Key takeaways
- Title tags at scale are templates: a fixed pattern filled with the page's own data.
- Write one template per page type and add hand-written titles for the pages that matter most.
- Google rewrites titles that are half-empty, inaccurate, outdated or nearly identical to siblings.
- Test on a group of similar pages and measure clicks and CTR in Search Console before rolling out.
Writing title tags at scale is one of the fastest wins on a large client site, and one of the easiest to get wrong. We supervise SEO on 18 US online stores with more than 3 million products, and on a dealership team we published 81,210 pages across a fleet of stores. None of those titles were written one by one. This guide explains how to build the rules, how to check them and how to sell the work to clients. It is part of the on-page work we deliver through white label SEO.
What does writing title tags at scale mean?
It means defining a pattern for each page type, such as product, category or location, and filling it with variables from the page's data: product name, brand, model, city. The pattern sets order, separators and branding. Good templates produce titles that are unique, accurate and readable on every page, including the ones nobody will check by hand.
How do you build title tag templates with variables?
Start with the main search for each page type, then place the most distinctive variable first. A product title might be product name, key attribute, brand of the store. Add fallbacks for empty fields, trim long values and avoid fixed words that repeat on every page. Then print a sample of real outputs and read them.
| Page type | Template example | Watch for |
|---|---|---|
| Product | {{product}} {{key attribute}} | {{store}} | Empty attributes, very long product names |
| Category | {{category}}: {{top subtypes}} | {{store}} | Overlap with subcategory titles |
| Location | {{service}} in {{city}} | {{business}} | Pages with no real local content |
| Vehicle (dealership) | {{year}} {{make}} {{model}} {{trim}} for sale in {{city}} | Same title on near-identical units |
| Article | Hand-written | Templates rarely work for editorial pages |
Category titles deserve extra care because those pages carry most of the broad searches; our guide on ecommerce category page SEO covers the rest of the page. Templates are also the core of programmatic SEO, where the dataset behind the template matters more than the wording.
Why does Google rewrite title tags?
Google generates title links automatically and uses the title element as one source among several, including headings, prominent text and anchor text. It is more likely to replace a title that is half-empty, outdated, inaccurate, nearly identical across many pages or in a different language from the page. Fixing those patterns is the main lever.
Google's documentation on influencing your title links lists those cases. Two matter most at scale. "Micro-boilerplate" is when many pages share almost the same title, so Google adds a detail to tell them apart. "Obsolete" is when a recurring page keeps an old year in its title. Both come straight from templates that were never reviewed.
How should meta descriptions work on large sites?
Google builds snippets mostly from page content and uses the meta description when it describes the page better. On large database-driven sites, Google says programmatically generated descriptions are appropriate and encouraged. The key is uniqueness: identical descriptions across many pages do not help. Use the page's own data, not a sitewide sentence.
The details are in Google's guide to controlling your snippets, which also says there is no fixed length limit; snippets are truncated as needed. A useful product description template mentions what is specific to the item (size, material, price range, availability) and one reason to buy from this store.
How do you test title tag changes on thousands of pages?
Pick a group of similar pages, change the template on part of it and keep a comparable part unchanged. Record clicks, impressions, CTR and average position in Search Console for a fixed period before and after. Compare the two groups, not just before and after, so seasonality and updates do not fool you.
- Choose page groups with enough impressions to show a difference.
- Change one thing at a time: order, a variable or the branding.
- Request recrawling of a sample with the URL Inspection tool.
- Allow time: Google says title link changes can take a few days to a few weeks to show.
- Filter the Performance report by page and compare groups over the same dates.
- Roll out the winner, then check a sample of live titles in search results.
Which pages should get hand-written titles?
The pages that carry the most impressions or revenue: the home page, top categories, best-selling products, core service pages and anything that ranks just below the first results. Templates cover the long tail well, but a few hundred hand-written titles on the pages that matter most often move more clicks than another template change.
A practical split for large clients: export pages from the Performance report sorted by impressions, write titles by hand for the top slice and let templates handle everything else. Keep a field in the CMS that overrides the template, so a hand-written title is never overwritten by the next import. Document which pages have overrides, because a future redesign or migration is the moment those titles tend to disappear.
On dealership fleets and multi-store groups, the same template runs on many domains at once. That multiplies both the benefit and the risk: one wrong variable produces the same broken title on every site. Test on one store first, then roll out store by store.
How do you QA titles before they go live?
Export every generated title with its URL and run simple checks: empty or duplicated titles, missing variables, titles that start with the brand, unusual length and leftover placeholders. Then read a random sample of a hundred aloud. Most template bugs are visible in minutes once the output is in a spreadsheet instead of a CMS screen.
Clients rarely notice title work until something breaks, so show them the before-and-after sample and the Search Console comparison in the monthly report. If you need this delivered under your brand, our white label on-page SEO covers templates, QA and testing. Or book a 30-minute call.
Frequently asked questions
How long should a title tag be?
Google does not set a character limit; long titles get truncated, so put the distinctive words first.
Should the brand name go in every title?
Google suggests including it concisely, at the start or end, separated by a delimiter.
Can AI write title tags for thousands of pages?
It can draft them, but outputs still need data checks and human review before launch.
Do title tags still matter if Google rewrites them?
Yes. The title element is still one of Google's main sources for title links.
Should meta descriptions be written for every product?
On large catalogs a data-driven template is fine, as long as each output is unique.