SEO Services: Technical Foundations, Architecture and Content
Which page cannot be crawled, which heading answers which query and which technical blocker is holding rankings back — written down instead of guessed at.
We treat SEO as machine readability rather than a keyword list. Before a page can rank it has to be crawled, indexed and understood as the answer to a particular question. If any one of those three steps is broken, work built on keyword choice is wasted.
The pattern we see most often on inherited sites is this: the content exists, but redirect chains, duplicate addresses, missing or contradictory canonical tags and text hidden behind JavaScript mean the search engine never sees half of it. Those blockers get cleared before any new content is written; otherwise every new article hits the same wall.
SEO and GEO are not the same job, but they share the same foundations. This page covers the classic search engine side: crawling, architecture, speed, internal linking and Search Console. Being cited by generative engines is separate work with its own page; when both run together, the same technical fixes pay off on two channels at once.
What this service covers
The order is not fixed; the state of the site decides it. Starting content work on a site that cannot be crawled is like arranging furniture in front of a locked door. The items below make up a full engagement.
Technical SEO audit and crawl issues
We crawl the site the way a search engine sees it: status codes, redirect chains, duplicate addresses, canonical conflicts, robots and sitemap consistency, content stuck behind JavaScript. The output is not a score but a fix list ordered by impact.
Site architecture and internal linking
How pages link to each other decides which page is treated as the centre of a topic. We map topic clusters, connect orphaned pages and, where two pages target the same query, decide which one stays.
Page speed and Core Web Vitals
We find the source of slowness by measuring it: image weights, font loading, third-party scripts, render-blocking resources. We work from measurement, not assumption; on this site, dropping the Google Fonts connection alone brought first contentful paint down noticeably.
On-page optimization and query mapping
We put every important query into a table against the page that answers it. Heading hierarchy, page title and description, image alt text and structured data follow from that map. Queries with no matching page become the case for a new one.
Search Console setup and reporting
Property verification, sitemap submission, index coverage monitoring and query-level reporting are put in place. The monthly report is not a screenshot of rankings: which page shows for which query, where clicks are being lost and what we look at next month.
How we work through it
Timing depends on the size of the site, its technical debt and how competitive the sector is. The schedule below describes a typical start; it is not a commitment.
- Stage 1 — first week
Crawl and technical audit
We crawl the site end to end and review Search Console and analytics data. The output of this stage is a written findings list, with an estimate of how much work each item is.
- Stage 2 — weeks 2-4
Clearing the blockers
Whatever stops crawling and indexing gets fixed first: broken redirects, duplicate addresses, missing sitemap entries, server errors. We do not move on to content production before this stage is finished.
- Stage 3 — months 1-3
Architecture and query mapping
Topic clusters are built, the internal link network is completed, missing pages are opened and pages that cannibalise each other are merged. In this period we watch coverage and click trends more than rankings.
- Stage 4 — ongoing
Measurement, upkeep and expansion
Search Console data is read on a regular cycle, declining pages are revisited and new query opportunities go into the page plan. Every change is logged with its date so a movement can be traced back to a cause.
What we do not promise
Trust in SEO usually breaks on a ranking promise nobody could keep. We write the limits down at the start:
- We do not commit to a specific position for a specific keyword; Google sets rankings and we only fix the variables we can control.
- We do not give a date for results. A technical blocker being cleared can show quickly, while architecture and content work takes months to register.
- We do not buy links and we advise against buying them; the practice breaks Google’s own guidelines and the risk stays with the business.
- We do not mass-generate content with AI and publish it. Tools can be used while drafting, but every page that goes live is checked by a person.
- If the product or service itself has no demand, SEO will not make up for it; in that case we suggest revisiting the offer rather than opening more pages.
What we measure
The position of a single keyword says nothing about the customers a business won. The list is deliberately short:
- How many pages are indexed and what share of pages report errors in Search Console.
- Impressions, clicks and click-through rate by query, read as a trend rather than a single month.
- Which page appears for which query, and whether two pages are competing on the same one.
- The direction of Core Web Vitals field data and the gap between mobile and desktop.
- Forms, calls and WhatsApp starts that originate from organic search.
Frequently Asked Questions
Let us crawl your site the way a search engine does
We work out where you stand on crawling, indexing, speed and query mapping, and hand back a fix list ordered by impact.
Get in touch