Meta Description Checker

Check a single page or crawl an entire site from its sitemap to find every missing, duplicate, too short, too long or truncated meta description.

Check your meta descriptions

One page or a whole site

Paste any page address. We read the live page and report the meta description exactly as Google would see it.

What a meta description checker is for

A meta description is the couple of lines of text under your link in a search result. It is not a ranking factor and has not been one for many years, which is exactly why it gets neglected. But it is the advert that decides whether the ranking you worked for turns into a visit, and on most sites it is quietly broken in the same handful of ways: missing on the pages nobody remembered, identical across a hundred product pages because a template wrote them, or thirty words too long so the interesting half never appears.

This tool finds those pages. Give it one address and it reads that page and reports its description the way Google would render it. Give it a bare domain and it finds the sitemap, reads every page listed in it, and reports the whole lot in one table you can filter and export.

Why the whole-site check matters

Almost every free meta description checker takes a single URL. That is fine when you already know which page is wrong, and useless when the question is “which of my four hundred pages are wrong”. To answer that most people are pushed towards a paid site audit or a desktop crawler they have to install and configure.

There is no good reason for that to cost money. A sitemap already lists the pages, and reading a page to pull one tag out of the head is cheap. So the site check here is free, needs no account, and runs from the address bar: type the domain, wait, read the table. It covers up to 500 pages from up to 50 sitemap files in a run, reads them a few at a time so nothing times out, and can be cancelled at any point while keeping whatever it has already found.

Measured in pixels, because that is how Google cuts

Advice about description length is almost always given in characters: keep it under 155, or 160, or 120 on mobile. Those numbers are averages of a thing that is not measured in characters at all. Google fills a line of fixed width and stops when it runs out of room, so a description full of capitals and wide letters is cut well before one made of narrow ones, at the same character count.

The figures this tool uses were read off a live results page rather than copied from an article. Desktop gives a snippet two lines of a 652 pixel column. A phone gives it three lines of 343 pixels. Every description is measured against both, because the two cut in different places and a description that survives a laptop can still lose its ending on a phone. When something is going to be cut, the preview cuts it at exactly the word Google would, and adds the same ellipsis.

Everything it flags

  • Missing. No description tag at all, so Google will lift a sentence from the page instead.
  • Open Graph only. The page has an og:description but no plain name="description". Social networks are happy; Google still has nothing to use. This is a common side effect of a social plugin filling in half the job.
  • Two tags. More than one description tag in the head. Google picks one and you do not get a say.
  • Duplicate. The same text on more than one page, grouped so you can see whether it is a pair or a pattern.
  • Cut off. Longer than the space desktop gives it.
  • Cut on mobile. Fits a laptop, loses its ending on a phone.
  • Too short. Well under the available space, which is a wasted chance to be persuasive.
  • Copies the title. The description repeats the title word for word and says nothing new.
  • Contains markup. Stray HTML or an unresolved entity, which will be printed as written.
  • Ends in an ellipsis. Usually the fingerprint of a description generated by chopping the first paragraph of the page.

How to write one worth clicking

Be specific. “Learn about running shoes” tells a reader nothing they had not assumed from the title. “We tested 42 pairs over six months; these five suit new runners best, from $55” tells them what they get, how much work went into it and what it costs. Numbers, dates and prices survive scanning in a way adjectives do not.

Front-load it, because if it is cut, the end is what goes. Put the reason to click in the first sentence and treat the second as a bonus that may never be seen. Write a different one for every page, even if that means writing a short one: a unique short description beats a long one shared with fifty siblings. And do not stuff keywords into it. The description is not weighed for ranking, so a keyword list buys you nothing and costs you the persuasion.

What to do with the report

Work down the filters rather than the whole list. Start with Missing, because those pages are currently letting Google write their advert. Then Duplicate, which is usually a template problem and often fixable in one place rather than fifty. Then Cut off, which is normally a matter of moving the good sentence to the front. Too short is the lowest priority and the one to leave until you have time to write rather than patch.

Export the filtered table as a CSV and it becomes a work list: address, current description, length and the faults found, ready to hand to whoever is rewriting them.

Where this fits with the other tools

  • Once you have rewritten a description, drop it into the SERP Snippet Preview Tool to see the finished result drawn out, with the title and the address, on desktop, tablet and mobile.
  • If the title needs work too, the Blog Title Analyzer scores a headline out of 100 and lets you compare several.
  • The Word Counter has a character limit preset for a 160 character meta description, which is handy while you are drafting.
  • To check the page behind the snippet actually covers the term you are targeting, run it through the Keyword Density Checker.
  • No sitemap to check against? Build one with the XML Sitemap Generator, or pull the topics out of an existing one with the XML Sitemap Keyword Extractor.

Is anything stored?

No. Reading a page means asking our server to fetch that address, because your browser is not allowed to fetch other people’s pages directly. Nothing about the result is written to a database, kept in a log or stored on your device: the report lives in the browser tab and is gone when you close it. Export the CSV before you leave if you want to keep it.

What it checks

Most free checkers take one address and count the characters. This one takes a whole site and measures the pixels.

Your whole site, not one page

Enter a bare domain and we find the sitemap, read every page it lists and report all the descriptions in one table. No account, no desktop crawler, no checking pages one at a time.

Pixels, not characters

Google cuts a snippet when the line runs out of room, not at a character count. Every description is measured against the real space: two lines of a 652px column on desktop, three lines of 343px on a phone.

Duplicates across pages

The same description on twenty pages tells a searcher nothing about which result to click. Every repeat is grouped and flagged, which is normally a paid feature.

The faults that hide

Two description tags on one page, an og:description with no real description behind it, a description that just repeats the title, stray HTML, or copy that already ends mid-sentence.

Seen as Google draws it

Single pages come with a desktop and a mobile snippet rendered at Google's own type sizes and column widths, cut at exactly the word Google would cut.

Filter and export

Narrow the table to just the missing ones, the duplicates or the ones being cut off, then take the lot away as a CSV to work through.

Questions about meta descriptions

How long should a meta description be?

Long enough to fill the space and short enough to survive it. Desktop gives you two lines of a 652 pixel column, which is about 155 to 160 characters of ordinary English. A phone gives you three narrower lines, which works out about the same. Writing to roughly 150 characters keeps you safe on both. The character figure is only a guide though: wide letters eat the line faster, which is why this tool reports the pixel width as well.

Do meta descriptions affect rankings?

Not directly, and Google has said so for years. They matter because they are your advert in the results. Two pages can rank in the same place and get very different traffic depending on which one gives a searcher a reason to click. Treat the description as copywriting rather than as a tag to fill in, and treat a missing one as a wasted slot rather than a penalty.

Why does Google ignore the description I wrote?

Google rewrites descriptions on a large share of results, usually because it has found a passage on the page that matches the search more closely than your tag does. That is not a fault to fix so much as a signal: if your description is routinely ignored, it is often because it describes the page in general terms while searchers are arriving with specific questions. Descriptions that are unique, specific and genuinely about the page get replaced less often.

Is a duplicate description really a problem?

It is not a penalty, but it is a wasted opportunity and often a symptom. If fifty product pages share one description, either the template is writing them or nobody has been near them. Google will usually replace a duplicate with its own text, so you have handed over control of your own snippet. The site report groups every repeat so you can see at a glance whether it is two pages or two hundred.

An og:description is not enough?

No. og:description is read by social networks and messaging apps when they build a link card. Google builds its snippet from the plain name="description" tag. Plenty of sites have one and not the other, usually because a social plugin filled in the Open Graph tags and nothing filled in the standard one, so the tool reports that case separately.

How many pages can the site check cover?

Up to 500 pages in a run, collected from up to 50 sitemap files. Each page is a separate request, read a few at a time so nothing times out, and you can cancel at any point and keep whatever has been read so far. If a site is larger than that, run it against the individual sitemaps in turn: most large sites split theirs by section or by post type anyway.

More free SEO tools

XML Sitemap Keyword Extractor Pull every URL out of a sitemap and turn each slug into a clean keyword you can filter and export. XML Sitemap Generator Crawl a website and build XML, index, image, video, news, TXT or HTML sitemaps ready to submit. Blog Title Generator Generate titles from formats that are documented to perform, then see every one scored and ranked by the same engine that powers our title analyzer. Blog Title Analyzer Score a headline out of 100, see exactly where every point came from, and compare up to ten titles side by side. Word Counter Count words, characters, sentences and reading time as you type, convert between seven letter cases, and clean up messy text. Paraphrasing Tool Rewrite padded prose into plain English by rule, with every change listed, explained and reversible. No AI, no word limit. HowTo Schema Generator Build valid HowTo JSON-LD with the ISO 8601 durations handled for you, read the steps off a live page, and be told when Recipe would serve you better. FAQ Schema Generator Build valid FAQPage JSON-LD, or read the questions straight off a live page, with an honest account of what Google still does with it. Duplicate Word Checker Catch doubled words, echoes, repeated phrases and pasted sentences, with thresholds measured from real edited prose. Free Readability Checker Score any text or page on seven published formulas at once, with syllables counted against a real pronouncing dictionary. SERP Snippet Preview Tool See how your title and description will really look in Google, measured in pixels, on desktop, tablet and mobile. Keyword Frequency Counter Count how often every word and phrase appears, compare up to four texts or pages side by side, and track your own keyword list. Sentence Length Checker Find every long sentence, see the rhythm of your writing laid out, and fix the ones that are hard to read. Page RPM Calculator Work out page, session and impression RPM from one set of figures, see what a lift is worth, and check it against published benchmarks. Writing Time Calculator Work out how long a piece will take to write, phase by phase, from composition rates rather than typing speed, and turn the hours into a quote. Reading Time Calculator Work out how long a page takes to read, using speeds from published research rather than round numbers, adjusted for how hard your text is. Paragraph Length Checker Find the walls of text, see how many lines each paragraph takes on a phone, and get every paragraph listed longest first. Heading Structure Checker Check your H1 to H6 outline for missing, duplicated, skipped or empty headings, from a live page, pasted HTML or plain text. Text Case Converter Convert any text to sentence case, lower case, UPPER CASE, camelCase, kebab-case and six more. Pick a case, copy the result. Title Capitalization Tool Capitalise a headline in AP, APA, Chicago or MLA style, and check it fits a Google result before you publish it. Keyword Density Checker Analyse any page or block of text for keyword frequency, density, prominence and where each phrase actually appears.