Why does Google rewrite your title tags and how can you stop it?

What exactly does a “global site title” mean?

Google analyzes all your titles to identify a general pattern that characterizes your site. This overall approach allows the algorithm to detect repetitions, generic formulations, or titles that do not reflect the actual content of the page. If your titles all display “MySite | Category”, Google understands that only the word “Category” truly varies.

The engine then compares this structure with the actual content of each page, H1 headings, and internal and external link anchors pointing to the URL. When it detects a mismatch between your title and what it considers the true theme of the page, it reserves the right to rewrite. This is not a bug; it has been a feature for years.

Why does title similarity matter?

Titles that are too similar create semantic confusion for the algorithm. Google no longer knows which page to prioritize for a given query if two or three pages have almost identical titles. The result: probable internal cannibalization and dilution of your perceived relevance.

The confusion also impacts your users in the SERPs. If three of your URLs appear with similar titles, the click-through rate drops mechanically. Users cannot discern which page truly answers their query. Google indirectly penalizes this poor experience by favoring competitors with clearer titles.

What constitutes a “descriptive” title for Google?

A descriptive title accurately reflects the unique content of the page. Avoid generic formulas like “Welcome” or “Home”. Google wants to read the subject matter immediately, ideally with vocabulary that matches target queries. If your page is about “boiler repair in Paris 15”, your title should include those terms, not just “Our Services”.

Descriptiveness also implies the absence of over-optimization. A title stuffed with repeated keywords or artificial variations will be perceived as spam. Google prefers a natural, readable title that invites clicks while being informative. Striking the right balance can be tricky, but it’s exactly what an SEO title must achieve.

• Strict uniqueness: no title should be duplicated on your site, even partially.
• Content reflection: the title must correspond to the H1, subtitles, and main body text.
• Controlled length: between 50 and 60 characters to avoid truncation in the SERPs.
• Strategic keywords: include priority terms at the beginning of the title without keyword stuffing.
• Positioned brand: place the brand name at the end of the title unless that is what is being searched.

Does this recommendation align with field observations?

Yes, substantially. Audits show that Google rewrites titles in 60 to 70% of cases on poorly optimized sites. Often, the engine pulls from the H1, a popular link anchor, or a visible text snippet. When your titles are truly unique and descriptive, the rewrite rate drops dramatically, sometimes below 10%. The figures confirm Mueller's statement.

However, be cautious: even a perfectly optimized title can be rewritten if Google believes another formulation improves CTR for a specific query. This is rare, but it does happen. The engine sometimes tests variations and keeps the one that performs best. You never have absolute control, but you can drastically reduce unwanted rewrites.

What nuances should be added to this statement?

Mueller mentions “confusion” but does not specify the similarity threshold that triggers rewriting. Does having the brand name as a suffix on all pages pose a problem? Probably not if the variable part is genuinely descriptive. However, titles like “Product 1 | MySite”, “Product 2 | MySite” with a poor semantic corpus on each page are problematic. [To be verified]: Google has never published specific metrics on this point.

Another nuance: the statement suggests that a unique title is sufficient, but the reality is multifactorial. A unique title on a thin content page, with an H1 different from the title and few backlinks, will still be susceptible to being rewritten. The uniqueness of the title is a necessary condition, but not sufficient. The overall content of the page, its relevance, and authority also play a role.

When does this rule not fully apply?

On high-authority sites, Google is more tolerant of shorter or more branded titles. If you're Amazon, a title “Product Name” without a long description passes without issue. For an average site, it would be rewritten. The brand itself becomes a signal of relevance and trust that compensates for less descriptive titles.

Multilingual or multi-regional sites also pose challenges. Pages that are nearly identical in content but translated can have structurally similar titles. Google generally understands the linguistic context, but rewrites sometimes occur. Again, the solution remains the same: maximize semantic differentiation in each language, not just translate word for word.

What concrete steps should be taken to avoid rewrites?

Start with a thorough audit of your titles using Screaming Frog, Sitebulb, or Google Search Console. Identify exact duplicates, near-duplicates, and titles that are too short or too long. Export everything to a spreadsheet and segment by page type: products, categories, blog, institutional pages. You'll quickly see problematic patterns.

Next, rewrite each title following this structure: main keyword + unique differentiator + brand. The differentiator is what makes your page unique: the exact product model, the targeted city, the year, or the type of problem solved. Never settle for “Our Products” or “Services”. Be surgical. Test your formulations in a SERP simulator to check readability and avoid truncation.

What mistakes should be absolutely avoided?

The first mistake is to duplicate the H1 entirely in the title. They should be consistent, but not identical. The title is visible in the SERPs, the H1 is on the page. Use this opportunity to slightly vary the phrasing or order of terms. Google detects this redundancy and may consider you lack creativity or semantic relevance.

Another pitfall is automated formulas like “[Category] | [Subcategory] | [Brand]”. If your subcategories are not very varied or if the content does not justify this granularity, you artificially create similarity. Prefer manual or semi-automated formulas with truly distinctive variables. A poorly configured CMS generates thousands of nearly identical titles without your notice.

How to check if your titles are accepted by Google?

Use the “site:” operator in Google targeting specific URLs. Compare the title displayed in the results with the one declared in your source code. If Google shows something else, it’s rewriting. Test this on a representative sample of strategic pages, not just the homepage.

Google Search Console also provides a “Pages” report where you can cross-reference indexed URLs with their displayed titles. Unfortunately, GSC does not directly flag rewrites, but you can export the data and cross-reference it with a crawl. If you notice systematic discrepancies, that’s a red flag. Correct, wait for recrawl, and check again. Iteration is essential.

• Audit all titles via an SEO crawler and export to a spreadsheet.
• Identify and eliminate all exact duplicates and near-duplicates.
• Rewrite each title with a unique differentiator and strategic keywords.
• Check for consistency between titles, H1, and content of each page.
• Test rewritten titles in a SERP simulator to control truncation.
• Monitor rewrites using the “site:” operator and Google Search Console after going live.