SEO ARCHITECTURE AND SCHEMA
Structure first, so the content you publish has somewhere to land.
SEO work splits into two halves and most people buy the wrong one first. The content half — what to write, how often — is worthless on a site whose URLs collide, whose pages all claim the same keyword, and whose markup tells Google nothing about what the business is. This page is the structural half: clusters, URLs, markup, links.
In practice: a keyword cluster map where each cluster owns exactly one URL, a JSON-LD graph with @id-linked Organization, Person, Service and FAQPage nodes so entities resolve instead of floating, canonicals and hreflang where they apply, an XML sitemap generated from the content source rather than hand-maintained, and internal links that follow the cluster map instead of a designer's whim.
I also build the answer-engine layer, because a share of the traffic now never reaches a blue link. Generated llms.txt routes, FAQ markup that matches the question as typed, and a baseline run against the assistants to see what they currently say about the brand. On this site that baseline turned up name collisions nobody had noticed, which is the kind of thing you only find by looking.
WHAT I DO
- Keyword cluster map — grouped by intent, one cluster to one URL, with an explicit avoid list and the reason each term is on it
- Template architecture: which pages are hand-written, which are generated from a data source, and the rule that stops the generated ones being thin
- JSON-LD graph — Organization, Person, Service, FAQPage, BreadcrumbList, @id-linked so the same entity joins up across every page
- Programmatic page templates where there is genuinely distinct data per row, and killed where there is not
- Sitemap, robots, canonical and redirect map generated from the content source so they cannot silently drift from reality
- Search Console and Bing Webmaster verified, submitted, and the first round of crawl and markup errors cleared
- Titles and descriptions written inside the limits that are actually enforced — 48 characters before the brand suffix, descriptions 25 to 160, because Bing flags the rest
HOW I WORK
- 01
MAP
Research the clusters, assign one URL per cluster, and list the terms worth ignoring so effort stops leaking into traffic that never converts.
- 02
ARCHITECT
Fix the URL shape, the internal link paths and the canonical rules, including the redirects needed for anything that has to move.
- 03
MARK UP
Write the JSON-LD graph as generated routes rather than pasted blocks, so every new page inherits correct markup automatically.
- 04
SUBMIT
Verify in Search Console and Bing Webmaster, submit the sitemap, watch the first crawls, and clear what the reports flag.
BEST FOR
- Sites with content already published that is not ranking because the architecture buries it
- Businesses with a real inventory of locations, services or products, where programmatic pages have substance to work with
- Brands whose name collides with something else and who are losing that collision in both Google and the assistants
- A site about to triple its page count, where getting the URL rules right now avoids a migration later
NOT THIS
- Not a monthly ranking retainer. I build the structure and hand over the cluster map; I do not run an ongoing link-building or publishing programme.
- Not the speed work. LCP, CLS and INP are a different discipline with different tools, and they live on the performance and Core Web Vitals page.
- Not paid search. Nothing here touches ad accounts, bidding or budget.
THE PROOF
Published case studies from digital experience work. Each one names the client and shows what was delivered.
STRAIGHT ANSWERS
REST OF DIGITAL EXPERIENCE
MAKE A
WISH.
Tell me what you're building. I'll tell you how loud it can get.
- /01
YOU WRITE
Tell me what you're building and what's in the way. Two lines is enough.
- /02
WE TALK
A 30-minute call, booked within 7 days. Real questions, no pitch theatre.
- /03
YOU GET A PLAN
Scope, timeline and an honest number, usually within a week.