Briefing and review time, and keyword research that quietly turns into translation.
Craft and depth in its home market, and clear set projects.
Thin coverage beyond its main markets, and Apple Ads often run by another team.
Tiered coverage of 20+ store-locales, one keyword map for organic search and Apple Ads, a test plan with control markets, and continuity.
A program of at least 6 months and setup time for access and approvals. Not built for one app in one market.
Two numbers, four triggers
Count your apps and live store-locales, then check four triggers: the same two signals we use when a portfolio asks us.
Your footprint
If two models fit, start with the one your team can run next Monday.
Pick both answers, or tap a cell in the map at the top.
1 app, under 10 store-locales
Usual fitModel 02In-house plus a tool
An in-house marketer with a self-serve tool covers this. Check the listing with the ASO Checker each quarter. An agency would cost more than the work needs.
Run the ASO Checker1 app, 10 to 19 store-locales
Usual fitModel 02 + 03In-house plus a tool, with freelancers
Keep the work in-house with a tool, and brief native freelancers for keyword research and screenshots in the languages your team does not speak.
Run the ASO Checker1 app, 20 or more store-locales
Usual fitModel 01 or 05In-house team or a portfolio partner
Staff an in-house team with native reviewers per language, or bring in a partner that tiers the markets and works with your lead. One person cannot research and review 20+ languages alone.
Book a portfolio review2 to 4 apps, under 10 store-locales
Usual fitModel 02 or 04In-house plus a tool, or a single-market agency
An in-house lead with a tool is usually enough. Add a single-market agency for set projects such as a relaunch.
Run the ASO Checker2 to 4 apps, 10 to 19 store-locales
Usual fitModel 05Portfolio partner with your lead
Here a portfolio partner starts to pay: one keyword library per language, one calendar, and checks so your apps do not compete, working with your in-house lead.
Book a portfolio review2 to 4 apps, 20 or more store-locales
Usual fitModel 05Portfolio partner with your lead
Tier every app and storefront, research natively per language once, and plan custom product pages and Apple Ads from one keyword map. Your lead keeps the decisions.
Book a portfolio review5 or more apps, under 10 store-locales
Usual fitModel 01In-house team
A dedicated in-house ASO team with a tool can run this, because the markets are few. Add outside help for new markets or cover.
Run the ASO Checker5 or more apps, 10 to 19 store-locales
Usual fitModel 05Portfolio partner with your lead
Shared research across apps saves the most here. A partner runs the libraries and the calendar, your team keeps product knowledge and approvals.
Book a portfolio review5 or more apps, 20 or more store-locales
Usual fitModel 05Portfolio partner with your lead
The case Admiral Media's model is built for: tiers per app and storefront, one keyword library per language, custom product pages and Apple Ads from one map, and a named senior owner.
Book a portfolio review4.9 on Clutch from 48 reviews
Each one moves the answer toward more help, whatever your size.
Momentum depends on what was written down. Start the week 1 handover below and arrange cover before the next release.
Week 1 handoverTier the new markets and research keywords natively before launch, so each store page is ready on day one.
Plan a test against matched control markets, with the success rule agreed before anything changes.
How to prove itCommon in fintech and banking. Ask any provider for its approval steps and a sample change log before you sign.
2 or more apps in 10 or more store markets, or one app in 20 or more? Get one plan across every market.
4.9 on Clutch from 48 reviews
Fit
Three of these six common situations need no agency at all; two point to a portfolio partner.
Model 02In-house team plus a self-serve ASO tool
Buy a self-serve tool and check the listing with the ASO Checker each quarter. An agency would cost more than the work needs.
Run the ASO CheckerModel 01In-house ASO team
Keep it in-house and add a tool for data. Buy outside help only when you open new markets.
Model 03Freelancers plus a translation vendor
Brief native freelancers for keyword research, not only translation, and let your lead review and publish.
Model 04Generalist or single-market ASO agency
A set project in one market is what a single-market agency does well.
Model 05Portfolio ASO partner tied to Apple Ads and UA
Week 1 handover of access, keyword sets, asset library and test log, then first approved updates in tier 1 markets in weeks 3 to 6. It also works as cover while you hire a new lead.
See the week 1 handoverModel 05Portfolio ASO partner tied to Apple Ads and UA
Every text change goes through your compliance review with a change log, and results are read against matched control markets.
How the proof worksOur in-house ASO lead just left
Access, keyword sets, the asset library and the test log. Collect them before the notice period ends; each saves weeks of rebuilding, whoever takes over.
How momentum is kept
4.9 on Clutch from 48 reviews
Buying ASO
Ask every bidder the same eight questions, then run a 90-day pilot with the success rule signed off before anything changes.
At Admiral Media the pilot is the first 90 days of an ASO program of at least 6 months, and the success criteria are written down before work starts. The full method
Published client results
Each figure is from a published Admiral case study, with how it was measured on the line under it. None of the three is a control-market test yet.
LowerProperty appCustom product pagesApple Ads“Admiral Media were extremely helpful and efficient in making Apple's Custom Product Pages work for us.”
Alexandra ZinnowLead PPC, Digital Marketing, ImmoScout24−43%cost per acquisitionAgainst the default page, same ad groups, 4 weeksRead the ImmoScout24 case study 
HigherBrain training appKorean App Store+93%App Store search downloadsBefore and after localized metadata in KoreaRead the NeuroNation case study 
LowerMeditation appApple AdsNew markets−38%cost per install+54% click-through rateRead the Petit BamBou case study Questions
One app in a few markets? Check it first with the ASO Checker. Many apps? See enterprise ASO.
Use an in-house team with a self-serve tool under 10 store-locales, add native freelancers for one app in 10 to 19, and use a portfolio partner from 2 apps in 10 or more store-locales. One app in 20+ needs native reviewers in-house or a partner.
Pick a partner that can take over from your consoles and files in week 1: role-based access, the live keyword sets per app and store-locale, the source files for screenshots and videos, and the test log. Ask for a named senior owner and written playbooks. Admiral Media takes over this way and ships first approved updates in core markets in weeks 3 to 6.
Look for a partner that sorts every app and storefront into tiers, keeps one keyword library per language shared across apps, plans store updates on one calendar and checks that sister apps do not target the same terms. Admiral Media runs portfolio ASO this way on the App Store and Google Play, with custom product pages and Apple Ads planned from one keyword map.
Ask every bidder for scope by app, store-locale and tier, who does native keyword research, how custom product pages tie to Apple Ads, the compliance workflow, access model and continuity plan. Run the pilot in 2 or 3 test markets against matched control markets, write the success criteria before the first change, and read the result out at day 90.
For data, yes. A tool shows rankings and search data per storefront. It does not choose the words locals type, design localized screenshots or link keywords to custom product pages. With 20+ store-locales the work itself is the bottleneck, so a tool pays off only next to people who can act on it in each language.
Each store-locale is a listing to research, write, design and keep current, and few teams have native speakers for each. In Admiral Media's benchmark of 25 app portfolios on 2 November 2026, 271 of 1,107 language-supported app and storefront pairs (24%) were not fully localized, and 151 of those gaps still showed the base screenshot set.
By cost drivers, not one number. In-house cost is salaries and hiring time, and grows in steps with new languages. Tools charge a subscription plus your team's hours. Freelancers and translation vendors charge by the day or by the word. Agencies charge by scope. Admiral Media publishes no prices: its fee is scoped by apps, store markets, stores and custom product pages, and plans run at least 6 months.
No page written by one of the options is fully neutral. Admiral Media is one portfolio partner among several, and model five is how we work. We describe the other four as categories without naming any agency or tool, show no prices for any model, take our own facts from our published pages, and say plainly where another model is the better choice.
Method
The four other models are described as categories, without naming any agency or tool. Costs are shown as cost drivers, with no prices for any model, ours included. Admiral Media facts come from our enterprise ASO and ASO incrementality pages, read 4 November 2026. Client results come from published case studies, and the localization figures from our portfolio benchmark of 2 November 2026. If anything here is out of date, tell us through the contact page and we will review it within 5 working days.
More comparisons
Book a portfolio review
Share your apps, stores and the markets that matter most. A senior strategist replies within one business day, and will tell you if another model fits better.
What happens next
Start an enquiry