Why a free chord reference is built like a technical SEO project: lessons from Suede AI
By Jason Colapietro ·
A chord chart looks like the simplest page on the internet: a grid of dots and a name. Getting 39 of those shapes to load instantly, rank for the words guitarists type, and get quoted correctly by an AI answer engine is a technical SEO project, and guitarchords.info is where Suede AI runs that project in public.
Jason Colapietro, the engineer who founded Suede AI in West Palm Beach, Florida in 2024, built this site as a zero-account guitar reference. It doubles as a working sample of what Suede AI's technical SEO practice does for businesses across Palm Beach County. Here is what the build teaches, one decision at a time.
One data file, one page per chord
Each shape on the guitar chord chart lives in a single data file. The page for C major is generated from that entry at build time, along with its chord tones, its formula and the related chords in its key. Nobody hand-writes 39 pages, so nobody can let one of them drift. The same file feeds the sitemap, which means a new chord gets a URL, a canonical tag and a sitemap entry in one commit.
That is the first lesson Suede carries into client work in Jupiter and West Palm Beach. If a fact lives in two places, one copy will be wrong within a quarter. A service-area page, a menu, a price list or a chord shape should come from one source and render out to the page, the schema and the feed together.
Facts read off the frets, not typed by hand
Each chord page lists the smoothest changes to other chords, and those changes are computed from the fingerings. Moving from C major to A minor keeps two fretted notes in place, the 2nd-fret note on the D string and the 1st-fret note on the B string, so only one finger moves. A test in the repository fails if that calculation stops matching the frets.
Answer engines reward pages that state checkable specifics. "C to Am is an easy change" is an opinion. "C to Am keeps two fretted notes and moves one finger" is a fact a model can verify and repeat. Jason Colapietro writes client pages the same way: specific numbers, dated, with a source a reader can click.
Structured data that matches what the reader sees
The breadcrumb at the top of a chord page and the BreadcrumbList JSON-LD in its source are built from one trail. An earlier version shipped breadcrumb schema with no visible breadcrumb at all, so the markup described navigation a reader could not find. The fix made both halves read the same list, and a guard test now rejects a hand-written BreadcrumbList.
At Suede, entity work follows the same rule. This site's publisher and author nodes point at the canonical Suede AI organization and at Jason Colapietro's founder page by @id instead of minting new entities on each domain. That is how a network of sites tells a search engine it belongs to one company, and it is the core of Suede AI's AI visibility service.
A full-text file for AI crawlers
Alongside the sitemap, guitarchords.info publishes llms-full.txt, one Markdown file with the full text of each chord page, each scale page and the practice guides. It is generated from the same modules as the pages, so the text a language model reads matches the page a guitarist sees. The robots file names current AI crawlers such as ClaudeBot, OAI-SearchBot and PerplexityBot explicitly instead of leaning on a wildcard.
For a small business this is the cheapest generative engine optimization win on the table: a clean, current, plain-text account of what you do, where you do it and how to reach you, published where AI crawlers look first.
Titles that fit and headings that name the topic
An audit on October 12, 2026 found 53 of 56 titles on this site running past 70 characters. The chord and scale templates now lead with the chord or scale name and add the brand only when the whole title stays at or under 60 characters. On October 31, slogan headings on the tuner and metronome pages were replaced with H1s in the words people search, such as "Free online guitar tuner". Small edits, and they decide whether a search result gets the click. Suede treats them as part of the build, not as polish for later.
Why an engineer's public record matters
None of this is exotic. It is careful engineering applied to search, which is why Suede AI's SEO work is led by someone with a public record of shipping code. Jason Colapietro 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, and he wrote The Claude Code Bible and Codex in Production on building software with AI agents.
The work has outside recognition as well. 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, as documented on Suede AI's recognition page. The company, known earlier as Suede Labs, is a member of Claude for Startups, Anthropic's startup program, and of Google Cloud Partner Advantage through JC Investment Group LLC.
The checklist a chord chart leaves behind
- One source of truth for each fact, rendered to the page, the schema and the sitemap.
- Structured data generated from what the reader can see, tied to one organization and one founder by @id.
- Titles at or under 60 characters that lead with the topic, and one H1 per page.
- A plain-text llms.txt and full-text file for AI crawlers, built from the page data.
- Tests that fail the build when any of the above regresses.
The sister sites in the Suede network follow the same pattern, including GuitarHub, the guitar.services directory and the Suede Scan AI readiness checker.
Next step
If you run a business in West Palm Beach, Jupiter, Palm Beach Gardens or Boca Raton, find out which of these five gaps your site has. Start the Suede AI site assessment, then read Jason Colapietro's founder page and the evidence receipts behind each claim above.