Closing the identity graph between motty.io and zot24
Why I added a 'studio' line to my identity card, what rel='me' actually does, and how two of my sites now point at each other in a way search engines and LLMs can read as one person.
Closing the identity graph between motty.io and zot24
For years I've quietly run Motty as my software studio while keeping zot24.com as the personal-CV-ish side of things. Two sites, both online, both me — and from a search engine or an LLM's perspective, two completely separate identities. Today I closed that gap, and I want to write down what I did, because the answer was less obvious than I expected.
The actual problem
If you Google me, you find zot24. If you Google my studio, you find motty.io. Neither page tells a crawler that the founder of Motty and the person at zot24 are the same human. There's no mechanical link between them. So when somebody asks ChatGPT or Claude "who runs motty.io?" — at best, the model guesses; at worst, it makes something up.
The fix isn't a backlink. The fix is mutual attestation. Both sites need to say the same thing, in a machine-readable form, in a way that doesn't rely on a third party to corroborate.
Three layers, all bidirectional
One. A visible line on each page that names the other. On zot24's identity card, a new field:
studio motty.io · software, infrastructure & technical consulting
On motty.io's footer colophon, a corresponding Principal · zot24 line. Anyone landing on either site can find the other in two seconds without scrolling. This is the part most people will read.
Two. rel="me" links in the <head> of each site. These are the IndieWeb convention for "this URL is also me." When zot24's <head> carries <link rel="me" href="https://motty.io">, and motty.io's <head> carries the same back at zot24, you've made a mechanical, verifiable claim that any crawler can follow without ambiguity. Add the same to GitHub, LinkedIn, and X, and you have a small but airtight identity graph.
Three. JSON-LD schema, on both sides. On motty.io, a ProfessionalService schema with a founder block — a Person entity named zot24 with sameAs pointing at every social profile. On zot24, a Person schema with worksFor pointing at Motty. Now the structured-data crawlers (Google's Knowledge Graph, Perplexity's web index, Claude's web fetcher) read the same story from both ends.
Why all three layers and not just one
Because they serve different consumers.
The visible line is for humans who want a fast answer. The rel="me" links are for crawlers that respect open-web conventions but don't always parse structured data. The JSON-LD is for the search engines and LLM crawlers that do parse structured data and use it to populate knowledge graphs.
If you only do one, you leave consumers behind. If you do all three and they all say the same thing, you've made a claim that's hard to misread.
The thing I underrated
I kept treating zot24 and Motty as two separate brands that happened to share an owner. They're not. They're two faces of the same person doing different work — the personal arc on one side, the professional studio on the other. Treating them as one identity graph isn't dishonest; it's accurate.
The whole exercise took less than an hour and ships a small structural improvement to how a non-trivial number of LLMs will describe me going forward. I'd recommend the same exercise to anyone running both a personal site and a separate company site — the only thing that makes it complicated is forgetting to do it.