How to Write and Test Title Tags That Earn the Click

Ranking on the first page is only half the job. The other half is persuading a searcher to choose your result over the nine others around it, and the main tool you have for that is the title. A clear, specific title can lift clicks without any change in position, while a vague or truncated one can waste a ranking you worked months to earn.

This guide covers how search engines treat titles today, how to write them, and how to run simple before and after tests so that changes are based on evidence rather than taste. It is aimed at site owners and small teams who manage their own pages. If you want to see how your current titles appear in live results before editing anything, a free toolkit like Rank Crown lets you check rankings and preview search results in one place.

What the title link actually is

Google calls the clickable headline of a search result the title link. Its documentation on influencing title links explains that the main source is the HTML title element, but that Google may also use other text on the page, such as the main visible heading, when it judges that text to describe the page better. In other words, the title you write is a strong suggestion, not a guarantee.

Knowing this changes how you write. A title that matches the page's main heading and accurately summarizes its content is much more likely to be shown as written. A title stuffed with keywords, padded with boilerplate, or disconnected from the visible content is more likely to be rewritten, and the rewrite may not be the version you would have chosen.

The anatomy of a title that works

Most effective titles contain three elements, in roughly this order.

  1. The topic, stated plainly. Use the words a searcher would use. If people search "invoice template for freelancers", say that, rather than an internal product name.
  2. A reason to choose this result. This might be a specific angle, a format, a time frame, or an audience: "for first-time employers", "with examples", "step by step".
  3. The brand, if it helps. For well-known brands, the name adds trust. For smaller sites, it often adds little and can be shortened or moved to the end.

Length matters because long titles get truncated in results. There is no fixed character limit, since display depends on pixel width and device, so the practical rule is to put the most important words first and keep the whole title concise. If the end gets cut, the reader should still understand what the page is about.

Common title problems worth fixing first

Before writing new titles, audit the ones you have. Export the title of every indexable page from a crawler or your content management system, and look for these issues.

Duplicate titles

Several pages sharing the same title make it hard for searchers and search engines to tell them apart. This often happens with category pages, paginated archives, or product variants. Give each page a title that reflects what makes it different.

Boilerplate-heavy titles

Titles that start with the brand name and a long tagline push the useful words to the end, where they get cut off. Move the topic to the front.

Titles that do not match the page

If the title promises a "complete guide" and the page is three paragraphs long, searchers bounce and Google may rewrite the title anyway. Either expand the page or make the title honest.

Missing or empty titles

These are rarer on modern platforms but still appear on old landing pages and template-generated pages. They are the easiest wins of all.

Meta descriptions: the supporting line

The text below the title link is the snippet. Google's documentation on controlling snippets notes that snippets are mainly generated from page content, and that a meta description may be used when it describes the page more accurately than other parts of the content. As with titles, your meta description is a suggestion.

A good description expands on the title rather than repeating it. It tells the reader what they will find, who it is for, and what they can do with it. Write it as a short, natural sentence or two, and make sure every page has its own. Duplicate descriptions across many pages are a common reason why snippets get generated from random page text instead.

Choosing which pages to work on

Rewriting every title on a large site at once is a mistake, because you will not be able to tell which changes helped. Start with pages where a better title is most likely to make a measurable difference.

Pick ten to twenty pages from these lists for your first round. That is enough to learn something without making the results impossible to interpret.

Running a simple before and after test

True split testing of titles requires large sites and specialised tooling, because you cannot show two titles for the same page to different searchers. For most sites, a careful before and after comparison is the practical alternative. The method below keeps it as fair as possible.

  1. Record the baseline. For each chosen page, note clicks, impressions, click-through rate, and average position over the previous four weeks from Search Console.
  2. Change only the title and description. Do not update the content, internal links, or anything else at the same time, or you will not know what caused the result.
  3. Log the change date. Keep a simple sheet with the page URL, old title, new title, and the date you published the change.
  4. Wait for the change to be picked up. Check the live result for each page until the new title appears. Then wait another four weeks.
  5. Compare like with like. Look at click-through rate for the four weeks after the change against the four weeks before, and check that average position stayed roughly stable. If position moved a lot, the comparison is less reliable, because click-through rate naturally changes with position.

Seasonality can distort these comparisons. If your business has strong seasonal patterns, compare with the same period in the previous year as well, or keep a group of similar pages unchanged as a rough control.

Interpreting the results honestly

Not every rewrite will win. Some will improve click-through rate clearly, some will make no difference, and a few will perform worse. Treat each outcome as information. When a change wins, look for the pattern: did adding the audience help, did putting the number first help, did removing the brand help? When a change loses, roll it back and note what you tried.

Over several rounds, you build a small library of what works for your audience. That library is worth more than generic advice, because searchers in different industries respond to different cues. A legal services audience may respond to reassurance and clarity, while a software audience may respond to specifics like "free" or "no signup".

Three illustrative rewrites

The examples below are invented for illustration, not results from a real test, but they show the kind of reasoning that tends to produce better titles. Use them as a pattern for your own drafts rather than as templates to copy.

A bakery's page titled "Home | Sunrise Bakery | Fresh Baked Goods Since 1998" puts the brand and a slogan first and says nothing a hungry searcher is looking for. A rewrite such as "Sourdough and Custom Cakes in Leeds, Order by Noon for Next Day" leads with the products, names the place, and gives a concrete reason to choose this result over a generic listing. The founding year can move to the page itself, where it still builds trust.

A software company's help article titled "Exporting Data" is accurate but anonymous among dozens of similar pages. "How to Export Invoices to CSV in Two Minutes" names the specific object, the format, and the effort involved. Someone who searched for exactly that task will recognise the answer immediately, which is the whole point of a title link.

A blog post titled "The Ultimate Complete Guide to Everything About Email Marketing in 2026" tries to claim every keyword at once and ends up saying very little. Something narrower, like "Email Welcome Sequences for Small Shops: What to Send in the First Week", promises a defined outcome for a defined reader. It may match fewer searches in total, but it matches the right ones far better, and those searchers are the ones most likely to stay on the page.

In each case the improvement comes from the same habit: replacing general claims with specifics that a searcher can check against their own need in under a second.

Writing titles at scale without losing quality

Templates are useful for large sets of similar pages, such as product listings or location pages, but they need care. A template should pull in the unique attribute of each page, such as the product name, the city, or the key specification, and place it at the front. Avoid templates that produce near-identical titles differing only in a single number, because those look like duplicates in results.

For editorial content, write titles by hand. Draft three or four options for each important page, read them aloud, and ask which one you would click if you saw it among competitors. Checking the live results page for your target keyword helps here, because it shows what you are competing against and where there is room to stand out.

A short checklist for every new page

Summary

Titles are your strongest lever for turning rankings into visits. Write them plainly, put the topic first, give a specific reason to click, and keep them honest about what the page contains. Fix duplicates and boilerplate before anything else, then improve high-impression pages in small, logged batches and compare click-through rates before and after. Over time, the evidence from your own site will tell you more about good titles than any rule of thumb.