Most apps do not fail because nobody could find a marketing channel. They fail because the team spends on channels before the app and its store page can turn attention into people who stay. Promotion multiplies what is already there. If one visitor in a hundred installs and most of them leave on the first day, more traffic only makes the loss arrive faster.
So this guide starts with what to fix first, then goes through the channels in roughly the order most apps should try them, from free to paid, and ends with a simple way to work out what is actually producing users. It is written for any app. If yours is a game, how to advertise a mobile game goes deeper on that market.
Before you promote anything
Three checks take a few days and decide whether anything after them pays off.
- Say who the app is for in one sentence. Not a category but a person with a problem: freelancers who lose track of invoices, parents planning a week of school lunches. Every headline, screenshot and targeting choice comes from that sentence.
- Look at the store page as a stranger would. The icon, name, subtitle and first screenshots do most of the selling. If they describe features rather than the result someone wants, rewrite them before buying a single click.
- Measure whether people stay. Watch how many new users are still active after one, seven and thirty days. The numbers vary widely by category, so compare with your own previous releases and fix the first session before scaling. An onboarding flow that loses half its users is the cheapest growth problem you will ever solve.
Free and low-cost channels
These cost time rather than media budget. They rarely scale on their own, but they produce the first users, the first reviews and the first honest feedback.
1. App store optimization
Store search is one of the largest sources of installs for many apps, and it keeps working after you stop paying. Research the words people use, put the most important ones in the title and subtitle, and treat screenshots as ads. App Store Optimization covers the method, and screenshot sizes and design covers the images that do most of the converting.
2. Ratings and reviews
A low rating quietly damages every other channel, because it sits beside your listing wherever people see it. Ask for a rating at a good moment, after someone completes something, using the platform's own in-app review prompt rather than a custom one. Reply to bad reviews, because future visitors read the replies too.
3. Your existing audience
A website, newsletter or customer list converts better than any stranger you can buy. Add a store link to your site, your email signature and your product emails. On iOS a Smart App Banner shows a small install prompt to people visiting your website on an iPhone, and it takes a single meta tag to enable.
4. Communities and launch sites
Find where your users already talk: subreddits, Discord servers, forums, Facebook groups. Join and be useful before mentioning the app, and read each community's rules on self-promotion, because a removed post costs more than a missing one. A launch on a site like Product Hunt suits apps aimed at early adopters, and is less useful for mass-market ones.
5. Content and search
Write the page someone would find while looking for the problem your app solves. A budgeting app can answer questions about budgeting for years. It is slow, but it compounds, and those visitors arrive with intent. Short videos on TikTok, Reels and Shorts work the same way for apps with something visual to show.
6. Referrals
If your app gets better when friends use it, or if a reward can be given to both sides, a referral programme can bring the cheapest users you will ever get. It only works once people genuinely like the product, so build it after retention is healthy.
7. Cross-promotion and partnerships
Other apps with the same audience but no direct competition can swap promotion at no media cost. If you already own more than one app, cross-promotion lets each one send users to the others.
Paid channels
Paid acquisition buys speed and predictability. Start with one channel and a small budget, because you are buying information first and users second.
8. Apple Ads
Ads inside the App Store reach people already looking for an app like yours, which makes this the natural first paid channel on iOS. You bid on search terms and the ad is built from your store page. Apple Ads explains campaign structure.
9. Google App Campaigns
One campaign runs across Search, Play, YouTube, Display and more, and Google automates the targeting. You supply text, images, video and, on some inventory, playable ads. Google App Campaign creative specs lists what to prepare.
10. Meta
Facebook and Instagram offer broad reach and are often the most scalable channel for consumer apps. Creative does most of the targeting now, so plan to make and test many ads. See Meta app install ads.
11. TikTok
A strong fit when your app is easy to show in a few seconds and your audience is young. Ads have to look native to the feed, as covered in TikTok app install campaigns.
12. In-app ad networks
AppLovin, Unity, Liftoff, Moloco, Mintegral and others place your ad inside other apps, mostly games. This inventory is built around interactive formats and suits games first, but it also works for finance, retail, food delivery and other apps. A playable ad lets someone try the core idea before installing: a loan calculator that gives a real answer, a meal planner that builds a sample week. Playable ads for non-gaming apps shows what that looks like.
What to make for paid channels
The ad itself is the biggest variable you control. Three formats cover almost every channel:
- Short vertical video, 15 to 30 seconds, showing the app solving the problem in the first three seconds. Feed platforms run on this.
- Static images, which are cheap to make and useful for testing messages quickly before committing to video.
- Interactive ads, such as playable ads and interactive end cards, where the viewer does something before installing. People who choose to install after trying the idea usually know what they are getting.
You do not need a developer for the last one. Playable Ads Maker builds interactive ads from templates or from scratch in a browser and exports the right package for each of 12 ad networks, so one idea can be tested in several places without being rebuilt.
A simple first budget
Forget percentages of revenue when you are starting out. Decide how much you can afford to lose to learn something, then split it up:
- Pick one channel that matches your audience. Apple Ads for iOS search intent, Google App Campaigns for Android reach, Meta for consumer apps with a clear visual story.
- Prepare three to five different ads, each with a different message, rather than variations of one idea.
- Run until each ad has enough impressions to compare fairly, then cut the weakest and write new ones in the direction of the strongest.
- Before spending more, check what the users who came from that channel did. If they stay and pay, or convert to whatever you care about, scale slowly. If they leave, go back to the product and the message.
A mobile game marketing plan and budget goes further into splitting spend, and most of its logic applies to any app.
How to know what is working
Store consoles show you sources in broad strokes, but they cannot compare paid channels on what users went on to do. Once you spend on more than one channel, a mobile measurement partner connects each install to its campaign and to later events.
Judge channels on the value of the users they bring rather than the cost of an install. A channel with installs at twice the price can still be the best one if those users pay three times as much. ROAS and LTV explains how to work that out. And remember that paid ads often lift organic installs too, so switching a channel off is sometimes the only way to see what it was really doing.
Common mistakes
- Buying traffic before fixing the store page and first session.
- Spreading a small budget across five channels so that none of them produces enough data to read.
- Testing small variations of one ad instead of genuinely different messages.
- Stopping promotion after launch week. Most apps grow through steady testing rather than a single spike.
- Judging success by downloads when the business depends on people who stay.
Create playable end cards in minutes—no code required.
Open Playable Ads Maker