How to promote your app: 14 ways that still work in 2026

Promoting an app is not one activity. It is a dozen small ones, and most guides list them all as if they were equal. They are not. Some return traffic the same afternoon and nothing after. Some return nothing for a month and then keep paying for years. A few are worth doing only once.

This is the ranked version. Fourteen ways, each with what it gives you, roughly what it costs in hours, and where it belongs in the order. Nothing here needs a budget.

The short answer

If you have one afternoon: submit the app to five or six directories that carry a followed link, write one honest launch post on the community your users already read, and add an llms.txt to your site. If you have a week: add a Product Hunt or Show HN launch, publish two comparison pages for the tools people already search for, and ask your first twenty users for a review.

Everything below is that answer, expanded.

1. Get listed in directories that link back

Gives you: followed links, and a trickle of search traffic that does not stop. Costs: about ten minutes a submission.

A directory listing is the cheapest permanent asset an app can have. The page describes your product, links to it, and stays up whether or not you ever log in again. Pick directories that carry a followed link and are reviewed by a person; the rest are link farms and Google knows.

Full disclosure: SpacerrApps is one of these, and it is free. Every listing here gets a page, a write-up and two followed links, and a person reads every submission first. Read the ten worth an hour for the others.

2. Launch on Product Hunt, once, properly

Gives you: a day of traffic, a few hundred sign-ups if it goes well, and a badge. Costs: a week of preparation and the launch day itself.

Product Hunt still works for the right product on the right day. It does not work as a thing you repeat, and it does not work unprepared. Have the gallery, the first comment and the maker story ready before midnight Pacific, and read the launch guide before you book the date.

3. Post a Show HN

Gives you: the most technical audience on the internet, and honest feedback. Costs: an hour to write, a day to answer.

Hacker News rewards a plain title, a working demo and a founder in the comments. It punishes marketing copy. If your app is for developers this is worth more than Product Hunt; if it is not, it can still send a few thousand visits when the post lands.

4. Write comparison pages for the tools people already search for

Gives you: search traffic from people who are already looking to buy. Costs: two or three hours a page.

Nobody searches for your app by name yet. They search for "alternative to X" and "X vs Y". A page that compares your app to the incumbent, honestly, ranks for exactly the query that ends in a sign-up. Write two, measure which one moves, then write more of that kind.

5. Add an llms.txt and clear structured data

Gives you: a chance of being the answer when somebody asks an AI what to use. Costs: an hour.

ChatGPT, Perplexity and AI Overviews now recommend software by name. They read plain pages that say what a product is, who it is for and what it costs, in that order. An llms.txt at your root and a description that answers those three questions in the first paragraph is most of the work. Here is how to get cited.

6. Ask your first users for a review

Gives you: the social proof every later channel needs. Costs: twenty minutes of emails.

A listing with three reviews outperforms the same listing with none, on every directory and in every store. Ask the people who already replied to you. Ask for the honest version, not the five-star one, because a page of identical praise reads as bought.

7. Answer questions where they are asked

Gives you: trust, and links from threads that rank for years. Costs: fifteen minutes a day.

Reddit, Indie Hackers, Stack Overflow and niche forums all have threads asking for exactly what you built. Answer the question first, mention the app second, and only when it is the honest answer. One good reply in a thread that ranks is worth more than a hundred posts about yourself.

8. Publish a changelog and ship in public

Gives you: a reason to post every week without inventing news. Costs: ten minutes a release.

A changelog is the easiest recurring content there is. Every release becomes a post, a tweet and an email, and a public record of momentum is what a visitor checks before they trust a new product with their data.

9. Put a badge on your own site

Gives you: referral visits from the directories you are on, and a signal that you are real. Costs: two minutes.

Every directory that lists you offers a badge. Put the one or two that matter in your footer. They send visitors back to a page that says good things about you, and they are the cheapest credibility a new site can buy, because they cost nothing.

10. Get on the app stores and marketplaces that fit

Gives you: distribution from inside the tools your users already open. Costs: a day per store, then review time.

If your app can live in the Chrome Web Store, the Slack marketplace, the Figma community or the Shopify app store, it should. These stores have search traffic of their own and a review process that, once passed, is a permanent listing.

11. Write one piece of original research

Gives you: the kind of page other sites link to without being asked. Costs: a week, once.

Guides get read. Data gets linked. Count something nobody has counted, publish the method with the numbers, and that page will earn links for as long as the numbers stay current. It is the single best thing a small site can do for its own authority.

12. Send a launch email to everybody who ever asked

Gives you: the first hundred users, if you collected addresses on the way. Costs: an hour.

The waitlist, the beta testers, the people who replied to a tweet. One email, one link, one sentence about what changed. Getting the first hundred users is mostly this.

13. Speak to one community, not the whole internet

Gives you: users who tell other users. Costs: nothing but restraint.

An app for accountants should be promoted where accountants are. Every channel above works better pointed at one audience, because the copy can say the thing that audience cares about instead of the thing everybody can agree with.

14. Repeat the ones that moved

Gives you: a marketing plan that is yours rather than borrowed. Costs: a spreadsheet.

Track where sign-ups came from for a month. Two or three of these fourteen will account for most of them. Do those every week and drop the rest without guilt.

The order to do them in

Directories and the badge first, because they take an afternoon and keep working. Then the comparison pages and llms.txt, because search and AI answers compound. Then a launch, once, with the preparation it deserves. Everything else fits around those three.