A RIG DIAGRAM IS A PHOTO UNTIL YOU GIVE IT FIELDS
Ask an answer engine what Tony Iommi played through in 1971 and it will either cite a page it can parse or guess from whatever forum thread it half-remembers. A scanned rig diagram, on its own, is an image with a filename. Nobody can quote it, filter it by decade, or tell whether the name attached to it was ever checked. Suede DNA exists to close that gap for 409 guitar rig diagrams from the Guitar Geek Archives on the Internet Archive, and the way Suede AI structured it is a working pattern for any archive that wants to be read by search engines and AI systems instead of skipped.
Jason Colapietro built the archive from Suede AI’s studio in West Palm Beach, Florida. This note walks through the decisions in the order they matter: what each record holds, which pages are offered to crawlers, how the machine-readable indexes stay in step, and what a Palm Beach County business site can borrow from all of it.
STEP 1: ONE MANIFEST, FIVE FIELDS PER RIG
Every rig in the archive is one entry in a generated manifest. Each entry records an artist association, a year, a file format, a structured source URL pointing back to the original archive.org file, and a stable identifier. That is deliberately modest. The manifest does not describe the gear in the photo, because Suede does not interpret the photographed equipment or re-verify the upstream attribution. It records what the source says and where the source lives.
The payoff is that every page renders from the same data. The Tony Iommi rig page and the Dave Murray rig page are not hand-written; their titles, descriptions, timelines, and source links all come from the manifest, and a validation step in continuous integration fails the run before the site builds if the manifest breaks its own schema. When one source of truth drives the HTML, the structured data, and the indexes, they cannot quietly disagree with each other, which is one of the most common reasons an AI answer misquotes a site.
STEP 2: DECIDE WHAT DESERVES TO BE INDEXED
The archive publishes 361 artist pages. Only 69 of them are offered to search engines. A page is indexed when its name has been reviewed by a person or when it documents at least three rigs. Single- and two-rig pages with unreviewed names render mostly shared template copy, so they stay published and linked for visitors but carry a noindex, follow robots directive.
The important part is not the threshold. It is that one function decides it, and the robots tag, the sitemap, and llms.txt all read that same function. Most sites that leak thin pages into the index do it because three systems each had their own rule. At Suede we treat “should this be indexed” as domain logic with test coverage, not as a checkbox in a plugin. That is also why the archive never pads the index to look bigger: an answer engine that finds thin pages learns to trust the whole domain less.
STEP 3: KEEP THE MACHINE-READABLE INDEXES IN STEP
Three files tell machines what the site is. The sitemap lists every indexable URL. The llms.txt file lists the same pages and describes each one with its own meta description, so a model reading it gets the page’s words rather than a paraphrase. The JSON-LD on every page resolves the publisher and author by reference into a single graph: the organization is always the canonical Suede AI identity and the author is always the canonical Jason Colapietro identity, never a fresh copy minted per page.
Reviewed artist pages describe their subject as a Person with an image gallery of rigs credited to the Guitar Geek Archives. Unreviewed pages make no identity claim at all and say plainly that the name is an upstream archive label. That restraint is what makes the structured data citable. A crawler that catches one false assertion has a reason to discount the rest.
WHAT A PALM BEACH COUNTY BUSINESS CAN TAKE FROM THIS
Swap “rigs” for “services,” “projects,” or “locations” and the pattern is the same one Suede AI uses for clients through its West Palm Beach SEO, GEO, and web development practice. A contractor in Jupiter with forty project pages, or a practice in Palm Beach Gardens with a dozen service pages, has the same three problems: records that live in people’s heads instead of structured data, thin pages competing with strong ones, and indexes that drift out of step with the site. The fix is the same too. Put the facts in one source, let a single rule decide what is indexed, and generate the sitemap, llms.txt, and schema from that source. The technical SEO service and the AI visibility service are built around exactly that sequence.
The tradeoff, as Jason Colapietro frames it: this approach shrinks what you show crawlers before it grows it. Suede DNA indexes fewer than a fifth of its artist pages on purpose. Expect a smaller, cleaner footprint first and better citations second.
THE RECORD BEHIND THE BUILD
Jason Colapietro does not ask anyone to take the method on faith, so the proof is public. He has 54 pull requests merged across 46 open-source repositories he does not maintain, including Jest, Adobe React Spectrum, Backstage, and the Polkadot SDK. He has shipped nine iOS apps under his own Apple developer account. On October 31, 2026, Suede AI SEO ranked #1 among Palm Beach County-based companies listed in The Manifest’s Florida AI Deployment directory and in the top 25 statewide, recorded on the recognition page. Suede AI is a member of Claude for Startups, Anthropic’s startup program, and Jason holds 21 earned course credentials across eight providers, including Semrush Academy’s Local SEO Essentials and HubSpot’s AEO Fundamentals, all shown in the evidence gallery. The team started out as Suede Labs, and the same habit of shipping receipts with claims runs through the free book The Screenshot, which explains how answer engines decide who to name.
YOUR NEXT STEP
Pick the one page type on your site that matters most for revenue and list the five facts every one of those pages should carry. If you cannot fill that list from a single source today, that is the first thing to fix. If you want Suede to look at it with you, start the free Suede AI assessment, or read more about how Jason Colapietro works on the Suede AI founder page. For more on how this archive handles names it has not verified, see rig diagram provenance and human authenticity.