Keyword Analysis Example: Density, Placement and Gaps

This report leads with density, so start here. Google does not have a keyword density target. There is no ideal percentage, and 1–2% is not a rule — it is a number that circulated in the SEO industry two decades ago and never left.

The deeper problem is that density is a badly-behaved metric, because it is a ratio and you can move it by changing the denominator:

A metric that rewards you for deleting good writing is not measuring quality. "SEO tools" at 1.03% tells you the term appears twenty-four times in two thousand words. That is a fact about word frequency. It is not a diagnosis, and there is nothing to optimise.

Ignore the density column entirely and read the last three. "schema markup" appears fourteen times in the body and is in no title, no H1 and no meta description.

That is a real finding, and it has nothing to do with percentages. It says: this page covers a topic it never announces. A reader scanning the SERP cannot tell the page addresses schema markup. Neither can a crawler assembling a summary of what the page is for. You have done the work and forgotten to put it on the label.

The prominence idea — that terms appearing early carry more weight — survives contact with reality better than density does, though the mechanism is not what people assume.

It is not that Google applies a positional multiplier to the first hundred words. It is that a page which is genuinely about a topic mentions that topic immediately, because that is what writing about something looks like. Prominence is a consequence of relevance, not a cause of it.

Which is why moving the keyword to the first paragraph, without changing anything else, achieves nothing. You have not made the page more about the topic. You have relocated a word.

The genuinely useful version of this: if your primary term does not appear naturally in the opening paragraph, that is a signal the page has not decided what it is about. The fix is editorial — write an opening that states what this page answers — and the prominence follows for free.

Gap analysis is genuinely useful. The way this table frames it is not. Comparing densities — "competitor 1.2%, you 0.1%" — implies the fix is to raise a percentage until it matches. That is imitation, not strategy, and it is how pages end up mentioning "backlink checker" four times in paragraphs that did not need it.

The signal is not the number. It is the absence. "Structured data", "backlink checker" and "redirect chain" are topics your competitor covers and you do not — and the question is not how often should I say this, it is:

1. Should this page cover that topic at all? Often the honest answer is no. Your competitor may cover it because their page is a different page, aimed at a different query. Copying their vocabulary drags your page towards theirs and away from whatever it was best at.
2. If yes — does it deserve a section, or a page? A term worth ranking for is worth a page. A term that genuinely belongs in the story this page is telling deserves a section, written properly, because you have something to say about it.
3. What is the search intent behind it? "Backlink checker" is someone looking for a tool. If your page is a guide, you cannot win that query by mentioning the phrase more often — you are the wrong kind of page, and no amount of vocabulary will fix that.

Add the term only if you can say something true and useful about it. A paragraph inserted to close a gap reads exactly like a paragraph inserted to close a gap.

Search has not matched strings for a long time. It matches meaning — through systems that understand that "how do I make my site faster" and "reduce page load time" are the same question, and that a page can be the best answer to a query without containing the query.

This is why the terms you never thought about are the ones you rank for. Search Console will show you queries you did not target and would never have chosen — and they are frequently your best traffic, because they are specific, and someone with a specific question found a page that answered it.

And this is the point where SEO and AEO converge. An answer engine assembling a response is looking for a passage that directly answers a question. It is not counting your keywords — it is looking for whether you actually said something. A page written to hit a density target has, almost by definition, nothing to quote.

What is keyword density?

Keyword density is the percentage of times a keyword appears relative to total word count. It is not a ranking factor and Google has no density target — the familiar 1–2% figure is an SEO-industry convention from two decades ago, not a Google guideline. It is also a badly-behaved metric, since deleting half your content doubles it while adding a useful new section lowers it. Treat it as a description of word frequency, not a diagnosis.

What is keyword prominence?

Keyword prominence measures how early in the page content a keyword appears. Keywords in the title, first H1, first paragraph and first 100 words carry stronger ranking signals than keywords buried further down the page. The Keyword Analyser scores each keyword by its prominence position.

What is a keyword gap?

A keyword gap is a term that a competitor ranks for but your page does not mention or target. Keyword gap analysis compares your page against a competitor URL to identify terms you are missing. Adding these terms as natural mentions or FAQ answers can increase your topical coverage and ranking potential.

Is there an ideal keyword density I should aim for?

No. Google does not have a density target, and the metric is a ratio you can move by changing the denominator — delete half your content and it doubles, add a thorough new section and it falls. A metric that rewards you for deleting good writing is not measuring quality. There is nothing to optimise here; the number is a fact about word frequency, not a finding.

My keyword is not in my title. Should I add it?

If the page is genuinely about that topic, yes — a page that covers a subject it never announces cannot be identified as being about it, by a reader scanning the SERP or by a crawler summarising it. But if you have eight significant topics and cannot fit them in sixty characters, the title is not the problem: the page is trying to be eight pages, and a page about everything ranks for nothing because there is no query it is the best answer to. The honest fix is usually to split it.

Should I add every keyword my competitor ranks for?

No. Comparing densities implies the fix is to raise a percentage until it matches, which is imitation rather than strategy. The useful signal is the absence, not the number — and the questions are whether this page should cover that topic at all (often it should not; your competitor's page may be aimed at a different query), whether it deserves a section or its own page, and what the search intent behind it is. If someone searching "backlink checker" wants a tool and your page is a guide, no amount of vocabulary will win that query — you are the wrong kind of page.

If not keywords, what should I optimise for?

Topics. Search has matched meaning rather than strings for a long time, which is why the terms you never targeted are frequently the ones you rank for — and why those queries are often your best traffic. Ask what someone with this problem actually needs to know, then answer all of it: the follow-up, the objection, the thing they will get wrong, the case where the standard advice fails. The vocabulary takes care of itself, because you cannot write thoroughly about a subject without using its words. This is also where SEO and AEO converge: an answer engine is looking for whether you said something quotable, and a page written to hit a density target has nothing to quote.