What Is a Title Tag in SEO?
The most important single element on a page, though not the controlled output most guides present it as. Google frequently shows something other than the title you wrote, which changes how everything else here should be read.
What The Page Says About Itself
These tags are a page's own account of what it is. Some are read and used, some are read and treated as a suggestion, then several are ignored entirely. Knowing which is which saves a great deal of pointless work.
Where they live. Out of sight.
In a part of the page a visitor never sees, which is why they are easy to neglect and easy to get wrong without anybody noticing.
The three categories. Used, considered, ignored.
The title and the indexing instructions are acted on. The description is considered and frequently replaced. Most of the rest do nothing at all, which block seven covers.
Why the distinction matters. Effort allocation.
Businesses spend time on tags that have no effect because a plugin provides a field for them. A field existing is not evidence that anything reads it.
What none of them can do. Change what the page is.
These describe a page rather than improve it. A poor page with an excellent title is a poor page that has been accurately labelled, which is worth remembering before spending an afternoon here.
What this page will not contain. Syntax or counts.
No code, plus no character count stated as a rule. Block three explains why the second of those is not merely our preference.
Title Tags
The title element states what a page is called. It is the single most important on-page element, doing two different jobs that people frequently conflate.
The first job. Describing the page.
It tells a search engine what the page is about, in the page's own words. That description is used in understanding the page regardless of what is eventually displayed.
The second job. Being the headline.
It is the basis of the clickable headline somebody sees in results. That is where the pressure to make it appealing comes from, plus where the tension in block four begins.
Why it outranks everything else here. It is the summary.
Every other tag on this page is supplementary. The title is the page's one-line statement of its own subject, which makes accuracy here worth more than anywhere else.
Where else it appears. Several places.
Browser tabs, bookmarks, shared links and anywhere else something needs a short name for the page. Those uses are unaffected by anything a search engine does.
What it is not. The heading on the page.
The title element and the visible heading are different things that may reasonably differ. The headings pillar covers the relationship, since it is the question everybody actually has.
It Gets Rewritten
A search engine frequently substitutes its own headline when it judges yours a poor match. That is normal rather than a fault. It means the tag is an input rather than an output.
What Google published. A dated change.
In August 2021, Google announced a change to how it generates the headline shown in results, which its documentation calls the title link. It states it may generate an alternative using visible content on the page where the supplied title is unsuitable.
The distinction that matters. Two different things.
The title element on your page and the headline shown in results are separate. A rewrite changes what is displayed and leaves your title element exactly as written.
Why that is reassuring. It is not a ranking event.
Your title element is still what gets used in understanding the page. A rewrite is a rendering decision rather than a judgement about the page's quality.
What triggers one. Documented reasons.
Titles that do not describe the page accurately, repeated boilerplate that fails to distinguish similar pages, stuffing, then pages with no clear main heading. Google publishes these as things to avoid.
What you can do about it. Improve the input.
Google has stated it cannot change title links for individual sites on request, adding that following its published practices improves the quality of what is shown. Influence rather than control.
Why this is not a crisis. A rewrite is usually a signal.
A consistently replaced title is telling you the page's own description was not the best available. That is information rather than an insult. Block four is what to do with it.
Writing Title Tags
Describe the page accurately first and appeal second. That order is not a moral position, it is what reduces the chance of the title being replaced.
The primary test. Does it describe this page.
Somebody reading only the title should know what they will find. A title describing a page adjacent to this one is the commonest cause of a substitution.
Distinctiveness. Different from your other pages.
Titles differing only by a word, across pages that are genuinely different, fail to help anybody choose. That is the boilerplate problem Google names directly.
Why we state no character count. There is not one.
Google publishes no official limit. What gets shown depends on rendering rather than on counting characters, so a target number is a convention people have adopted rather than a rule anybody set.
The useful version of that guidance. Front-load it.
Long titles get cut off somewhere, so the important part should come first. That achieves what the count rule was reaching for without pretending to a precision nobody has.
Where the business name belongs. Usually at the end.
Useful for recognition and rarely the most important thing about the page. On a page somebody is searching for by name, that reverses.
What to avoid. Stacking phrases.
A title assembled from every term you hope to be found for reads badly to a person and is one of the documented triggers for replacement. It fails on both counts at once.
Meta Descriptions
The description offered for the snippet beneath your result. It is not a ranking factor, it is frequently replaced, yet it still matters because it affects whether somebody chooses you.
What it is for. The summary under the headline.
A short account of what the page contains, offered for display. Purely a presentation element, with no bearing on how the page is assessed.
Why it is not a ranking factor. Long settled.
It has not been used in ranking for many years and nothing has changed that. Anybody presenting description writing as a ranking activity is describing something that does not happen.
Why it is replaced so often. Relevance to the search.
Where a passage of your page answers the particular search better than your written summary, that passage gets shown instead. One description cannot suit every search a page answers.
Why it still matters. The click.
When yours is used, it is what a person reads before deciding. Being seen and not chosen achieves nothing, which is the whole subject of the click through pillar.
Which pages deserve the effort. The ones people reach.
Writing descriptions for pages nobody finds is work with no possible return. Concentrate on the pages already appearing, which is a theme the audit pillar develops.
Writing Descriptions For Click Through
Our own standard is plain and descriptive, with no hooks and no statistics. That is a deliberate choice rather than an oversight. It is worth stating because it differs from common advice.
What the standard is. Say what the page contains.
A direct account of what somebody will find, in the words they would use. No question openers, no promises of secrets and no numbers inserted for effect.
Why we chose it. The reader is deciding about a supplier.
Somebody comparing businesses is assessing whether to trust one. A result reading like a headline written to be irresistible works against that, whatever it does to click rates elsewhere.
Why we are flagging it. It differs from the usual advice.
A great deal of guidance recommends techniques we deliberately do not use. Stating that lets you weigh the argument rather than assume we simply had not heard of them.
The one universal rule. It must be true.
A description promising something the page does not deliver produces a visit and a departure. That is worse than not being clicked, which the click through pillar argues properly.
Where the full argument sits. Its own page.
Click through rate covers which techniques genuinely work, where they work and why we still avoid them.
Meta Data Generally
Most historical meta tags do nothing. The keywords tag in particular has had no effect for many years. It is still being filled in on sites we are asked to look at.
What the term covers. An umbrella.
Any of the descriptive tags a page carries about itself. The word makes them sound like a single system. They are an accumulation of things added at different times for different reasons.
The keywords tag. Genuinely nothing.
A list of terms a page wanted to be found for. It was abused immediately and abandoned as a signal long ago. Filling it in achieves precisely nothing today.
Why it persists. Software still offers the field.
Plugins and platforms still present a keywords box, so people conclude something must read it. The field is a legacy of the interface rather than evidence of a function.
Other tags with no current effect. Several.
Author tags, revisit instructions, distribution and rating declarations. All present on older sites, all doing nothing, none worth removing urgently.
What does still function. A short list.
The title, the description, indexing instructions and the tags controlling sharing previews. Everything on this page is one of those, which is why the list of what to attend to is short.
How to tell whether one functions. Look for the documentation.
Anything currently used is documented by the search engine that uses it. A tag whose only support is blog posts recommending it, with nothing from the people who would read it, is a convention rather than a mechanism.
Why this is worth saying. It is sold.
Meta tag optimisation appears in proposals as a deliverable. A business paying for keywords tags to be completed is paying for something with no possible outcome.
Open Graph
These tags control how a page appears when shared on social platforms and in messaging apps. They have nothing to do with how it appears in search, which is the confusion worth clearing up.
What they control. The sharing preview.
The image, headline and summary shown when somebody pastes your link into a social platform or a message. Without them, the platform improvises, usually badly.
Why the confusion arises. They look identical.
They sit alongside the search tags, carry similar names and describe the same things. Nothing about their appearance signals that a different audience reads them.
What happens without them. An unappealing preview.
A shared link with no image, a truncated headline or a fragment of navigation as the summary. That is what somebody's contacts see, which affects whether anybody follows it.
Where it matters most. Anything shared privately.
Links sent between people in messages, which is a substantial amount of sharing and entirely invisible in reporting. A poor preview costs you there without leaving a trace.
What it does not do. Affect search at all.
No search engine uses these to decide anything. A business told that Open Graph tags will help rankings has been told something untrue.
Whether they are worth setting. Yes, once.
Set sensibly at template level, they need no ongoing attention. That is a small piece of work with a lasting return.
Meta Robots And Noindex
These instruct a search engine how to treat the page. The critical distinction is that blocking a page from being fetched and telling a search engine not to list it are different instruments with different effects.
What these tags instruct. Treatment, not access.
Whether to list the page, whether to follow links from it, plus related instructions. They sit on the page itself and are read when the page is fetched.
The distinction from the robots file. Fetching against listing.
Our technical material covers a file that asks crawlers not to fetch parts of a site. That is a different operation: it controls access rather than appearance, so a blocked page can still be listed.
Why that catches people. The wrong instrument.
Somebody asked to keep a page out of search reaches for the file whose name suggests exclusion. The reasoning is sound and the instrument is wrong. The result is an address listed with no description.
The self-defeating combination. Doing both.
Blocking a page while also instructing that it not be listed prevents the second instruction from being read at all. The page must be fetchable for the instruction on it to have any effect.
When listing should be prevented. A short list.
Thank you pages, internal search results, duplicate filtered views and anything genuinely not intended for strangers. Fewer pages than most people apply it to.
What we are not publishing. The syntax.
How either instruction is written is specified and a developer has it. The decision about which pages deserve it is the part needing somebody who knows the business.
The Noindex Mistake
An instruction not to list pages, applied across a development site and shipped to the live one at launch. The site disappears from search, nothing on it looks any different, then checking takes seconds.
How it happens. Correctly, then carelessly.
A development site is kept out of search deliberately, which is right. At launch the templates are copied across and the instruction travels with them. Nobody added it to the live site, so nobody thinks to remove it.
What makes it worse than the alternative. It is obeyed precisely.
Unlike a blocked page, which may still be listed, this instruction is honoured. Pages already listed are removed, so an established site loses presence it had rather than failing to gain any.
Why nobody notices. A working site.
Every page loads, the design is right, the enquiry form works. Decline appears over weeks by which point attention has moved on. Nothing connects it to the launch.
Who usually finds it. Somebody arriving later.
An agency taking over months afterwards, checking basics as a matter of course. That is an uncomfortable conversation for everybody involved in the original launch.
Its counterpart. The robots file.
Our technical material describes the same failure with a different instrument. Both are invisible on the live site and both are checked in under a minute, which is why site migrations opens with both.
What to do. Check at launch.
On the day a site goes live, confirm no page carries an instruction against being listed. One check, immediately, on a sample of each template.
Check for a noindex
the day you go
live.
An instruction not to list pages, carried over from a development site, removes an established site from search. Every page loads, the design is right, the form works, then presence declines over weeks with nothing connecting it to the launch.
What we set on every page:
We state no character count, because Google publishes no official limit.
Every guide.
One practice.
What a page says about itself, headings, keywords, search intent, images, page types, user behaviour, credibility and how to review an existing site.
