← All companies

Coverage & method

Every chart on this site has its own denominator. Hosting is measured across the companies we could crawl; Companies House across those matched to a UK filing; GitHub across those with an attributed organisation. Those are three different populations, and comparing them without saying so would be misleading. This page states them.

What each collector has reached

Every column except Domains found counts companies, not domains: a company is counted once if at least one of its domains carries that fact. Percentages are of the companies in that index. “Domains found” is a count of domain names, which is why it is larger than the company count — most companies have several.

Subdomain discovery (Certificate Transparency) has found 50351 hostnames across all indices, of which 50351 have been resolved and probed.

Where the gaps come from

Sites that decline to be crawled

45 of the 177 published companies have no crawl data at all, and never will under the current rules. This is not a backlog. Our crawler asks for /robots.txt before fetching anything, and these sites' CDNs answer that request by refusing it outright — a 403 aimed at our user agent, not an absent file. We treat that as the refusal it plainly is, even though a strict reading of RFC 9309 would let us crawl anyway. We do not guess, and we do not work around it. See about our crawler.

Corrected 2 September 2026: this page previously said we treat any unreadable robots.txt as a refusal. That described the code and was stricter than the standard — a merely absent robots.txt now correctly means “no rules apply”. The count above is unaffected, because these sites refuse the request rather than lacking the file.

Everything else on those companies' pages — DNS, certificates, registries, Companies House — comes from sources that never touch the company's own servers, which is why those profiles are not empty.

Companies with no infrastructure of their own

Some constituents are investment trusts and closed-end funds: listed legal wrappers whose day-to-day web presence lives on their manager's domain rather than on one of their own. They are counted in the index totals above but have no domain to collect against, which is a real property of what they are rather than a collection failure.

Collectors that have not been extended yet

Companies House and GitHub enrichment currently run against the FTSE 100 only. Companies House is a UK register, so it has nothing to say about an S&P 500 constituent; the GitHub gap is simply work not yet done. Both are visible in the table above rather than hidden behind a percentage of a smaller denominator.

Pages with a narrower scope than the rest of the site

The digital sovereignty analysis is scoped to the FTSE 100 alone and says so on its own page. The FTSE 250 comparison is aggregate only — no individual FTSE 250 profile is published here.

What we deliberately do not collect

• Anything requiring authentication. No logins, no forms, nothing behind a paywall or a session.
• Active scanning. No port scans, no vulnerability probes, no attempts to find unlinked paths. Out of scope by policy.
• Search-engine result snapshots. Built, but switched off: the balancing assessment for storing third-party search results has not been completed, so it does not run.
• Any cross-linking of people between datasets. Persons of significant control are shown on the company they control and nowhere else — never aggregated across companies into a profile of an individual, and never joined to any other corpus.

How current this is

Most recent page crawl: 2026-08-28 16:46 UTC. Most recent subdomain probe: 2026-09-02 14:01 UTC. Collection runs in background passes rather than continuously, so individual facts carry their own collection timestamp on the company page they appear on.