Get in touch

Title tag and meta description templates at scale

Updated August 19, 2026 · 8 min read · Search intent : informational · Architecture

On a generated site the title tag is the most sensitive variable in the template. It drives click-through, it is the first thing compared between two similar pages, and it is the element most often rewritten automatically when built badly.

Title tags produced from a template
Title tags produced from a template

Useful lengths

ElementTargetWhat happens beyond
Title tag50–60 charactersTruncated on display; the tail is lost
Meta description140–160 charactersTruncated, or rewritten from body copy
H1No display constraintPast 70 characters readability drops

These are guides, not limits: actual truncation depends on pixel width, not character count. The practical rule is to put the distinguishing information in the first 45 characters.

Four variable title formulas

1. {Entity}: {main benefit} ({modifier})
   → XML sitemaps: splitting 100,000 URLs (2026 guide)

2. {The question as typed}
   → How many internal links per page?

3. {Number} {objects} for {task}
   → 7 title patterns for a generated site

4. {Entity A} vs {Entity B}: {decision criterion}
   → Programmatic SEO vs editorial: how to choose

Formula 2 deserves special mention: reusing the question exactly as typed works well on long tail queries, where users recognise their own phrasing. It is also the least likely to be rewritten.

Never use a single formula for an entire batch. Three or four formulas assigned by intent type prevent a uniform pattern across hundreds of results.

Why Google rewrites a title

Google replaces the displayed title when it judges the tag a poor description of the page. On generated sites four causes recur:

  1. Keyword stuffing — the query repeated twice in the same tag.
  2. Over-generic titles — "Home", "Article", or a bare entity name.
  3. Content mismatch — the title promises a comparison, the page is a definition.
  4. An overlong brand suffix — a fifty-character tagline eating half the usable space.

The last one is fixed immediately: a short brand suffix, or none at all on long tail pages where every character counts.

Handling title collisions

if title in already_generated_titles:
    add_secondary_variable()      # city, year, segment
    if still_colliding:
        do_not_publish()          # the two pages are duplicates

A title collision is almost always a symptom of a content collision. Appending a number to make the title unique hides the problem instead of fixing it — a case covered in canonicals on templated pages.

What the meta description is for

It plays no part in ranking. It influences click-through, which matters a great deal on a batch where each page collects a few dozen impressions a month.

A good generated description contains three things: what the page answers, the differentiator (a figure, a proprietary data point, a method), and a reason to click rather than stay on the results page. Impersonal phrasing scales better than imperatives, which become repetitive across hundreds of snippets.

Five automated checks

These run at generation time and block publication on failure — the same principle as validating heading hierarchy.

Frequently asked questions

What is the ideal title tag length?

Roughly 50 to 60 characters, though real truncation depends on pixel width. What matters most is placing the distinguishing information in the first 45 characters.

Does the meta description affect rankings?

No, it is not a ranking factor. It affects click-through from the results page, which stays significant on low-impression long tail pages.

Why does Google show a different title than mine?

Because it judges the tag unrepresentative: too generic, keyword-stuffed, truncated by a long brand suffix, or mismatched with the actual content.