StackScope Discovery

How we find new launches ourselves, and what we do and don't publish about them.

What Discovery is

Most launches on StackScope come from the public places founders announce them: Product Hunt, Hacker News, PeerPush, and the launch directories we track. Those launches are self-announced. Discovery is the other channel: StackScope finds newly launched websites itself, from publicly available internet infrastructure, and checks whether a real product has gone live.

It exists because plenty of founders never post to a launch platform. They put a product on its own domain and get on with it. Discovery lets the catalogue include those launches too, usually within days of the site going live. Launches found this way are labelled StackScope Discovery on their pages and carry a Discovery badge when browsing. The crawler that does the finding is FossickBot (details and opt-out at fossick.bot).

What earns a public page

The public catalogue is for indie product launches: a real product, on its own website, from an independent maker or team. Discovery sees far more of the web than that, so what it finds is filtered before it counts.

• Automatic checks discard the obvious noise first: parked domains, placeholder pages, dead sites, and spam.
• Sites on major platforms and marketplace listings are excluded. We focus on independent products with their own websites.
• A launch that only Discovery has found joins the indie catalogue through review: clear cases are screened automatically, ambiguous ones are decided by a person. Nothing goes straight from the crawler into the headline listings. A discovered site that later shows up on a launch platform joins the same way every platform launch does.

What stays aggregate-only

Most new websites are not product launches. A dental practice, a restaurant, a company rebrand: new sites, not new products. Discovery observes them, but they don't become catalogue launches. They feed aggregate statistics only, under the label StackScope Wider Web, where they help answer questions like what share of brand-new websites run on a given platform.

Sites that fail review don't sit in a private catalogue. The record itself is deleted; what we keep is a short note of what was rejected and why, which trains the filters that screen future finds, and spam domains are blocked from re-entering.

What we publish about a listed site

Everything StackScope publishes about a site comes from what a regular browser would load from its public pages: the frameworks and tools visible on the page, hosting and DNS configuration, response headers, and a screenshot. We never log in, submit forms, or look at anything behind authentication.

We publish conclusions, not raw evidence. A launch page states what a site appears to be built with and how confident we are, and leaves it at that.

The catalogue records facts about websites, not people. StackScope has no user accounts and does not collect personal data. The privacy page covers data practices in full.

Your site, your call

If Discovery found your site and you would rather it had not, everything below is self-service and free.

• Block the crawlers. Discovery fetches are made by FossickBot and full analysis by StackScopeBot. Both identify themselves and obey robots.txt. Opt-out rules are on the bot page and at fossick.bot.
• Claim your launch. Open your launch page, request a fresh snapshot, and follow Claim this launch. Verification takes a meta tag, a file, or a DNS record, and needs no account. A claimed launch can be re-scanned after fixes or wear a score badge.
• Remove your site. A claimed launch can be removed entirely, immediately, with one click. The full removal flow is on the bot page.

For how the scores are calculated, see the methodology page. For the project itself, see about.