You wrote a title tag, published the page, and then searched for it and saw something different in the result. Maybe a word is missing, maybe the brand name moved, or maybe the whole phrase is one you never wrote. This is Google rewriting title tags, and it happens often enough that it's worth learning to read the signs. This piece helps you work out whether a rewrite is actually happening, find the cause on your page, and change the tag so Google is more likely to keep it as written.
The short answer is that Google generates the clickable title link for every result, and your HTML title tag is one important input rather than an instruction. That means no edit can guarantee Google shows your tag word for word. What you can do is find out what Google is seeing, remove whatever gives it a reason to swap your wording, and give it time to recrawl the page.
First, check that it really is a rewrite
Before you attempt any title tag rewrite fix, make sure you're looking at a rewrite and not something simpler. A lot of "title tag not showing" reports turn out to be truncation, or a page Google hasn't recrawled since you edited it.
Start by finding the result for the exact URL you care about, and write down three things: the title Google displays, the search you used, and the device. Then compare that with the real HTML title tag on that URL. Use the tag itself, not the headline in your CMS editor and not the browser tab, since a template can change what gets published.
If the displayed title is your tag with an ellipsis at the end, or with the last few words missing, that's most likely truncation. Google says title links are shortened as needed to fit the width of the device. A different phrase, a product name pulled from your heading, or an old year swapped for a new one is much stronger evidence that Google generated different wording.
Next, check what Google has actually seen. In Google Search Console, paste the full URL into the URL Inspection tool search bar. The default view shows the indexed version of the page, which isn't always the same as today's live page. Look at the last crawl date, the Google-selected canonical, and the crawled HTML. If the Google-selected canonical is a different URL, the page you're inspecting may not be the page whose result you're looking at, so sort that out first.
Then check today's rendered version separately, using the option to test the live URL and view the tested page HTML. In Chrome, Inspect shows the rendered HTML, while View source usually shows the code as it was first returned, before scripts run. If a CMS, a template, or a script might be changing the title, compare the indexed version and the live version side by side.
Two cautions apply. If you edited the tag after the last crawl date, Google may still be working from the older version, and that's a timing issue before it's anything else. And Search Console tells you which URL and which HTML Google knows about, but it doesn't show you the exact title link served for each search, so you still need to look at the real result to see the SERP title mismatch for yourself.
How Google picks a title
When Google rewriting title tags looks random, it helps to know what it is choosing from, because most of the fixes below come down to giving it fewer reasons to look past your tag. According to Google's guidance on title links, it can draw on the HTML title element, the visible main title on the page, headings, the og:title tag, other prominent text, anchor text on the page, text in links pointing to the page, and WebSite structured data. Its stated goal is to represent and describe the result.
In August 2021 Google said its updated system would generally stop changing titles according to the query, which is something the earlier system sometimes did. Its current advice still asks publishers to think about whether the alternative title Google chose fits the query better than theirs. So it's not accurate to say every search produces its own rewrite, and it's not accurate to say the query never matters either.
There's also no single honest rate to quote. Google said in September 2021 that its system used the HTML title element around 87 percent of the time. Ahrefs looked at 953,276 top-ten pages in November 2021 and found about a third had a different title in the results, and Zyppy's early 2022 study of 2,370 sites found a much higher share, though it noted many changes were minor. They measure different samples, years, and definitions of a mismatch, so don't average them. What they show is that rewrites are common and plenty of pages keep their tag. If you wonder whether SEO fundamentals still matter as AI answers grow, that piece covers it.
Cause 1: The title is too long, or is just cut off
What it looks like. The result starts with your intended wording and ends early, sometimes with an ellipsis. Or Google skips your tag and picks a shorter phrase from the page.
How to confirm it. Open the full HTML title in the crawled or live HTML in Search Console, then compare it with the displayed result on the same kind of device. Work out whether the extra words were hidden because there was no room, or whether Google replaced them with something else. Also look for a long repeated brand phrase or a run of trailing modifiers using up the space.
The fix. If it's truncation, cut the wording you can live without and keep the description short enough to be read in one glance. Google gives no fixed character limit for a title element, and the visible width depends on the device. That means you shouldn't promise yourself that a 60-character tag, or a set width in pixels, will force the exact display, and moving one keyword to the front won't do it either.
The studies do point to length as something worth watching. Ahrefs reported that titles over 600 pixels were rewritten about 46 percent of the time after Google's 2021 update, compared with roughly 29 percent before. Zyppy saw far higher rewrite rates for titles over 60 characters than for shorter ones. Both are correlations in particular samples, and neither is a rule from Google, so treat length as a reason to tighten the tag and not as a magic number.
Cause 2: The template produced a missing, vague, or repeated title
What it looks like. The title reads something like "| Site Name," "Home," or "Profile." Or lots of pages share the same tag, or pages in a series leave out the one thing that tells them apart, such as a season number. Google fills the gap, often with a more specific heading from the page.
How to confirm it. Check the rendered title tag on the affected URL, then check a handful of related URLs. You're looking for an empty template variable, one generic title repeated on every page, or near-identical titles across one section of the site. Then compare each page's heading with its tag.
The fix. Repair the template variable, and give each page an accurate phrase that separates it from its neighbors. Google has a name for repetition across a subset of pages, micro-boilerplate, and its own examples show a missing product name being filled in from prominent content and season numbers being added to otherwise identical episode titles. So this is a template repair, not a reason to generate a new keyword variation for every URL, which leads straight into the next cause.
Cause 3: The title repeats keywords or promises things the page doesn't deliver
What it looks like. The same keyword shows up in several variations, or the tag names products, topics, or features that aren't on the page. Google shows a broader or more accurate phrase instead.
How to confirm it. Read the actual title tag next to the main heading and the main content. Ask whether each repeated term adds something useful, and whether a visitor could find every specific thing the title promises. Google's own example is a title listing several kinds of stuffed animals for a page that no longer sells them.
The fix. Take out the repeated terms and any claim the page can't back up. Say what the page covers today, and then make sure the visible main title and the body support that description. Google discourages keyword stuffing, and swapping it for a shorter string of synonyms doesn't fix a mismatch between the tag and the page.
Cause 4: The title is out of date and the page has moved on
What it looks like. The tag advertises an old year or a previous edition, while a prominent headline on the page names the current one, and Google's title link uses the newer wording.
How to confirm it. Compare the title element, the visible heading, and the page content. This comes up most on URLs that get reused every year and on dynamic product or category pages.
The fix. Update the tag so it matches the page as it is now, and keep the headline consistent with it. Google's documented example changes "2020 admissions criteria - University of Awesome" to "2021 admissions criteria - University of Awesome" when the main heading says 2021. Those years are Google's illustration, not guidance about a real page. Don't change the year on a legitimate historical page just to look fresh, since that puts you back in Cause 3. If you're updating many old posts at once, this is how to refresh old SEO articles without losing rankings. If old posts are the real problem across your site, the same review can go into a content audit to find decaying pages.
Cause 5: A prominent heading is competing with the tag
What it looks like. The title link closely matches your H1, or another large heading, instead of the HTML title. Sometimes a page has several headings that look equally important, and it isn't clear which one is the main title.
How to confirm it. Compare the result with the H1, the other big headings, and the og:title. Look at the rendered page and not only the unrendered source, and check whether the heading you meant to be the main one really is the most distinctive and prominent thing at the top.
The fix. Make the true page heading clear, usually as the first visible H1, and reconcile any real differences between that heading, the tag, and what's on the page. Google doesn't require the H1 and the title to be identical, so exact matching is not the goal. Alignment helps, but copying a weak tag into the H1 doesn't cure anything.
Ahrefs found that in the cases where the displayed title differed from the tag, the results matched the wording of the H1 about half the time, at 50.76 percent. That's an observation and not proof that Google took the title from the H1, so use it as a prompt to check your headings.
Cause 6: The brand or site name is repetitive or inconsistent
What it looks like. Google removes the site name from the end of the title, changes how the brand appears, or adds one when your tag has very little in it. Sometimes a site name is already shown separately in the result, which can make a second copy in the title link look redundant.
How to confirm it. Compare the title tag, the site identity visible on the page, and the displayed result. Check whether every page repeats a long homepage-style slogan, whether a template adds a brand that is missing or inconsistent, and whether the only difference is that Google dropped a duplicate site name.
The fix. Keep the brand short. Google's example allows more explanatory text on a homepage, and suggests using just the site name at the start or the end of other pages, with a delimiter such as a colon, hyphen, or pipe. Google may leave out a domain-level site name if it already appears elsewhere in the result, so don't count on any one delimiter to keep the brand in place. Zyppy saw brand names added and removed, but Google has published no rule that the brand will always be kept or dropped.
Cause 7: The tag's language, or Google's access to the page, doesn't match what you expect
What it looks like. Google picks title text in the language or writing system of the main page content and not the language of your tag. Or the generated title looks like anchor text from other sites, because Google couldn't get at the page itself.
How to confirm it. Compare the language and alphabet of the tag with the page's main content. Then use URL Inspection to check that crawling, fetching, and indexing are allowed, and that Google could really retrieve the page you meant.
The fix. Match the tag to the main language and writing system of the page. Clear away any crawl barriers you didn't mean to set up, so Google can look at the page itself. A technical SEO checklist covers the other crawl and render basics. A robots.txt block can stop crawling without necessarily stopping indexing, and in that case Google says it may rely on text from other places for the title link. If you're not sure how your file is set up, this guide to configuring robots.txt walks through the rules. Please don't add noindex to fix a rewrite, since that directive is for keeping a page out of the index and has nothing to do with the wording of the title.
When none of these matched
If you've gone through all seven and found no match, go back over the basics. Make sure the result is for the canonical URL you meant. Make sure Google has recrawled and reprocessed the changed page. Make sure the indexed HTML, the live rendered HTML, the visible heading, and the content all agree with each other. Then look at the og:title and the prominent text, and ask whether the title Google chose describes the page better for that search than yours does.
If it still looks wrong after all that, take the exact URL, the search, the tag you intended, the title you see, and what URL Inspection told you to the Google Search Central Help Community and ask for feedback. Google says it cannot manually change the title links for an individual site, and there is no supported override that guarantees the exact string.
Once you've made a fix on a single page, you can ask for a recrawl. An owner or full user of the Search Console property can request indexing in URL Inspection after checking the live page. For many changed URLs, submit a sitemap instead. Google's page on how to ask Google to recrawl your site says crawling can take a few days to a few weeks, that asking again for the same URL won't make it faster, and that a request guarantees neither indexing nor a particular displayed title. So recheck the result after the new crawl, and don't read an unchanged result the same afternoon as proof that the fix failed.
If your team also reports on drops in clicks, a title that has been swapped for a weaker one is one of the causes worth ruling out when you diagnose a CTR-driven traffic drop. A wider organic traffic drop framework built on Search Console shows where this check fits.
What to do this week
Pick the five pages where the result matters most, or use an AI workflow that flags low-performing pages to find them, the ones that bring in leads or already get a fair amount of search traffic. For each one, run the checks in the first section, then find the cause that matches and make that one title tag rewrite fix. Request indexing on each page once the change is live, and put a reminder in your calendar to look again in two or three weeks.
The order of the causes above isn't a ranking by how often each one happens. Google doesn't publish how common each is, and the two studies measure particular title characteristics and not the causes overall. I put the ones that are easy to spot and confirm first so you can rule them out quickly.
If you publish at volume and want titles and metadata to be written along with every article, DeepSmith's 7-day free trial is one way to see how that works in practice. No tool can dictate what Google shows, so the checks here still apply to every page.
