← Journal

Designing for the machines too

2026-06-28

People are not the only ones reading your site anymore. Search crawlers read it. AI assistants read it. If you want to show up when someone asks an assistant "who builds websites in East Tennessee," the machine has to be able to understand the page.

What that actually means

Three things, none of them flashy.

First, structured data. Both sites ship JSON-LD: a Person for me, a ProfessionalService for the studio, cross-linked so a machine reads them as one identity across two domains. The studio's FAQ is real visible content and machine-readable schema from the same source, so the two never drift apart.

Second, an llms.txt. A plain-text summary at the root of each site, written so an assistant can lift the facts verbatim: who I am, what the studio does, where I am based, how to get in touch.

Third, the crawl basics done right. A sitemap that lists every real route. A robots file that explicitly welcomes the AI crawlers instead of leaving it to chance. Canonical URLs, so there is one clean address per page.

Why bother

Because the question is changing. Less "I'll Google a designer," more "who should I hire for this." If the machine can read your site cleanly, you land in the answer. If it cannot, you do not.