Data and privacy

We did not get it from you. Everything here is read from places that are already public: launch posts on Hacker News and Product Hunt, App Store and Google Play listings, domain registration records, and the pages apps publish about themselves. Because it was not collected from you directly, we are required to tell you it exists. That is what this page is.

A root domain, a name and tagline, when we first saw it, whether it still responds, what its site links to, and a category we infer. Every inferred label is stored with the evidence behind it, so anything we assert can be checked and argued with.

Business profiles only. If a founder attached a professional handle to their own public launch post, that handle may appear against the app. Personal accounts, personal photographs and anything about someone's life outside their product are out of scope permanently. Not scoped down for now, excluded.

Role addresses like hello@ or support@ are treated as business contact details. Anything that looks like a personal address is used only to verify a claim over an app and is never displayed; those are deleted after 12 months.

The news page reproduces a headline, a publisher's name, a publication date, the short summary the publisher broadcasts in their own feed, and the thumbnail they attach to it. The thumbnail loads from the publisher's own site; we keep only its address. We never store or show the article itself — every item links out to the publisher, and the traffic goes to them. We only read feeds from a list of outlets we maintain by hand, and an outlet that would rather not appear is removed from it on request.

Legitimate interest: keeping an accurate public record of which products exist and which have stopped. That record is the entire point of the site, it concerns businesses rather than private life, and the personal data involved is the minimum that identifies who is behind a product. Our reasoning is written down and we will share it on request.

Use our contact form and say what should go. We suppress rather than delete: the row stops being published anywhere on this site, immediately and permanently, while the historical record of what happened stays intact and unpublished. That is deliberate: deleting a domain would destroy years of observations about other things, and suppression removes you from public view just as completely.

You can also ask for a copy of what we hold, ask us to correct it, or object to us holding it at all.

Use the Report an error link on any app page, or our contact form. We take funding and team-size labels especially seriously: those are inferred, they are the ones most likely to be wrong, and a confidently wrong “venture-backed” on somebody's side project is the most damaging thing this site could print.

Signing in sets a cookie. That one is not optional: without it there is no way to stay signed in, so it is set whether or not you agree to anything else.

Analytics is optional and we ask first. If you are in the EU, the UK or Switzerland, Google Analytics and PostHog load only after you accept, and declining means neither script is ever fetched. Elsewhere they load unless you decline. Wherever you are, a Do Not Track or Global Privacy Control signal from your browser counts as a decline. To change your answer at any time, use Cookie settings at the foot of this page. While analytics is on and you are signed in, PostHog links the pages you view to your account, by its account number and never your email address, so we can see what brings people to sign up. With analytics off, nothing is linked. We also count how often an app appears in results and how often its record is opened, so that founders who have proved a listing is theirs can see what it did. Those counters hold a date, an app and a number, with no cookie, no address and nothing that could identify you.

When something breaks on our servers, the error is reported to Sentry, which stores it in the EU for 30 days. Nothing runs in your browser for this, and addresses, identifiers and anything you typed are removed from the report before it is sent.

Ask AppWatch, the chat for signed-in accounts, sends what you type to Anthropic, which answers with its Claude models. With it go the earlier messages of the conversation, the page you asked from, the facts you saved for it to remember and, when a question is about your own ideas, apps or watchlist, what that question needs from them. Anthropic does not train its models on this, and deletes it within 30 days, unless its systems flag a request as breaking its usage policy, when it may keep it for up to 2 years, or the law requires longer.

A question that searches the web is answered without your ideas, your apps, your watchlist or what Ask AppWatch remembers, so a search can carry what you typed but never those. A question about search interest sends up to five search terms, a country and a date range to the search-data provider we buy from, which reads them from Google Trends. Nothing that identifies you goes with them, and we keep the terms, without your account, for 14 days after the week you asked.

Your conversations stay in your account for 12 months after their last message, and you can export or delete them from your account at any time. Nobody at AppWatch reads them unless you report one, which sends us a copy of that conversation. We keep a report and its copy for 12 months, even if you delete the conversation, and delete it with your account.

We respect robots.txt, identify ourselves as AppWatchBot, and use official APIs wherever one exists. If our crawler is causing you a problem, tell us through our contact form and we will stop fetching your site.