Coverage, depth, focus: a model for measuring content quality

"Improve the content" is the least actionable instruction in SEO. It points at everything and nothing. The page is too thin, or too broad, or missing something — but which, and where, and worth how much? A single quality score doesn't answer that. It just compresses the problem into one number you can't act on.

The fix is to break content quality into parts that each map to a different edit. That's what coverage, depth, and focus do. Together they describe how well a page satisfies the intents behind its real demand. Apart, each one tells you a specific thing to change.

Here's the model, component by component, with what each looks like and how to fix it.

The problem with one number

A page scores 64. Is it 64 because it's missing whole topics? Because it covers everything but answers each thinly? Because it's bloated with off-topic filler that drowns the real answer? All three produce mediocre scores, and all three need opposite fixes — add sections, deepen sections, or cut sections.

That's why the number alone is a dead end. Splitting it into three components turns the score into a diagnosis. Below, each component sits on a clear definition and a clear remedy.

Coverage: how many intents you address at all

Coverage asks: of the distinct intents in your real demand, how many does the page address in any form? Low coverage means searchers arrive with a question the page simply never takes up.

Example. A page on "email deliverability" ranks broadly and, in Search Console, receives impressions for sender authentication, bounce handling, spam-folder placement, and warming up a new domain. The page covers the first three well. It never mentions domain warming. That's an entire intent — drawing real impressions — with zero coverage. The searcher who came for it leaves immediately.

How to fix it. Add a section that answers the missing intent. Coverage gaps are the most straightforward to close because the diagnosis is binary: the topic is there or it isn't. Pull the page's query list, group it into intents (see finding content gaps in Search Console), and check each intent against the page's actual sections. Every intent with demand and no section is a coverage fix waiting.

Depth: how completely you answer what you cover

Depth asks: for the intents the page does cover, how completely does it answer them? A page can technically address an intent and still leave the searcher wanting. Depth catches the difference between a mention and an answer.

Example. Back to the email-deliverability page. It "covers" pricing of deliverability tools in a single sentence: "Most deliverability tools cost between $50 and $300 a month." The demand behind that intent — dozens of impressions for queries like "deliverability tool pricing comparison," "cheapest deliverability monitoring" — clearly wants a breakdown: tiers, what each unlocks, where the cutoffs are. One sentence is coverage without depth. The intent is technically present and practically unanswered.

How to fix it. Expand the thin section into a real answer. Depth fixes are about completeness relative to what the demand wants — a comparison intent wants a comparison, a how-to intent wants steps. The signal for "how deep is deep enough" comes from the queries themselves: the more specific and varied the queries behind an intent, the more the page needs to fully resolve it rather than gesture at it.

Focus: how little off-intent noise dilutes the page

Focus is the one most people forget, because it runs the opposite direction from "add more." Focus asks: how much of the page is off-intent — content that doesn't serve any of the real demand and dilutes the answers that do? It's the inverse of over-serving, and it has its own failure mode. More on that in why over-optimization and focus matter.

Example. That same page, in an attempt to "be comprehensive," opens with 600 words on the history of email and SMTP protocol internals — content no query in its demand asks for. The off-intent intro pushes the real answers below the fold, weakens the page's topical signal, and makes both readers and crawlers work harder to find what they came for. The page isn't missing anything; it's burying what it has under noise.

How to fix it. Cut or split. If the off-intent material serves no demand, remove it. If it serves a different intent that deserves its own page, split it out. A focused page answers its real intents and stops. The goal isn't shorter for its own sake — it's a high ratio of on-intent content to total content.

Why the three together beat any single score

Each component points at a different, opposite action:

• Low coverage → add sections for the missing intents.
• Low depth → expand the thin ones into full answers.
• Low focus → cut or split the off-intent noise.

A page scoring 64 on coverage with strong depth needs new sections. A page scoring 64 on depth with strong coverage needs expansion, not new topics. A page scoring 64 on focus needs subtraction. Same headline number, three completely different worklists — which is the entire argument for splitting it.

Prioritize the fixes by what's at stake

The three components tell you what kind of fix each gap needs. They don't tell you which to do first. That's where impressions come in.

Rank every gap by the traffic at stake — the impression volume behind the underserved intent. A coverage gap on a 3,000-impression intent outranks a depth gap on a 200-impression one, regardless of which is "easier." Combine the two views — what kind of fix from the components, which one first from the impressions — and the content audit stops being a wall of suggestions and becomes an ordered list of edits, each tied to a specific intent and a specific amount of demand.

That's the deliverable worth producing: not a score, but a ranked set of concrete changes, each justified by the demand behind it.