On-page SEO
What the bot and the reader actually digest
Templates, internal linking, metadata at scale, cannibalisation, entity coverage. Technical SEO decides whether a page can be reached. On-page decides whether it's worth anything once it is.
Which one you need
Where the line sits
Technical SEO is infrastructure and plumbing — can Google discover, crawl, render and index the URL without choking on server errors, broken scripts or code bloat. On-page is what the bot and the user actually digest once they’re in the door.
Internal linking shows the overlap well. The technical side is making sure links aren’t buried behind JavaScript, that there are no redirect chains, and that high-priority pages don’t end up as crawl-depth orphans. The on-page side is strategic relevance: descriptive, contextual anchor text, directing authority into commercial priority pages and signalling topical relationships.
Technical makes the pipe work. On-page directs the value through it.
Substance
What it covers
Templates
On e-commerce platforms, programmatic directories or large multi-author publications, fixing things one page at a time is impossible.
Template optimisation means restructuring global layouts, improving semantic HTML hierarchy, adjusting where breadcrumbs live, placing dynamic internal linking modules like related categories or contextual product grids, and making sure critical content isn’t buried below three screens of UI.
Fix the template once, and you lift thousands of pages simultaneously.
Metadata at scale
On a site with 50,000 URLs, nobody is hand-crafting 50,000 title tags.
Metadata at scale means engineering dynamic template rules that pull from your database attributes — product name, colour, category — paired with fallback logic so titles don’t end up blank or awkward when a field is empty. It also means automated sanity checks to flag truncations, dynamic duplicate tags and missing descriptions across large catalogues.
Cannibalisation
I find it in Search Console by filtering queries and looking for URLs actively fighting each other. If one query shows three URLs splitting impressions, with rank bouncing between position 8 and position 24, that’s classic cannibalisation. Google doesn’t know which page is the authority.
The fix depends on the pages.
Merge and 301.
If two posts cover the same intent, take the best of both, build one dominant page, and redirect the weaker URL into it.
De-optimise and re-target.
If both pages need to exist — a category hub and a specific sub-guide — rewrite the supporting page’s titles, headers and body copy to target a narrower long-tail query, then link internally back up to the main page.
Canonicalise.
If they’re functionally near-duplicates that need to stay live for UX or filtering, set a clear canonical to the primary version.
Entity coverage
Search engines don’t rank documents on repeated keywords any more. They evaluate how well you cover the real-world concept — the entity — and its associated subtopics.
Write about commercial property insurance and Google expects to see underwriting, public liability, business interruption, premiums, policy limits.
Fixing entity gaps doesn’t mean stuffing in buzzwords. It means restructuring the page: dedicated subheadings addressing missing subtopics, structured FAQs answering real user questions, and structured data to define those entities explicitly to machines.
Your side
What I need from you
Direct CMS access — WordPress, Shopify, or a custom backend. Search Console access. An open line to whoever controls development or staging.
And an honest view of your business priorities: which categories or product lines actually deliver margin, and which are vanity traffic that never converts.
Fit
When this is the wrong thing to buy
If the site has foundational problems that make on-page work irrelevant.
If your site takes twelve seconds to load, your indexing is blocked by a bad robots.txt rule, or the brand has no external authority at all, spending on title tags and internal links won’t move anything.
Fix the burning house before you pick out the wallpaper.
Engagement
How it’s scoped
The baseline audit, template restructuring, metadata logic build and initial cannibalisation cleanup are project work. Once the framework is locked and the legacy mess is cleared, the project wraps.
If the site publishes regularly, introduces new product lines, or scales catalogues weekly, it moves to an ongoing cadence — monitoring keyword drift, auditing new templates, and stopping cannibalisation creeping back in.
Related services