Titles and descriptions are primary tools searchers use to determine the relevance of a search result. Before clicking on a page in SERP, they read through its meta tags to understand what it’s about and whether it’s useful. Only if the page is valuable and matches searchers’ intent, they visit it.

That’s why Google keeps encouraging website owners to create unique and descriptive meta tags.

“Our advice to webmasters has always been to write unique, descriptive page titles (and meta descriptions for the snippets) to describe to searchers what the page is about.”

Unfortunately, many SEO experts neglect this advice and don’t use meta tags as intended. Instead of putting effort into describing a page, they stuff a bunch of keywords into its title and description, hoping to rank better. Some also write vague and irrelevant information, which confuses users.

If Google turned a blind eye to this, its SERPs would become a huge mess. Luckily, the search engine addresses the problem by generating titles and descriptions automatically, often rewriting the meta tags specified by website owners.

Of course, the fact that Google rewrites meta tags is by no means new.

Back in 2014, Brian LaFrance, currently vice president at The Athletic, analyzed 111,000 search results and their title tags. He already knew that Google had been changing titles in SERP, so his goal was to compare the title tags website owners had manually specified to those displayed in search results. Here is what he found:

• 38,6% of titles remained intact.
• 36% were slightly altered: for example, Google truncated them or added brand names to titles.
• 25,4% of titles were completely changed: the generated titles were created with entirely different words.

As for descriptions, a recent study reveals that Google rewrites description meta tags more than 70% of the time.

So why did we write this article if this information is widely known?

The reason is that Google has started changing titles using a completely different approach.

Many SEOs have noticed it, and this topic is currently the subject of heated debate.

We would also like to participate in this and try to clear things up.

So in the following paragraphs, we describe:

• What is the reason Google rewrites meta tags in SERP
• What content from a page the search engine may pull out to generate its title and description
• What you should do about it

Below, we will explain the reason Google rewrites meta tags.

Let’s start with descriptions.

As you probably know, a single article can rank for multiple keywords. If its description was always static, it would be irrelevant to some queries searchers use to find the article.

Let us explain.

Here is a manually specified description of our blog post about Google Sandbox:

“You can’t improve rankings while your website is in Google Sandbox. But you can speed up its release from there. We’ll share six ways to do it.”

However, most of the time, Google just wouldn’t display it. Instead, it changes the description based on a query we used to find that article.

As you can see, Google rewrote the description to match the words from the specified query.

The logic behind it is simple: Google assumes that users will be more likely to consider the result relevant and click on it if the words from the query appear in the description. Consequently, Google rewrites description tags to make them relevant to different queries and thus help website owners attract more visitors.

Not so long ago, Google was using a similar approach when rewriting titles.

Initially, Google was changing title tags to also make them relevant to different keywords. But now, the approach has changed.

Here is an excerpt from a recent Google Search Central blog post that proves it:

“Last week, we introduced a new system of generating titles for web pages. Before this, titles might change based on the query issued. This generally will no longer happen with our new system. This is because we think our new system is producing titles that work better for documents overall, to describe what they are about, regardless of the particular query.”

So when it comes to titles, Google wants them to be relevant to the entire content of a page rather than to a particular keyword. Thus, the title tag of a specific page will remain static regardless of the queries searchers may use to find the page.

As Google reps state, the search engine rarely changes titles, using the information specified in title tags most of the time.

“Of all the ways we generate titles, content from HTML title tags is still by far the most likely used, more than 80% of the time.”

But what happens when Google doesn’t like content from HTML tags? What content from a page may the search engine pull out to generate its title and description? That’s the topic of the following paragraphs.

Experienced SEOs know that descriptions can be pulled out from any piece of text. The difference between them and titles is that description tags are dynamic and often change to match a specific query better, which is no longer the case with title tags.

To generate a relevant description, Google pulls out a passage where the words from the query (or their synonyms) show up. By the way, the search engine uses a similar approach when crafting featured snippets and passage indexing snippets.

But what content does Google pull out when generating titles?

SEOs have begun suspecting that Google is using a new approach to creating title tags after a tweet from Lily Ray, a senior director, SEO & head of organic research at Amsive Digital. She noticed that Google replaced her title with anchor text from an internal link, which never happened before.

Most SEOs knew that the search engine could replace a title with an H1 header, but using anchor text for this purpose was something new.

That raised a reasonable question: if Google can now use anchor text to generate a title, where else can it pull titles from?

Luckily, Google has the answers.

If we get back to the Google Search Central blog post mentioned above, we’ll find the following paragraphs:

“Also, while we’ve gone beyond HTML text to create titles for over a decade, our new system is making even more use of such text. In particular, we are making use of text that humans can visually see when they arrive at a web page. We consider the main visual title or headline shown on a page, content that site owners often place within h1 tags or other header tags, and content that’s large and prominent through the use of style treatments.

Other text contained in the page might be considered, as might be text within links that point at pages.”

So what Google is basically saying is that it can use pretty much any piece of text on a page to generate the page’s title. It can pull out content from:

• H1, H2, H3, H4, and other headers.
• Large and prominent pieces of text (for example, from phrases in bold).
• Anchor text from internal links.
• And even normal text.

But still, if your title is descriptive enough, Google won’t change it.

Now that you know what content from your page Google may use to generate its meta tags, you’re probably asking yourself: should I do something about it? We think you definitely should.

Given that the search engine can pull out any passage from your content to generate meta tags for a page, you should start taking care of the quality of your content.

As mentioned above, Google will change your description most of the time, so it’s reasonable to make every line of your text meaningful and thoughtful. If the search engine pulls out some vague line full of grammatical errors (which sometimes happen), searchers won’t click on your result. So our advice here is to always proofread your content before publishing it.

Pay close attention to the lines with keywords — you should always place keywords naturally. There is no need to break grammar rules by using long-tail terms in the exact match form, as explained in our latest article.

Even though Google frequently changes descriptions, it doesn’t mean you shouldn’t write them at all. Google may still display the tag you specified, especially for your main keyword. And if you don’t want the search engine to rewrite it, follow these simple steps:

• Ensure your meta description is not too long or too short: try to keep it between 50-160 characters.
• Don’t stuff it with a bunch of keywords: placing one main keyword you want to target should be fine.
• Make sure the content of your description tag is descriptive (sorry for the tautology).
• Ensure it doesn’t have any grammatical errors.
• Create a unique description for each page on your site: in case it’s impossible, prioritize your pages and create distinctive descriptions for the critical and popular ones.

As for a title, Google can also pull the information from any content on your page to generate it, so our advice doesn’t change here — ensure every line of your text is meaningful. Pay special attention to your headers, anchor text, and any passage that visually stands out from normal text. That way, even if Google decides to rewrite your tag, you’ll make sure it will be thoughtful.

As we said, Google won’t change your title more than 80% of the time. But it still may happen if the search engine assumes your tag isn’t well-written. To prevent it, follow Google’s guidelines:

• Ensure your title is not too long: try to keep it between 50-60 characters so that the search engine won’t truncate it.
• Don’t put too many keywords in a title.
• Avoid repeated or boilerplate titles. First off, it leads to keyword cannibalization. Secondly, Google doesn’t like such titles and prefers to rewrite them. For this reason, make each title unique and distinctive. In case you have similar product pages with different versions of the same product, add additional attributes to each title — for example, the product’s ID.
• Keep your titles descriptive and concise. Google recommends avoiding vague descriptors like “Home” for your home page or “Profile” for a specific person’s profile.

We understand that it’s hard to keep every title and description tag healthy, especially when you deal with big websites. To help you with such a time-consuming task, we developed Site Auditor. This tool will automatically spot meta tag errors on any site, saving you plenty of time.

In its Meta tags section, you will see the number of pages with:

• empty titles and descriptions;
• too long and too short titles and descriptions;
• duplicate meta tags, and other issues.

You can click on any number to discover the URLs of the pages with a specific meta tag error.

After that, you can visit any page and fix the problem, improving your website.

As you can see from this article, it’s all about relevance. Google remains the most popular search engine because it does a great job providing relevant results to users. And to do so, sometimes it has to rewrite the title and description you diligently wrote for your brand new page.

We understand that it may make you angry or cause distress, but there is nothing you can do here — if you’ve decided to optimize your site for Google, you have to play by its rules.