Glossary
Meta description
A meta description is a short summary of a page, placed in the HTML head, that search engines may use as the snippet beneath a result link. It is not a ranking factor and has not been one for many years. What it influences is whether someone clicks, and search engines use it only when they judge it a better description of the page than text they could extract from the page itself.
When it is used and when it is not
Google documents that it uses the meta description when it describes the page better than an extract would, and generates a snippet from the page content otherwise.
Snippets are frequently generated per query: a page that answers several questions gets a different extract depending on which was asked, which no static description can match.
A missing description is not an error. The search engine generates one, and often generates a good one.
A generic or duplicated description across many pages is more likely to be ignored than a specific one, so the effort is only repaid where it is specific.
What makes one worth writing
It describes what is on this page specifically, in terms someone scanning a list of results would recognise.
It reads as a sentence rather than a list of keywords. Keyword stuffing here achieves nothing, since there is no ranking effect to gain.
It is honest about the content. A description that oversells produces clicks that leave immediately.
Length matters only in that snippets are truncated by width. Roughly 150 to 160 characters usually survives on desktop, and less on mobile, but no limit is defined anywhere.
Where it is worth the effort
Pages that receive substantial search traffic and where the automatic snippet is poor — typically pages whose useful summary is not near the top of the content.
Commercial and category pages, where the description is doing the work of a short advertisement.
Pages whose content is a list or a table, from which an extracted snippet reads badly.
It is not worth the effort on very large sites to write one per page by hand. A good template is better than a duplicated generic sentence, and no description at all is better than a bad template.
How to check them
`curl -s https://example.com/page | grep -i 'name="description"'` reads the tag from the served HTML.
Across a site, look for missing tags, duplicated text across many URLs, and descriptions that are only a few words long.
Duplication is the finding that matters most, because it means one sentence is being offered as the summary of many different pages.
Compare against what actually appears in search results for the page. A description consistently ignored is a description that is not doing its job.
Frequently asked questions
- Is the meta description a ranking factor?
- No. It has not been used for ranking for many years. It can affect click-through, which is a different thing.
- What happens if I have no meta description?
- The search engine generates a snippet from the page content, which is frequently a good result and is often what happens even when a description exists.
- How long should a meta description be?
- No limit is defined. Snippets truncate by width, so around 150 to 160 characters typically survives on desktop and less on mobile.
Sources
Related
VeriFixScan crawls a site and applies its checks to every page it reaches, keeping the evidence behind each finding. Scanning one website is free.
Scan a website