How to submit your AI tool to directories (and which ones are worth it)

There are a lot of AI tool directories. Most of them will accept your submission. A small number will send you anyone.

The submission mechanics are almost identical everywhere, so the interesting question is not how but where — and the honest answer is that you cannot know in advance, only narrow it down and then measure. This post covers both: how to make the submissions cheap enough that being wrong does not cost much, and how to work out which ones are worth repeating.

Prepare once, submit many times

The single biggest time sink is re-deciding your copy at every form. Every directory asks for roughly the same eight things. Write them once, in a text file, and paste.

Some surfaces also want a first comment — 200 words on why you built it. Write that once too.

With that file open, a submission is genuinely five minutes. Without it, it is thirty, and by the fourth one you are writing worse copy than you did for the first.

Make your own page do the work

Most directories that auto-fill their form are reading your HTML, and they all read the same tags. If your <title> is "Home | Acme", a lot of directories will call your product "Home" until you correct it by hand, once per directory.

So fix it at the source. The five things worth getting right before you submit anywhere: a <title> that names the product and what it does, a <meta name="description"> that reads as a tagline, an og:image at 1200×630, a square logo a crawler can find, and a sentence somewhere on the page that states your pricing.

Our readiness audit will tell you what a crawler currently makes of your page, with a specific fix for each thing it cannot find. It is free and does not require an account. Other directories run something similar; the checks are broadly the same because the tags are.

The tiering that actually matters

Directories fall into four groups, and the strategy differs for each.

Ranked launch boards. A daily or weekly leaderboard with votes and comments. Product Hunt is the largest, and launching there is free — their own guide says so: "Product Hunt is free for everyone to use." These are worth doing once, properly, on a day you have chosen. The launch day is the product; the listing afterwards is secondary.

Category directories. Permanent listings organised by what a tool does. There is no launch day. The value is that the listing ranks for a term for years, so accuracy and completeness matter far more than timing. This is the group with the widest quality spread and the one where a bit of triage pays — partly because the AI-specific ones have mostly gone paid. Toolify charges $99, There's An AI For That charges a $49 review fee for the minimal listing, and Futurepedia's verified listing is $497. At those prices "submit everywhere" stops being a strategy and starts being a budget.

Community surfaces. Hacker News, subreddits, Discord servers. Not directories, and they behave nothing like them. Show HN has published rules worth reading before you post — projects should be "something you've worked on personally and which you're around to discuss", and you should "make it easy for users to try your thing out, ideally without barriers such as signups or emails". A post that ignores that gets flagged, not ranked.

Review sites. Places built around ratings and buyer comparison rather than discovery. Slow to pay off, meaningful for B2B software with a sales motion, largely irrelevant for a $9/month tool.

Most people spend all their energy on the first group and almost none on the second, which is backwards for anything with a searchable category.

How to tell whether a directory is worth it

You cannot tell from the directory's own claims. Traffic numbers on a "submit your tool" page are marketing, and third-party estimates are estimates. What you can do in five minutes, per directory, before spending twenty:

1. Search a term it targets. Take a category it lists — "AI meeting transcription", say — and search it. If none of the directory's pages show up anywhere in the results, it is not going to rank you either.
2. Check whether the listings are alive. Open five entries at random. Dead links and products that shut down two years ago mean nobody maintains it, and an unmaintained directory does not attract readers.
3. Look for evidence of humans. Comments, votes, recent additions. A directory with 4,000 tools and no activity is a database, not an audience.
4. Check whether submissions are reviewed. If everything goes live automatically, the catalogue will fill with junk, and being in a junk catalogue is worth about what it sounds like.

We are deliberately not publishing a table of traffic estimates for named directories. We would have to source them from third-party tools whose numbers we cannot verify, and a made-up number that looks precise is worse than no number. What we do keep is a list of directories worth knowing about, grouped by what each one is for, with no invented metrics attached.

Do the measuring yourself

This is the part that turns directory submission from a chore into something with a return.

Tag every submission URL. ?utm_source=<directory>&utm_medium=directory costs nothing and means that in a month you will know, exactly, which directories sent people and which sent none. Some directories strip parameters or use a redirect, in which case you still have the referrer — check that too.

After a month, you will have a list of three or four that actually work and a long tail of nothing. That list is the entire output of the exercise, and it is worth more than the listings themselves, because next time you launch you skip straight to the four.

Expect the results to be unintuitive. Small directories with a specific audience frequently beat large general ones, and the surface everyone told you was essential often sends a dozen visits.

The mechanics, per directory

For anything with a review queue:

1. Submit with everything filled in. Half-complete submissions get rejected or sit in a queue longer.
2. If it requires a badge or backlink, decide before you submit whether you will add it. Submitting and then not adding it usually means being delisted quietly.
3. Note where you submitted and when. A spreadsheet with directory, date, status and URL prevents the "did I already do this one" problem that eats an afternoon six months later.
4. When it goes live, check the listing. Auto-filled listings get things wrong — wrong pricing tier, wrong category, a logo scraped from a favicon. Fix it while you are still logged in.

Submitting here

Ours takes about five minutes in the browser: paste your URL at /submit, we audit the page, and the form arrives pre-filled with what we found. You correct it and send it to review. A person reads it before it publishes.

There is also a path that does not involve a browser at all. We run an MCP server, so if you use a coding agent you can tell it to submit the project it is already working in — it audits the page, drafts the listing and opens it for you. It cannot publish, pick a date or pay for anything; those stay with you deliberately. The API and MCP page has the connection command.

Free listings require our badge on your site, which is one line of HTML. If that does not work for you, there is a paid option that removes the requirement — we wrote up what the paid tiers actually buy rather than leaving you to guess.