A title tag is the title element in the head of a page. A meta description is the meta name="description" element beside it. People often talk about both as if they were the blue link and the gray text in Google. They are not. Google’s documentation on influencing title links says a title link is generated from several sources, and the title element is one of them. Google’s documentation on snippets says the snippet is created mainly from the page, and the meta description is used when it describes the page better than that text. The SEO content brief is where you write the promise before the draft. This page is what happens to that promise after the URL is public.
Write one clear promise, and put it in the HTML you send. The title element and the visible H1 should name the same page. The meta description should summarize that page in a sentence a person can use. Google may still show a different title or a sentence from the article. A character counter does not decide which one.
What a title tag is
The title element is a preference. Google’s title-link page says title links are often the main thing a person uses to decide which result to open, so the text should be descriptive and concise. Every page should have one. “Home,” “Profile,” and “Blog post” are the examples Google gives of titles that do not say what the page is. A long title is not forbidden. The same page says there is no limit on how long a title element can be, and that the title link is truncated to fit the device.
Keyword stuffing is called out separately. A few descriptive words can help. Repeating the same phrase in different spellings, the way “standup, stand-up, standups, stand up guide” does, does not help the person choosing a result, and Google says it can look spammy. The title is a label for one URL. It is not a place to list the cluster, the year, the brand slogan, and five modifiers. Discover uses that same title as preview content. A misleading or exaggerated headline is the pattern in Google’s Discover documentation, and a title that shocks instead of summarizing the page is a helpful content question. The title here is still the label for search.
Boilerplate is the other failure Google names. If every article on the site uses “Northwind — see guides, templates, checklists, and examples,” the words that vary are too small for anyone to tell the pages apart. Brand the home page with a fuller line if you want. On articles, a short site name at the start or the end, separated by a hyphen, colon, or pipe, is the pattern Google shows. That suffix is not the name Google shows beside the result. The site name is the home-page label. The distinctive part of the title is the page, not the slogan.
Write the title in the same language and writing system as the page. If the article is in Hindi, the title element should be in Hindi. Google says it tries to show a title link that matches the primary language of the page, and it may choose different text when the title element does not. A title that is only a Latin-letter transliteration of a page written in another script is one of the documented reasons for a replacement. Hreflang does not detect that language. It only names the alternate URLs, which is the hreflang guide. If you keep a separate mobile URL, that title and the meta description should match the desktop page. The mobile-first indexing guide is the rest of that split. A responsive page has one title.
What a meta description is
A meta description is a summary that is not shown on the page itself. That is why it gets neglected, and why a bad one is easy to ship. Google says it will sometimes use the element as the snippet if that summary is more accurate than what it would pull from the body. The snippet is still automatic. Different queries can produce different snippets from the same URL, because the snippet is meant to preview the part of the page that fits that search.
There is no published character limit. The snippet is truncated to fit the device, the same way the title link is. Guides that tell you the description must be 150 to 160 characters are describing a common phone preview, not a rule on Google’s snippet page. A sentence that says what the page decides is a better target than a sentence padded until a counter turns green.
Unique descriptions matter more than length. Google says identical or near-identical descriptions on every page are not helpful when several of those pages appear together. Write a page-level description for the articles people actually open. A site-level line belongs on the home page. If you cannot write one for every URL, write them for the home page and the pages that already get queries, not for a tag archive that should not be a result. That second case is a robots meta tag decision, not a description you still need to perfect. A paginated sequence is the exception Google names: those pages may share a title and a description. Unrelated articles may not.
What to put in the sentence: the job of the page, and a concrete detail a person can use before they click. Google’s examples for posts include the author or the publication date when that detail is relevant. A product page might gather a price or a location that is otherwise scattered. A keyword list is the example Google marks as bad, because it does not say what the page is. “Async standup, standup template, remote team, agile, scrum” is that failure. “How a six-person team runs an async standup in one thread, and when a live meeting is still the right call” is a description of a page.
Google also says a programmatic description can be appropriate on a large database site, if the result is human-readable and different for each URL. Swapping one keyword into a fixed sentence is the boilerplate problem again. An article that was written as an article can carry a one-sentence summary written with it. That is the case for a blog.
What Google may show instead
Treat both fields as proposals. The title link people click, and the snippet under it, are chosen when the result is built.
A rewritten title is not, by itself, a penalty. Google’s page lists the situations it adjusts for, and it also says the title in the results might simply be a better fit for the query. If you changed the element yesterday, Google has to recrawl and reprocess the page. The same documentation says that can take a few days to a few weeks. Requesting a recrawl does not force the old title back. It asks Google to look at the new sources.
The meta description has the same kind of gap. You can write a careful sentence and still see a paragraph from the article under the result, because that paragraph matched the query. That is a reason to make the opening of the article a real answer. If the only good summary lives in the meta description, and the page opens with a story that never states the point, Google has little on-page text worth using, and an answer engine has the same problem. The description does not repair a page that hides the answer.
Where the title link comes from
Google lists the sources it uses when it builds a title link. They are not ranked as a score you can add up. They are the places the system may look.
- The title element.
- The main visual title on the page.
- Heading elements, especially a clear H1.
- The
og:titlemeta tag. - Other text that is large or styled as a title.
- Other text on the page.
- Anchor text on the page, and anchor text in links that point at the page.
- WebSite structured data, which is a site-level source for the site name, not a place to hide a keyword.
If Google cannot crawl the page, it falls back to off-page text such as other people’s anchors. That happens when robots.txt disallows the URL. The title-link documentation says a disallow does not always keep the URL out of the index, and the title may then be whatever other sites called the link. Preventing indexing is a noindex rule, which only works if the page can be crawled. The robots meta guide is that decision. Do not disallow an article to “control the title.”
Anchors are a source, so the words you use in internal links can show up as a title if the page’s own title is weak. That is not a reason to freeze one exact phrase on every link. The topic cluster guide already says to describe the destination and vary the wording. A descriptive anchor helps a person. A stuffed anchor, repeated sitewide, is how a bad title spreads.
Why a title gets replaced
Google publishes the common problems. Matching your title to one of them is more useful than blaming a pixel width.
| What Google describes | What it looks like on a blog | What to change |
|---|---|---|
| Half-empty title | The element is only the site name, or a separator and nothing else | Name the page. Keep the site name short, if you keep it. |
| Obsolete title | The title still says 2020 and the H1 says the current procedure | Update the title with the page. A content refresh that changes the H1 and leaves the old title is this row. |
| Inaccurate title | The title lists topics the article does not actually cover | Describe the page you published, not the cluster you hoped to rank for. |
| Boilerplate | Twenty posts share one title pattern and differ by a single word | Put the distinctive promise first. Drop the repeated slogan. |
| No clear main title | Two headings on the page are the same size | One H1, visually first and larger than the H2s. |
| Language mismatch | The article is in one language and the title is in another | Write the title in the language of the article. |
| Repeated site name | The brand is in the title and Google already shows the site name on the result | Google may omit the extra copy. You do not need to add it twice to compensate. |
Flight prices are a special case on Google’s list: prices in titles for flights are likely to be dropped because they go stale. A blog about standups does not have that problem. A title that leads with a number you did not measure has a different problem, covered in citing sources. Do not put an invented statistic in the title so the result looks specific.
Length, without a fake limit
Tools that flag a title over 60 characters, or a description over 160, are showing you a rough phone preview. Google’s own pages say there is no limit, and that truncation follows the device. A title that is three screens long will be cut off, and it is also harder to read in a browser tab. Concise is still the instruction. The instruction is not “stop at character 60 even if the promise is unfinished,” and it is not “add words until you reach 60.”
Write the title so the page is identifiable in the first few words. Those are the words that survive truncation, and they are the words a person scans. “How to run an async standup” does that. “The complete ultimate 2026 guide to everything about async standups for remote teams” spends the visible space on words that every competing result also uses.
The same habit applies to the description. One or two sentences. The decision or the limit, not a list of keywords and not a copy of the title. If the description and the title say the same seven words, you have used the snippet to repeat the headline. Say what the page adds: who it is for, or the choice it recommends.
Pixel width changes. A title that fits your laptop can wrap on a phone. You will not win that by chasing the preview in three tools. You win it by being short enough that the promise is still there after a cut.
The H1 and og:title
Google looks at the main visual title and at headings when it builds the title link, and it says multiple headings with the same visual weight are confusing. The fix it describes is practical: one main title, larger, and the first obvious heading. On an article that is the H1. The H2s are sections. They should not be styled like a second H1.
The H1 and the title element do not have to be character-for-character identical. The title element can add a short site name. The H1 can be a little more like a sentence, because it has the page around it. They do have to be the same promise. If the title says “Async standup template” and the H1 says “Why status meetings run long,” you have labeled two pages. Google may pick the H1. A reader who clicked the template title and landed on an essay was misled even if Google keeps your title.
og:title is on the source list. It is the title some social posts use. If it is a different pitch from the title element, you have added another candidate. Set it to the same promise. Do not stuff it with a longer keyword variant “for Facebook.” Open Graph does not replace the title element, and it does not replace the H1.
The slug is a third label, and it is not a title. Keep it short and readable. Changing it later is a redirect, which the canonical URL guide separates from a tag. Do not put the title’s full punctuation into the path.
What answer engines actually read
Answer engines quote text they can retrieve from the page. The meta description is easy for them to miss, because many of them prefer the body, and because Google’s own snippet system already prefers the body unless the description is better. If the only place you state the answer is the description, you have hidden it from the person on the page and from a system that extracts passages.
Put the answer in the first paragraph, in HTML. The title can name that answer. The description can summarize it. The paragraph has to say it. That is the same habit as the answer engine guide: a passage that stands alone. A title written for a counter, sitting on a page that never repeats the promise, gives an overview nothing solid to cite.
Do not use a nosnippet rule on an article you want quoted. That rule is for pages whose text should not become the snippet, and Google’s robots documentation says it also applies to AI Overviews and AI Mode. It is the wrong switch for a guide. It is explained on the robots meta page so this one can stay on the words you want shown.
“Read more” links inside a snippet, when Google shows them, depend on the section being visible in the HTML, not tucked in a control that starts closed, and not fought by a script that jumps the scroll position on load. Google’s snippet documentation says that directly. An article that is readable without extra script is easier to snippet and faster to paint. The title element itself is a few words in the head. It does not need a library. Adding a script whose only job is to rewrite the title after load creates a second version of the title. Put the final title in the HTML response.
A title tag example
Northwind’s standup procedure was first published with this title element: “Async Standup Guide | Best Standup Practices | Remote Team Tips | Northwind.” The H1 was “How to run an async standup.” The meta description was “async standup, standup template, scrum, kanban, remote work, best practices 2020.” Search showed the H1, not the title element. The description in the result was the second paragraph of the article, which happened to mention a six-person team, because the meta description was a keyword list.
The replacement title element is “How to run an async standup — Northwind.” The H1 stays “How to run an async standup.” The meta description is “A six-person team runs the standup in one thread: three prompts, a cutoff time, and the one case that still needs a call.” og:title matches the title element. Nothing in the title repeats “standup” four times. The year is not in the title, because the page is a procedure, not an annual report. If the procedure changes, the refresh updates the sentence that changed, and the title only if the promise changed.
They do not add a second article titled “Async standup best practices 2026” to chase the words that used to be stuffed into the title. Those words were not a separate job. A second URL would be the overlap the content audit is meant to catch.
How this shows up in a BloGoose draft
BloGoose sends meta_title and meta_description with the article. They should be the promise and the summary, not a keyword string generated beside an article that says something else. When you review a draft, read the title, the H1, and the first paragraph together. If a stranger could not tell they belong to one page, send it back. The description should be a sentence you would not be embarrassed to see under the result, including if Google ignores it and lifts the opening paragraph instead. That is the test: the opening has to be good enough to be the snippet.
The title is decided with the brief, before the draft pads it. A model that is told to “include the primary keyword, three secondary keywords, and the brand” will produce the stuffed title Google’s documentation warns against. Tell it the promise in ordinary words, and a maximum that means “short enough to identify the page,” not a superstition about character 60. Then read what it wrote.
Questions about title tags
What is a title tag?
A title tag is the text in the title element of a page. It is one source Google may use for the title link in search results. It is not a guarantee of the words people will see, and it is not the same thing as the visible H1.
What is a meta description?
A meta description is an HTML element with a short summary of the page. Google may use it as the snippet under the title link when it describes the page better than the on-page text. Google often builds the snippet from the page itself, and the snippet can change with the query.
Is there a title tag character limit?
Google’s title-link documentation says there is no limit on how long a title element can be. The title link is truncated to fit the device. A concise title is still easier to read. A character count from an SEO tool is a preview, not a rule Google publishes.
Does the meta description affect rankings?
The meta description’s job is the snippet. Google uses it when it is a better description than the page text, and a clear snippet can earn the click. Stuffing it with keywords does not make the page a better answer, and Google is less likely to show a description that is only a keyword list.
Why did Google change my title?
Google may replace the title element when it sees a common problem: the title is half empty, outdated, inaccurate, boilerplate, in a different language or writing system than the page, or the page has no single obvious heading. Google can also choose a title that fits the query better. Check the visible H1 before you assume the character count was the cause.
Should the title tag match the H1?
They should describe the same page. Google looks at both, and at other prominent text. If the title element says one thing and the H1 says another, Google may pick the heading. Make the H1 the obvious main title on the page, and write the title element as the same promise, plus a short site name if you want it.
Write the promise once, in the title, the H1, and the opening paragraph.
Start the 1-day trial