Writing for a Space You Can't Measure
You've written a meta description. It's 156 characters — safely inside the limit everyone quotes.
It gets truncated anyway. Meanwhile a competitor's 168-character description displays in full.
The reason is that search result snippets aren't limited by character count. They're limited by pixel width, and characters have wildly different widths. A description full of i, l, t and f fits far more characters into the same space than one full of m, w, W and M.
The character counts everyone repeats — 60 for titles, 155 for descriptions — are rough proxies derived from average character widths. Useful as a guideline, unreliable as a rule.
Why Pixel Width Wins
Search results render in a proportional typeface. Compare:
Illinois litigation ← narrow characters
Warm summer meadow ← wide characters
Same character count, noticeably different rendered width.
The practical consequences:
Capitals are wide. A title in title case takes more space than the same words in sentence case.
Numbers are usually uniform width in the typefaces used, so they're predictable.
Punctuation is narrow. Pipes, colons and hyphens cost little.
Some languages are dramatically different. German compounds are long. CJK characters are typically full-width, so a Japanese title fits far fewer characters into the same pixel space.
The exact pixel limits change as search engines adjust their layouts, and differ between desktop and mobile. Treating any specific number as fixed is a mistake.
The practical rule: front-load. Put the essential information in the first 50 characters of a title and the first 100 of a description. Then truncation costs you the least important part.
What Truncation Actually Costs
A truncated title or description isn't a ranking problem. Search engines index the full page regardless.
It's a click-through problem. A snippet ending mid-sentence with an ellipsis looks less considered than one that ends cleanly, and it may cut the very thing that would have persuaded someone to click.
There's a second effect worth knowing: search engines frequently rewrite titles and descriptions anyway. Studies have consistently found that a substantial proportion of displayed titles differ from the page's title tag, and meta descriptions are rewritten even more often — search engines regularly pull a snippet from body content instead when it matches the query better.
This doesn't make writing them pointless. It means:
- Your tag is a strong suggestion, not a guarantee
- A well-written tag that matches likely query intent is more likely to be used
- A tag that's truncated, keyword-stuffed, or duplicated across pages is more likely to be replaced
The Meta Tags That Do Something
Generating a full set is easy; knowing which ones matter is the useful part. The Meta Tag Generator produces the standard set:
- Enter your title, description and URL.
- Generate the tag block.
- Add it to the page
<head>.
Worth knowing what each one does:
<title>
The most important single tag. Used as the clickable headline in search results, as the browser tab label, and as the default when someone bookmarks the page.
Not technically a meta tag, but it's what everyone means when they say "meta title."
<meta name="description">
Not a ranking factor. Influences click-through when it's used, which is a substantial fraction of the time. Worth writing well; not worth agonising over.
<meta name="robots">
Actually consequential. Controls indexing behaviour:
<meta name="robots" content="noindex, nofollow">
<meta name="robots" content="index, follow"> <!-- the default; unnecessary -->
<meta name="robots" content="max-snippet:150">
<meta name="robots" content="noarchive">
The one important caveat: a page blocked in robots.txt cannot have its meta robots tag read. If you want a page removed from the index, allow crawling and use noindex. Blocking it in robots.txt prevents the crawler from ever seeing the instruction, and the URL can remain indexed based on external links.
<meta charset>
<meta charset="utf-8">
Must appear within the first 1024 bytes of the document. Without it, the browser guesses the encoding, and guesses wrong often enough to produce mojibake.
<meta name="viewport">
<meta name="viewport" content="width=device-width, initial-scale=1">
A rendering instruction, not an SEO tag — but without it mobile browsers assume a desktop-width viewport and render your responsive design at a zoomed-out desktop size.
Open Graph and Twitter Card tags
Control link previews on social platforms and messaging apps. Separate system from search snippets, and worth setting independently — a good search description and a good social description often differ.
The Tags That Don't Do Anything
Plenty of legacy tags persist in templates and CMS defaults:
<meta name="keywords" content="..."> <!-- ignored by major search engines -->
<meta name="author" content="..."> <!-- no search effect -->
<meta name="revisit-after" content="7 days"><!-- never honoured by anyone -->
<meta name="distribution" content="global"> <!-- meaningless -->
<meta name="rating" content="general"> <!-- essentially unused -->
<meta name="generator" content="..."> <!-- fingerprints your stack -->
None of them help. The generator tag actively tells attackers which CMS and version you're running, which is worth removing.
Uniqueness Matters More Than Optimisation
The single most common meta tag problem on real sites isn't poor wording. It's duplication.
A CMS that applies a template title to hundreds of pages produces:
Products | Example Company
Products | Example Company
Products | Example Company
Search engines have no way to distinguish them, and users scanning results can't either. This is a much bigger problem than a description that's ten characters too long.
Audit for uniqueness first. Crawl your site, export titles and descriptions, and check for duplicates. Fixing the templates that generate them addresses hundreds of pages at once.
Common sources of duplication:
- Paginated pages with identical titles
- Filtered and sorted category URLs
- Templates that don't inject page-specific variables
- Pages where the description field was left empty and a site-wide default filled in
Writing Ones That Get Used
Match the query intent. A title that reads like an answer to a question people ask is more likely to be shown than a brand-first construction.
Put the distinguishing information first. Brand name at the end, if at all.
Avoid keyword repetition. A title repeating a term three times is a rewrite candidate.
Make the description a summary, not a sales pitch. Descriptions that describe the page content accurately are more likely to be used than ones that read as marketing copy.
Don't duplicate the title in the description. They occupy adjacent lines; repetition wastes the space.
Include a reason to click — a specific number, a distinguishing detail, a clear statement of what the page delivers.
Skip the boilerplate. "Welcome to our website" and "Learn more about our services" say nothing.
Practical Tips
Front-load everything. First 50 characters of the title, first 100 of the description.
Preview at both desktop and mobile widths. Truncation points differ.
Use sentence case for long titles. It's narrower than title case and fits more.
Check robots.txt isn't blocking pages you've noindexed. The contradiction is common and prevents deindexing.
Remove the generator tag and any other legacy meta tags your CMS adds by default.
Audit for duplicates before optimising wording. Uniqueness is the bigger win.
Don't chase exact character counts. Write something good, front-load it, and accept that truncation depends on rendering you can't fully control.
FAQ
How long should a meta description be? Around 150–160 characters is a reasonable target, but the real limit is pixel width. Front-load the important part rather than optimising to a count.
Do meta descriptions affect rankings? Not directly. They affect click-through rate when they're displayed, which is worth something on its own.
Why does Google rewrite my title? Usually because the tag is truncated, keyword-stuffed, duplicated across pages, or doesn't match the query well. A clear, unique, query-relevant title is more likely to be used.
Is the keywords meta tag still useful? No. Major search engines have ignored it for many years.
Should I noindex or block in robots.txt?
Use noindex and allow crawling. Blocking prevents the crawler from reading the noindex instruction.
Are Open Graph tags the same as meta descriptions? No. Open Graph controls social link previews; the meta description relates to search snippets. Set both, and they don't have to be identical.
The Takeaway
Meta tags are worth getting right and not worth agonising over. The two things that actually matter are uniqueness across pages and front-loading the important information — both are more consequential than hitting a character count that was only ever an approximation of a pixel limit.
Generate SEO meta tags free with the Meta Tag Generator at sadiqbd.com — no sign-up, instant results.
