App launch guide
App Launch Checklist: How to Get Your First Users
Your app is live and the download counter barely moves. Work out where users are lost, then follow a simple four-week plan without an ad budget.
An app does not market itself. If nobody downloads it, check three things in order: whether people understand what it does, whether the store page convinces them to install, and whether anyone sees it at all. Then fix the positioning and store page first, pick two channels where your users already spend time, and compare installs week by week.
Launch in the right order
The app is in the store, and the download counter barely moves. That does not mean the app is bad. Most first apps ship without a plan for how anyone will hear about them, and the store does not hand out attention just because something new appeared.
The fix is an order of work, not a list of tricks: first make the message clear, then make the store page convincing, then bring people to it, and finally measure which step loses them. Doing it the other way around—posting everywhere before the page explains anything—spends your best early attention on a page that cannot convert it.
Why nobody downloads your app: three checks
Zero downloads usually has one of three causes. They look the same from the outside—a flat install chart—but they need different fixes, so check them separately before you change anything.
| Check | What is going wrong | How to test it today | Where to fix it |
|---|---|---|---|
| 1 — Message | People do not understand what the app does | Show the store page to someone outside the project for five seconds and ask what the app is for | Rewrite the one-sentence promise |
| 2 — Store page | People see the app but do not install it | Compare product page views with installs in App Store Connect or Play Console | Fix name, subtitle, first screenshots, description |
| 3 — Distribution | Nobody sees the app at all | Look at impressions and store listing visitors: are they close to zero? | Pick two channels where your users already are |
Positioning: who it is for, in one sentence
Before touching the store page, write the promise as one line:
For [who], who [have this problem], [app name] helps [get this result].
A hypothetical example: “For freelancers who lose track of unpaid invoices, this invoice tracker shows every overdue invoice in one list and sends polite reminders.” If you cannot fill the template in a minute, the store page will not be able to explain the app either.
- Name one specific group of people, not “everyone.”
- Describe the problem in the words those people use.
- State a result they would notice, not a feature list.
- Read the sentence to three people from that group and ask them to repeat it back.
If they describe the app differently from how you meant it, change the sentence—not their understanding. This line becomes the source for your subtitle, first screenshot caption, and the first sentence of the description.
Fix the store page before sending traffic
Every channel you try later ends on the same product page. The name and subtitle mostly work for search; the screenshots and opening lines mostly work for the install decision. A useful order of fixes is: name, subtitle (or Google Play short description), first screenshot, then description.
- Name: brand plus a hint of what the app does, within the 30-character limit both stores use.
- Subtitle or short description: the outcome from your positioning sentence.
- First three screenshots: one message each—outcome, core workflow, differentiator—with captions readable at thumbnail size.
- Description: lead with the promise. Apple notes the first sentence is what people read without tapping “more”; Google advises putting the most important text in the top part of the description.
For the full picture of fields, limits, and how search works, read what is ASO. For the screenshots themselves, see screenshot best practices, the annotated examples, and how to create App Store screenshots.
Find your first users where they already are
Without an ad budget, the trap is trying to be everywhere at once and doing each channel badly. Choose two channels where the people from your positioning sentence already spend time, and take one concrete step in each this week.
| Channel | Effort | Good first step | Watch out for |
|---|---|---|---|
| Store search | Low once set up | Put the words users search for into the name and subtitle or short description | Slow to build; depends on ratings and downloads too |
| Online communities | Medium | Answer real questions in a forum, subreddit, or Discord where your audience already talks | Read each community’s self-promotion rules first |
| Launch sites and directories | Medium for one day | Prepare a clear one-line pitch, screenshots, and a short demo before you post | A spike of visits, not a steady channel |
| Short videos | Medium to high | Record the app solving one problem in under a minute | Needs repetition before it shows results |
| Personal outreach | High per person | Message people who have the problem and ask for honest feedback | Does not scale, but teaches you the most |
| Your own audience | Low if it exists | Tell your newsletter, followers, or existing customers what you shipped and who it is for | Only as large as the audience you already have |
Talk to the first people who install. Their questions show you what the store page failed to explain, and their words are often better captions than the ones you wrote.
The four-week plan
One month is long enough to try two channels and short enough to stay focused. Repeat the cycle afterwards and compare the numbers once a week.
| Week | Focus | Actions | Number to write down |
|---|---|---|---|
| Week 1 | Positioning and store page | Write the one-sentence promise, test it on three people, fix name, subtitle, and description | Baseline: impressions, page views, installs |
| Week 2 | Screenshots and first channel | Rebuild the first three screenshots around the promise, start channel one | Page views to installs |
| Week 3 | Second channel and measurement | Start channel two, tag your links, write down what each channel sends | Visits and installs per channel |
| Week 4 | Conclusions and store page fixes | Keep the channel that worked, drop or change the other, fix the weakest part of the page | Week 4 vs week 1 |
What to measure each week
Video views, likes, and followers are not installs. Track the path a person takes to your app as separate steps, so you can see where they drop off:
- Reach: how many people saw your post, video, or the app in the store. In App Store Connect this is impressions; in Play Console, look at where store listing visitors came from.
- Store page visits: product page views (App Store) or store listing visitors (Google Play).
- Installs: downloads, and how they compare with page visits. Apple reports a conversion rate; Google reports store listing acquisitions and click-through rate on the Install button.
- Real use: whether new users come back after the first day—check this in your own app analytics.
App launch checklist
- The one-sentence promise names who the app is for and the result they get.
- Three people from the audience can repeat the promise back.
- The name and subtitle or short description use words your users would search for.
- The first three screenshots each carry one message and stay readable at thumbnail size.
- The first sentence of the description works on its own.
- Two channels are chosen, and each has one concrete action this week.
- Links you share are tagged so you know which channel sent each visit.
- Baseline impressions, page views, and installs are written down.
- Each week ends with one change based on the numbers, not on a guess.
Official sources
- Apple: Creating your product page
- Apple: Search on the App Store
- Apple: App Store Connect metric definitions
- Google Play: Get discovered on Google Play search
- Google Play: Understand and grow your app’s user base
Specifications can change. We review these pages monthly and update the visible verification date only after checking the source material.