How to Acquire Users for a New Mobile App
A new app has a cold-start problem. No ratings, no download history, no keyword rankings, so the stores have nothing to rank it on.
2026 made this harder. Apple says nearly 65% of App Store downloads come straight from search, and since March Apple Ads has been placing paid listings further down the results as well as in the top slot. Fewer organic spots are left for an unknown app to claim. Paid installs got pricier too: Singular’s Q2 2026 Quarterly Trends Report shows global cost per install up 12% in one quarter.
We see the same pattern across launches: the teams that do well start working on acquisition months before release, and they do things in a certain order. Below is that order, from roughly three months out to the end of your first quarter live.
8–12 Weeks Before Launch
The store listing comes first
Whatever you run later, search ads, TikTok creatives, a newsletter mention, tasks for real users, ends on your store page. A page that doesn’t convert makes every one of those channels more expensive.
Apple ranks search results on text relevance (title, subtitle, keywords, primary category) and on user behavior (downloads, ratings, reviews). You’ll have none of the second kind at launch. So spend the effort on the first: research what your users actually type, put the two or three most important terms in your title and subtitle, and choose the primary category on purpose rather than by default.
Screenshots do most of the selling. Hardly anyone opens the full description.
Google Play indexes the description, though, so write that listing separately. Copying the App Store text over loses search traffic.
Pre-registration on Google Play
Android users can sign up before your app is live. On launch day they get a notification, and on eligible devices the app installs itself. That’s a small crowd of installs on day one, which is exactly when you need them.
A few rules from Google Play Console Help to plan around:
- A campaign runs 90 days per country. After that, the app has to launch.
- You get one pre-registration reward for the whole campaign, and it can’t be edited or deleted once created.
- Google recommends testing on a test track first.
You can buy sign-ups too. Pre-registration ads run on Google Play, YouTube, and the Display Network and point to your Play listing. We like using them as a cheap early test of creatives, weeks before anything is spent on install campaigns.
iOS has pre-orders instead. Apple Ads can promote an app during its pre-order period, and people who pre-order receive the app automatically at release.
A featuring nomination
Since late 2024, App Store Connect has had a Featuring section where you can nominate an upcoming launch, new content, or a new feature for the editorial team to consider (App Store Connect Help). Apple’s Getting Featured page describes what editors look for.
It’s free, it’s not guaranteed, and most practitioners submit six to eight weeks before the target date. Say clearly what’s new, who it’s for, and which Apple technologies you use well. Editors read a lot of these.
Your own audience
A waitlist. An email list. A Discord server, a subreddit, a build-in-public thread. None of these will deliver big numbers, but they hand you something no ad buys: the first hundred people who’ll rate the app, report bugs, and tell someone about it. Paid traffic stops when the budget does. These people don’t.
Soft Launch
Release in one to three markets first and watch whether people stay. Volume isn’t the goal yet. If onboarding is broken, you want a few thousand users to find out, not a few hundred thousand. Adjust’s guide to soft launch strategy goes deeper into planning.
Write your pass marks down before the first install. Day-1 retention above X, a working tutorial, some share of users reaching the core feature. Teams that skip this step tend to decide afterwards that whatever they got was fine.
For orientation, Adjust’s benchmarks put the median app at about 26% retention on day 1, 13% on day 7, and 7% on day 30. Category and business model move these numbers a lot, so compare yourself with apps like yours. If you’re far below your category, fix the product before buying growth.
Choose markets for the answers they give. Canada and the Netherlands are common for casual games: broad audiences, cheaper installs than the US. Singapore and the Nordics get used when a game needs to prove it can earn a lot per player. If the US is your real target, test somewhere its users behave similarly.
Keep acquisition modest and steady. When retention and conversion hold for several weeks in a row, go wide.
Launch Week
Stores pay attention to download velocity, the number of installs over a short window, so a launch-week wave of real installs helps you start ranking. We wrote about how paid and incentivized installs pull organic ones along in Incentivized Traffic vs. Organic Traffic.
Search ads
People searching the App Store already want an app, and Apple says ads in the top position convert at around 60%. Bid on your own brand name first. If you don’t, a competitor will, and people searching for you will see them above you. Then add category terms your listing genuinely fits. With ads now appearing at several positions, check impression share by position to understand where installs come from.
On Android, Google App campaigns cover Search, Google Play, YouTube, and partner apps from one setup. Switch from optimizing for installs to optimizing for an in-app event as soon as volume allows.
Social and video
Meta gained more ad spend than any other network in Q2 2026, per Singular, and it’s still the biggest install channel for many categories. TikTok and YouTube work best when the product makes sense in 15 seconds of screen recording.
At launch, we’d treat these channels as a creative lab as much as a source of installs. Run several concepts at once. When one clearly wins, put that message on your store page too.
Real users for early data
Ad platforms learn from conversion data, and on day one you have none. Tasks for real users help here. On IPweb you set the action, say, install and finish registration, or complete a first session of a set length, and English-speaking users do it for a reward. You pay per completed task, get people through onboarding in the first days, and see where they get stuck. Rewarded User Acquisition: How It Works covers the mechanics if this model is new to you.
Keep rewards tied to what happens inside the app. Never reward ratings or reviews: Apple’s App Review Guidelines and Google’s User Ratings, Reviews, and Installs policy both forbid it, however the reward is delivered.
The First 90 Days
Compare channels on who stays
Cheap installs mean little if the users are gone in a week. AppsFlyer found a 30-day uninstall rate of about 46% among Android apps in 2024. So line up your sources by day-7 and day-30 retention, cost per registered or paying user, and early revenue. Shift budget slowly toward whatever brings people who stick, and keep one or two new sources in testing at any time.
Untangle paid and organic search on iOS
App Store Connect reports Apple Ads downloads inside the same “App Store search” source as organic search. If ads start taking installs your listing used to win for free, the total can stay flat while organic quietly drops. Subtract Apple Ads installs from App Store search downloads for the same dates and countries, and read the trend week by week.
Keep editing the listing
Your ads are a free testing ground. A headline that beat the others in paid campaigns deserves a spot in your screenshots. On iOS, build custom product pages for different audiences and match them to keyword groups, so someone searching “budget planner” and someone searching “expense tracker” each land on a page made for them.
Ask for ratings at a good moment
Use the stores’ built-in review prompts, and trigger them after something went well: a finished task, a streak, a saved document. Not on first open. And no reward attached, ever.
What It Costs in 2026
Install prices range from a few cents to tens of dollars depending on category, platform, and country, so treat any average as a rough marker. Singular’s Q2 2026 figures: $1.24 global CPI, $0.76 on Android after a 15.1% jump in one quarter, $3.03 on iOS after a slight drop. In the US, Android installs now cost more than iOS ones.
The same report shows installs per thousand impressions down 10.5%. More money is buying fewer installs. Singular puts it down to the funnel rather than bids, and we agree: for a new app, a better store page and better creatives usually do more for the budget than higher bids.
When planning, start from retention. With 7% of users active at day 30, each retained user costs roughly 14 times your CPI. Compare that with what a retained user is worth to you, and you’ll know how much you can spend.
Before You Hit Publish
Three months out: two separate store listings, Google Play pre-registration, a featuring nomination, a list of people waiting for you.
During soft launch: written pass marks, one to three markets, analytics running before the first install.
Launch week: brand and category search ads, several creatives in testing, real users going through onboarding, no rewards for reviews.
After that: channels judged on retention, organic search measured separately, listing updated with what worked in ads.
The first users are the hardest to get, because every channel works better once you have some. Launching soon? Create your first tasks on IPweb and get real English-speaking users through your onboarding in week one, before the big campaigns start.
Internet Marketer
