How a game studio should pick a UA and ASO agency: soft launches, Day 7 retention, ROAS by network, and the ownership questions that protect your consoles.
Why an app or SaaS agency is not a game UA team
A generalist agency hears “mobile” and reaches for the app playbook: an App Store listing, some Meta ads, a newsletter. Mobile game user acquisition is its own trade with its own vendors. AppLovin, Unity Ads and ironSource, Liftoff, Moloco, and Vungle are game ad networks, not names a general agency bids on every day. Pocket Gamer Connects runs a whole conference circuit for game discovery, UA, and monetization. If a candidate agency does not know that world, your budget will be spent learning it.
The funnel is also judged differently. A game campaign is not profitable because installs were cheap on install day. It is profitable when Day 7 and Day 30 retention hold and players either buy in-app purchases or watch rewarded video. Return on ad spend gets checked at Day 0, Day 7, and Day 30, and a campaign that looks great at the click can lose money once real retention plays out. An agency that reports cost per install and stops there is measuring the wrong thing.
Genre changes the whole plan. Hyper-casual runs on thin margins, ad monetization, and huge volume. Casual and puzzle titles blend in-app purchases with ads. Midcore and premium games lean on deeper spending and slower, more careful acquisition. One funnel template does not fit all three, and a generalist usually only has one. Premium titles are different again: every install is a sale at the store price, while a free-to-play install is only the start of a relationship that may never pay. An agency has to know which one you are shipping before it writes a single ad.
The first question: how do they judge a soft launch?
Ask a candidate to describe how they would run a soft launch for your next title. This is standard practice in games: ship to a handful of test markets, measure cost per install, early retention, and monetization, and only then spend real money on a wide launch. A team that knows games will name the checkpoints they watch and the numbers that would make them pause. A team that skips straight to “launch day plan” is guessing with your budget.
Push on creative next. Games are sold with playable ads and rewarded video, and store listings are won with gameplay-accurate screenshots and a trailer that shows the real loop, not polished feature cards. Ask who makes the creative, how often they refresh it, and how they test it. Creative fatigue is where game UA spend quietly dies.
Finally, ask how they handle attribution after Apple's App Tracking Transparency. User-level tracking on iOS is largely gone, and honest teams now work with aggregated and modeled measurement through a mobile measurement partner like AppsFlyer or Adjust. An agency promising precise user-level iOS attribution is either behind or overselling.
Which channels fill a game's funnel, and in what order
The store listing comes first because paid and organic both land on it. On Google Play, discovery is keyword and algorithm driven, so genre and mechanic terms like “match 3,” “idle RPG,” or “battle royale” belong in the title and description. On Apple's App Store, discovery leans more editorial and curated, so featuring and a clean product page matter more. Two stores, two jobs.
Paid UA comes second and runs across Google App campaigns, Meta, TikTok, and the dedicated game networks. This is where most of a studio's budget goes and where the Day 7 checkpoint decides whether to scale. Third is search on the open web and in AI assistants: players type “best puzzle games 2026,” “offline puzzle game,” “alternatives to [competitor game],” and “[genre] game no ads,” and they ask Gemini or ChatGPT which game to download. Ranking for those queries produces installs nobody paid for.
Creative volume matters more here than in almost any other category. Winning teams ship new playable and video variants every week, because the networks' algorithms reward fresh creative and punish ads the audience has already seen. Ask how many variants the agency produces a month and who keeps the source files.
If you run more than one title, cross-promotion between your own games is a channel no outside vendor can copy for you; make sure the agency plans for it. Then live-ops messaging: push and email tied to your update calendar, a welcome-back bonus for lapsed players, an event reminder. Store ratings close the loop; a rating prompt timed after a good session lifts both conversion and rank. Local SEO and a Google Business Profile have no place in this plan.
The game calendar and what a player is worth
Game demand follows two calendars. The first is the device calendar: new phones under the tree mean a wave of fresh installs in the days after the winter holidays, and summer breaks put more hours into play. The second is the ad-cost calendar: the run-up to the holidays is when every retailer bids for the same attention, so acquisition costs climb in the fourth quarter and ease in January. A good agency spends into the cheap windows and tests creative before the expensive ones. Live-ops shapes the calendar as well. A seasonal event, a new world, or a collaboration gives you a reason to re-engage lapsed players and a fresh hook for ads, so the UA plan should be built around your content roadmap rather than run beside it.
Talk value in your own words. A free install is a cost, not a win. A player who finishes the tutorial and comes back on Day 7 has a chance of paying. A payer in a midcore title can be worth many times an ad-supported player in a hyper-casual one. The question to ask is “what does a Day 30 retained player cost, by network and by creative?” Cost per install by itself is how studios end up scaling a network that sends players who quit before the first session ends.
Ask how the agency reports lifetime value against payback. If they cannot show ROAS by cohort and by channel, they cannot tell you which network to cut.
Red flags, and what stays in the studio's name
App Store Connect, Google Play Console, your ad accounts, your MMP account, and your player data belong to your studio, not the agency. Some vendors run UA from their own ad accounts to keep the spend history. When you leave, you start from zero with every network. Check who owns the creative, too. Playable ads and trailers cost real money to make; if the agency keeps the source files, you pay for them twice.
Be wary of anyone promising a chart position or a set number of installs. Store featuring is not for sale, and bought installs from low-quality sources tank retention and can trigger store review problems. Be wary of reports that show installs and cost per install but hide Day 7 retention and ROAS. That is the report of a team paid to spend, not to return.
Ask what happens on the day you part ways, and whether you can log into every dashboard yourself. Month-to-month terms mean the agency expects the Day 30 numbers to keep you, which is the right pressure to have on them.
Six questions for any mobile game marketing agency
One: “How would you structure a soft launch for our next title, and what would make you stop?” Two: “Which game ad networks have you run, and which creative formats worked?” Three: “Show me a report with Day 7 retention and ROAS by channel, not just cost per install.” Four: “How do you handle iOS attribution after App Tracking Transparency?” Five: “Who owns the store consoles, ad accounts, and MMP?” Six: “What is the plan for genre searches and AI assistant recommendations?”
Specific answers name networks, checkpoints, and creative tests. Generic answers promise reach.
SearchPod is one team that runs this whole funnel for studios: store listing, paid UA, ASO and SEO, AI search, and live-ops messaging, judged on retained, paying players rather than downloads. Prices are posted: Google Ads management at 10% of budget with a $600 per month floor and no markup, content at $50 a page from 10 pages, sites and landing pages as one-time builds from $1,500 to $20,000+, no setup fee, month to month, and a 30-day guarantee where you don't pay if it isn't working. A free proposal comes back within one business day from /get-proposal. Ask us the six questions too.
