Home / SEO / On-Page SEO
On-Page SEO
On-page SEO is the work of making each page unambiguously the best answer to one specific thing people search for. Most sites fail at this not through neglect but through accumulation — pages added over years, each reasonable on its own, several now competing for the same query and splitting the signal between them.
Overview
One page, one intent. Map keywords deliberately so your pages stop competing with each other.
Keyword mapping is a decision, not a list
A keyword research document that lists a thousand terms is not a strategy. The output that matters is an assignment: this page targets this query, and no other page may target it.
That constraint is what prevents cannibalisation. When three pages address the same intent, search engines have to choose between them, links and internal signals divide across all three, and the one that gets picked is often not the one you would have chosen. Consolidating those into a single authoritative page — and redirecting the others rather than leaving them to compete — routinely produces movement faster than writing anything new.
Intent, then depth
Before writing anything, we look at what currently ranks for the target query, because that is search engines telling you what they believe the searcher wants. If the results are all comparison pages, a product page will not win the query no matter how well optimised it is.
Depth is then measured against those results rather than against a word count target. The question is not whether the page is two thousand words; it is whether it covers the sub-questions the ranking pages cover, and adds something they do not. Word count as a target produces padding, and padding is precisely what the last several years of ranking system updates have been aimed at.
Structure that machines and people both read
Headings are the outline of the argument, not a place to repeat keywords. A page whose H2s state the actual sub-topics is easier to skim, easier to parse, and considerably more likely to be pulled into a featured snippet or an AI-generated answer.
We also handle the elements that decide whether a ranking becomes a visit. Title tags written to earn the click rather than to hit a character count. Meta descriptions that pre-qualify, so the wrong people do not click and bounce. Schema markup where it is genuinely eligible rather than sprinkled everywhere, since invalid markup is a liability and irrelevant markup does nothing.
Internal linking is the lever people forget
Internal links are the one authority signal you control completely, and most sites use them almost by accident — navigation and footer links repeated on every page, with almost nothing contextual in the body copy.
We build a deliberate structure: important pages linked from more places, descriptive anchor text that says what the destination is about, and links placed inside body content where they carry more weight than another navigation repeat. On sites with existing content this is frequently the highest-return work available, because it needs no new pages and no new links from anyone else.
What's included
The scope of the engagement, stated plainly so there is nothing to discover later.
- Full keyword to page mapping with one search intent per page
- Cannibalisation audit with consolidation and redirect plan
- SERP intent analysis for every commercially important target query
- Content depth assessment measured against currently ranking pages
- Title tag and meta description rewrites focused on click-through
- Heading structure rebuilt as a readable outline
- Schema markup where genuinely eligible, validated before deployment
- Internal linking architecture with descriptive anchor text
- Content briefs for new pages, and rewrite specs for existing ones
How we run it
The order matters more than the individual tasks. Doing these out of sequence is what wastes months.
Inventory and map
Every indexable page catalogued against the query it should own. This is where cannibalisation surfaces, and it is usually worse than expected.
Consolidate
Competing pages merged or redirected before anything new is written. Adding content to a site that is already competing with itself makes the problem larger.
Rewrite and build
Existing pages brought up to the standard the query demands, then gaps filled with new pages. Existing pages first, because they already have history and links.
Link internally
The contextual link structure built last, once the destination pages are worth linking to. Pointing authority at a weak page just wastes it.
What you get
Concrete artefacts you keep, whether or not the engagement continues.
- Keyword to page map covering the full site
- Cannibalisation report with a specific consolidation plan
- Rewritten titles and meta descriptions for priority pages
- Content briefs with intent analysis and depth requirements
- Internal linking plan with source pages and anchor text specified
- Schema markup implemented and validated
Common questions
The questions that come up most often on discovery calls.
How long should a page be?
Long enough to answer the question completely and no longer. We measure against what currently ranks — if the top results all cover eight sub-topics, a page covering three will struggle regardless of length. Writing to a word count reliably produces padding, and padding is what recent ranking updates have targeted most directly.
What is keyword cannibalisation and how do we know if we have it?
It is when multiple pages on your site compete for the same query, splitting signals so none ranks as well as one strong page would. The quickest check is Search Console: if the URL ranking for a given query keeps changing, or several of your URLs appear for it, you have it. It is very common on sites with a few years of accumulated blog content.
Should we optimise old content or write new content?
Old content first, almost always. Existing pages have crawl history, internal links and sometimes external links already pointing at them. Improving a page that ranks on page two is usually faster and cheaper than getting a brand-new page to the same place.
Do meta descriptions affect rankings?
Not directly — they are not a ranking factor. They affect click-through rate, which affects how much traffic a given ranking actually delivers. A page at position four with a compelling description can outperform position three with a poor one, so they are worth writing properly even though they do not move the ranking itself.
Does adding schema markup improve rankings?
Not on its own. Schema helps search engines understand the page and can make it eligible for richer result formats, which affects visibility and clicks. Adding markup that does not describe the page accurately is worse than adding none, so we only implement types the content genuinely qualifies for.
Related work
This sits inside our SEO practice.
Want this done properly?
Start with a discovery call. We will tell you whether on-page seo is actually your bottleneck, or whether something else should come first.