Anyone can say a restore works. A case study here is where we show one: a single real job or experiment, written up after the run, with the numbers that run produced. What came back, what did not, what we did, and what we chose not to do.
What a case study here contains
- One job or one experiment per post, dated, so you know when it was true.
- Numbers from the run’s own records: URLs, redirects, images, checks, time taken. Nothing estimated, nothing rounded up.
- What we did and what we did not do, including the gaps the archive left.
- No ranking or traffic promises. A restored site is a starting point, not a result.
We describe the service itself at the level of the pricing page: the AI reads the archive and picks the snapshots worth trusting, drops junk and parked pages, keeps the original URLs and builds the redirects, brings the original look back, and checks the result against the original before a human signs it off. How each part is done is our own work, and it stays off the page.
What we never publish
- A client’s job without the client’s written consent. There are no client case studies yet; the first will appear only after a client has said yes in writing.
- Content we have no right to republish. When a case study is about a site we do not own, we show the numbers from the run, not its pages, its text or screenshots of it. The writing belongs to whoever wrote it, even when the domain is parked and the site is long gone.
How internal tests are labelled
Every case study opens, in italics, with the domain it is about and whether we own it. You should not have to read to the end to find out that nobody paid for it. All three case studies so far are internal tests.
- What happens when an AI agent restores a site without us: undertheleague.com, a British non-league football blog on a domain we registered as a test case, so it can show screenshots.
- A 2009–2010 FA Cup fan site, restored from the Wayback Machine: facupscene.com, a domain we do not own, so it shows numbers and none of the old site’s content.
- An Arsenal fan site built as a JavaScript app, restored to WordPress: thearsenalreview.com, a domain we registered, so it shows screenshots; 514 articles at their original addresses.
For the how-to side, the guides sit in three hubs: website recovery guides for lost and hacked sites, expired domains for finding, vetting and rebuilding them, and AI search for getting found in ChatGPT, Perplexity and Google AI answers. Each case study links back to the hub it belongs to.
The case studies are listed below. It is a short list, and everything on it happened.