Top queries not visible in page content

SEO Perception reads the actual page we crawled and checks whether the Search Console queries that land on that URL appear in the content. The same check shows up in two places: a Content match label on query tables, and a Search Console possibility when important queries are missing from the page.

What we compare

For each query we look for the phrase (and simple space/hyphen variants) in the crawled page:

Title
Meta description
Headings
Body text
Internal link URLs and anchor text
Image alt text

A hit in any of those counts as in content. A query only counts as missing when it is absent from all of them. Google search operators such as inurl: are ignored, because they are not phrases a page would contain. This is a text match against the crawl, not a guess from an AI model and not a live call to Google.

Where it shows in the workspace

On the website Queries tab, each row has a Content match column. The filter next to it is All terms, Found in content, or Not in content. On a single page, the Pages → Queries list shows the same result as an icon beside the query. Hover it for the explanation.

In content — the query was found in the title, meta description, headings, body, internal links, or image alt text from the last crawl.
Not in content — that crawl did not find the query in any of those places. The possibility opens from these misses on the page with the most clicks for the query.
Unknown — we have not confirmed a hit or a miss yet. The phrase may only be in body text we have not scored, or the match has not been evaluated.
Crawl pending — the HTML crawl for that URL is not finished, so the match is not reliable yet.

Use Not in content when you want the queries people already search that the landing page never states. See What you see on a website.

The possibility

The same comparison is a Search Console possibility: Search queries not reflected in visible content. It is one open item per indexable page (HTTP 200, not noindex) when stored Search Console data exists for that URL.

We look at the page’s top queries by clicks. Queries with at least one click that appear nowhere on the crawled page are listed on the possibility—up to five, so the row stays readable. Queries that contain a term saved under Exclude queries containing these terms on the website Overview are left out. When several URLs have clicks for the same query, the possibility opens on the URL with the most clicks. The other URLs stay on the cannibalization row. The detail shows each query with clicks, impressions, and average position when we have it. Fixing the copy and recrawling clears the item when those queries are present.

A related check, query alignment with title and meta description, is different: the phrase can already be in the body, and we still flag it when it is missing from the title or meta description.

What to do

Update headings and body copy so the page answers the intents behind those queries—without keyword stuffing. If two URLs split the same intent, merge or differentiate them instead of repeating the phrase on both.

Example

If people search for “vegan leather tote” and that phrase is not in the title, meta description, headings, body, internal links, or image alt text, the Queries row is Not in content and the page can get the possibility. Weave the idea in where it is true:

<main>

<h1>Leather tote bag</h1>
<h2>Vegan leather option</h2>
<p>
We offer a vegan leather tote with the same silhouette as our classic bag—
ideal if you want a cruelty-free tote for work or travel.
</p>
</main>

Technical details

Requires stored Search Console facts for the URL; low-traffic properties may have sparse data. The match is computed from the crawl and stored with the page, then read by the Queries tables and by this possibility. See Search Console data in SEO Perception and How we find possibilities.