en
/login
Koval SEO

The page-specific part of a title is missing

In the reportHalf-empty title

half_empty_title · Review item

Short answer

This appears when a page's title is only the site name its other titles share, or starts or ends with a separator that has nothing on one side, such as | Example Bakery. Add the words that say what this page is, and keep the site name beside them. The homepage is not checked, because a homepage titled with the site name alone can be intentional.

Severity: Medium. Plan it with your next round of title changes; this finding alone does not mean the page is missing from search.

Why it matters

Google lists half-empty titles, such as <title>| Site Name</title>, among the problems that can lead it to build a page's title link from other text on the page. A title that names the page gives it a clearer source, and tells a visitor which page they are about to open.

What to do

  1. Open the page's title or SEO-title field and add what this page is about before or after the site name, for example About our bakers | Example Bakery.
  2. If the field is already filled but the published title is still half-empty, ask your developer to check the template that joins the page's title to the site name.
  3. If several pages show this, fix the shared template first: one empty field in a template can leave the same half-empty title on every page built from it.

When to leave it as it is: if the text we took for your site name is really a word most of your titles share, such as Blog on a site where most titles end in | Blog, a page titled with that word alone may already say what it is. The finding shows the shared text we inferred, so you can tell which case it is.

What we read

We read the first <title> in the HTML each page returned, without running JavaScript. We compare pages that answered 200 with HTML, carry no noindex and either declare no canonical or name themselves as canonical. Pages with no title are left to the missing-title check. Pages that are only a frame for scripts, with almost no text of their own, are left out; pages where scripts add part of the content are still checked from the HTML we read.

A title counts as half-empty in two cases. It starts or ends with a separator — |, :, ·, », ›, –, —, or a hyphen set apart by a space — with nothing on its other side. Or it is exactly the text that opens or closes most of your titles, normally the site name: we take a text as shared when it is at least two characters long and stands first or last in titles made of two or more parts, on at least three pages and at least 60% of the titled pages compared. The homepage is not checked.

Example & evidence

Illustrative example — one page among six, not a live audit.

Address: https://bakery.example/about/

Of the six titled pages, the homepage is titled Example Bakery, four others have their own words before | Example Bakery: Bread, Cakes, Menu and Contact. The title we read on this page:

<title>| Example Bakery</title>

It starts with a separator and has nothing before it. Example Bakery is the inferred site name, so the finding shows this title and that shared text.

A possible correction:

<title>About our bakers | Example Bakery</title>

With the page's own words before the separator, neither case applies. A title of Example Bakery alone would still be half-empty on this page.

Confirm & check again

Check for yourself: open the page source and find the <title>. If it is only your site name, or begins or ends with a separator, the page-specific part is missing; compare it with the title field in your website editor.

Check again: Full audit — run a new audit of the site. The finding closes when the page is read again as a complete HTML page and its title has a page-specific part. If the new audit can no longer infer a shared site name from your titles, it cannot confirm the change on a page that has a title.

Limits & sources

The site name is worked out from repeated title segments, not read from your settings or structured data, so a word most titles share can be taken for it. The check reads the HTML title only: it does not see the title link Google shows, and it judges only pages in this audit's crawl.

Confidence: High when the pages compared were read in full; pages that failed to load or are only a frame for scripts are left out, and the result says its coverage was partial.

Sources

Editorial review: 29 September 2026