How to Write a Search Result That Earns More Clicks

Here's the plain answer. To get more clicks without ranking higher:

1. Find the queries worth rewriting for. In Search Console, look for queries with real impressions and a click-through rate well below what your own pages usually get at that position.
2. Look at the result as it's displayed, not as you wrote it. Run the query and write down what Google actually shows for each piece. Often the title you're about to "improve" isn't the title anyone sees.
3. Fix the piece that's failing. Usually that's a boilerplate title, a snippet pulled from the wrong paragraph, a stale date or a site name Google didn't accept. Each has its own fix, covered below.
4. Measure against pages you didn't touch. A CTR change only means something next to a control group, at a steady position, starting from the day the new text went live.

Four limits apply from the start. No rewrite guarantees a single extra click. Google can rewrite titles and snippets anyway. Position sets a ceiling that no sentence can lift. And a higher CTR isn't a ranking lever: Google has said publicly that clicks aren't a direct ranking factor, a record we go through in click data as a ranking signal. The payoff is more visits from impressions you're already getting, and that's reason enough.

Google's Visual Elements gallery names every part of a text result. The top line is attribution, which "can include various aspects of the source, such as the name of the site, favicon, and URL to the web page." Below that come the title link, the snippet and, sometimes, a byline date and rich attributes. What shows up varies by device, query and language.

What Google shows. Google's site names documentation says the site name is "for the entire site", unlike title links, which belong to single pages. You set a preference with WebSite structured data on the home page, and the documentation says "WebSite structured data is most important, if you want to specify a preference." It also reads og:site_name, the home page </code>, headings and other home page text. There's one site name per domain or subdomain: "Google Search does not support site names at the subdirectory level."</p> <p><strong>When Google replaces yours.</strong> The documentation says that "if our system is less confident in a name you provided, it may sometimes generate site names using other sources or show a domain or subdomain name." It warns against generic names: "A generic name like "Best Dentists In Iowa" is unlikely to be selected by our system as a site name, unless that's an extremely well-recognized brand name." The two fixes it gives are an <code>alternateName</code> (an acronym, a shorter form, or your lowercase domain as a fallback) and using the same name everywhere on your home page.</p> <p><strong>The favicon.</strong> It comes from a link tag on the home page. The rules are short and strict. "Your favicon must be a square (1:1 aspect ratio) that's at least 8x8px", and Google recommends one "larger than 48x48px". It has to be crawlable, and "The favicon URL must be stable (don't change the URL frequently)." The supported formats listed on that page are BMP, GIF, ICO, PNG, JPEG, PPM and TIFF, so a site that only ships an SVG icon should add one of those. Even then, "A favicon isn't guaranteed to appear in Google Search results, even if all guidelines are met."</p> <p><strong>What makes a searcher choose it.</strong> Recognition, mostly. Returning searchers spot the name and icon before reading a word, and a clean brand name reads as a business where a keyword phrase reads as a directory.</p> <p><strong>Before and after.</strong> Invented examples, not real brands:</p> <div class="table-wrap"><table><thead><tr><th>Piece</th><th>Before</th><th>After</th><th>Why</th></tr></thead><tbody><tr><td>Site name preference</td><td>"Cheap Tiles Online Best Prices"</td><td>"Example Tile Shop", with alternateName "ETS"</td><td>A generic phrase is unlikely to be picked; a brand name with a fallback is</td></tr><tr><td>Home page references</td><td>Header says "ETS Tiles", title says "Example Tiles Ltd"</td><td>"Example Tile Shop" everywhere on the home page</td><td>Google checks that the name is consistent across home page text</td></tr><tr><td>Favicon</td><td>A CMS default icon, served from a new hashed URL on each deploy</td><td>A square PNG of the logo mark, over 48 pixels, at one fixed URL</td><td>Meets the size rule and the stable URL rule</td></tr></tbody></table></div> <p>Nothing here touches individual pages, and changes take effect only after Google recrawls the home page, which the documentation says can take "several days to several weeks".</p>

What Google shows. Per the gallery, "A visible URL has two parts: domain and breadcrumb." The breadcrumb part is either taken from your URL path or built from BreadcrumbList markup. One detail changes how much this is worth to you: Google's breadcrumb documentation currently says "This feature is available on desktop in all regions and languages where Google Search is available." If most of a query's impressions come from phones, a breadcrumb rewrite reaches a minority of the people who see that result. Check the device split in Search Console before you spend time here.

What makes a searcher choose it. The path labels the page type. A trail ending in Pricing, Guides or Support tells a buyer or a researcher what they'll land on before they read the title. Query strings and ID numbers tell them nothing.

Before and after. Invented paths on our example domain:

Changing a live URL means redirects and a recrawl, so breadcrumb names are usually the cheaper fix. They must match the trail visible on the page, and the markup details are in how to add breadcrumb and organization schema.

This is the piece that decides most clicks, and Google's documentation says so: the title link is "often the primary piece of information people use to decide which result to click".

What Google shows. Your </code> element is the first of nine documented sources. The rest are the main visual title on the page, heading elements such as <code><h1></code>, <code>og:title</code>, other large and prominent text, other text on the page, anchor text on the page, the text of links pointing to the page, and WebSite structured data. Length isn't capped, but "the title link is truncated in Google Search results as needed, typically to fit the device width", so the end of a long title is the part that gets cut.</p> <p><strong>When Google rewrites it.</strong> The documentation is direct: "If we've detected an issue on the page, we may try to generate an improved title link from anchors, on-page text, or other sources." The issues it lists are the checklist to write against:</p> <div class="table-wrap"><table><thead><tr><th>Issue Google names</th><th>What it looks like</th><th>What to write instead</th></tr></thead><tbody><tr><td>Half-empty title</td><td>A title that is only a separator and the site name</td><td>The page's actual subject, then the brand if there's room</td></tr><tr><td>Obsolete title</td><td>Last year's date in the title, this year's on the page</td><td>Update both together, or leave the year out</td></tr><tr><td>Inaccurate title</td><td>A title listing things the page doesn't cover</td><td>Describe what the page actually contains</td></tr><tr><td>Micro-boilerplate</td><td>A set of pages sharing one title, differing only in a detail left out</td><td>Put the distinguishing detail, such as season or model, in each title</td></tr><tr><td>No clear main title</td><td>Several headings styled the same</td><td>One prominent `<h1>` that matches the title's subject</td></tr><tr><td>Language mismatch</td><td>An English title on a Hindi page</td><td>Title in the page's language and script</td></tr><tr><td>Site name repeated</td><td>Brand at the start of the title and in the site name line</td><td>Expect the brand to be dropped from the title link</td></tr></tbody></table></div> <p>The documentation says "Google may omit the site name from the title link, if it's repetitive with the site name that's already shown in the search result." A "| Example Tile Shop" suffix costs characters in your <code><title></code> and may never be shown, because the brand already appears in the site name line above. If you're short of space, the suffix is the first thing to cut.</p> <p><strong>What makes a searcher choose it.</strong> Google's starter guide puts the standard in one line: "a good title is unique to the page, clear and concise, and accurately describes the contents of the page." In practice a searcher is looking for four things, roughly in this order:</p> <ul> <li><strong>Their object, in their words.</strong> The noun they typed, not your internal name for it.</li> <li><strong>Their situation.</strong> A qualifier that tells them this version is for them: for renters, for small bathrooms, on a phone.</li> <li><strong>The kind of answer.</strong> A comparison, a price list, a calculator, a step list.</li> <li><strong>Something the other results lack.</strong> Your title is read next to nine others. The documentation criticizes boilerplate titles because they make it "impossible for users to distinguish between two pages", and the same test applies across a results page. Run the query before you write and look at the titles already there. If five of them say "Complete Guide", saying it a sixth time adds nothing.</li> </ul> <p>Put the words that do this work at the front, since truncation removes the end.</p> <p><strong>Before and after.</strong> Invented pages:</p> <div class="table-wrap"><table><thead><tr><th>Page</th><th>Before</th><th>After</th><th>What changed</th></tr></thead><tbody><tr><td>Comparison guide</td><td>"Tiles \</td><td>Example Tile Shop"</td><td>"Porcelain vs Ceramic Floor Tiles: Cost, Wear and Water"</td><td>Half-empty template replaced with the object and the three things buyers compare</td></tr><tr><td>Pricing page</td><td>"Pricing Plans - Best Value Accounting Software Solutions"</td><td>"Example Ledger Pricing: Three Plans, Monthly or Yearly"</td><td>Stuffing removed, the shape of the answer stated up front</td></tr><tr><td>Help article</td><td>"How To Fix It"</td><td>"Fix a Bank Feed That Stopped Syncing in Example Ledger"</td><td>Vague label replaced by the searcher's actual problem</td></tr><tr><td>Seasonal page</td><td>"2024 Tile Trends" on a page updated for 2026</td><td>"Floor Tile Trends for 2026" on the same, updated page</td><td>Removes the obsolete title issue Google says it corrects</td></tr></tbody></table></div> <p><strong>Don't write clickbait.</strong> A title that promises more than the page delivers is an "inaccurate" title, and Google says it may replace it. Google's helpful content questions ask: "Does the main heading or page title avoid exaggerating or being shocking in nature?" Its Discover guidance is blunter, telling publishers to "Avoid clickbait and similar tactics to artificially inflate engagement by using misleading or exaggerated details in preview content (title, snippets, or images) to increase appeal, or by withholding crucial information required to understand what the content is about." Making the title agree with the page's headline is its own discipline, covered in <a href="https://www.sparkcliks.com/blog/match-title-tag-to-landing-page-headline">how to match a title tag to your landing page headline</a>.</p>

What Google shows. From Google's snippet documentation: "Snippets are primarily created from the page content itself. However, Google sometimes uses the meta description HTML element if it might give users a more accurate description of the page than content taken directly from the page." And the line that changes how you write: "Google Search might show different snippets for different searches." Like the title, the snippet "is truncated in Google Search results as needed, typically to fit the device width."

When Google ignores your description. The documentation gives the reasons. The same description on every page doesn't help, and it says descriptions "comprised of long strings of keywords don't give users a clear idea of the page's content, and are less likely to be displayed as a snippet." A description that doesn't match the query is another, since Google picks the text that best fits each search.

What makes a searcher choose it. Google calls a good meta description "like a pitch that convince the user that the page is exactly what they're looking for." It has to confirm the title's promise with detail the title had no room for. The documentation suggests plain facts: for a product, the price, age and manufacturer; for an article, the author and publication date.

The practical consequence. Because snippets are chosen per query, a meta description can only serve one group of queries, normally the main one. For every long-tail query, Google is pulling a snippet from your body text. That makes the first sentence under each heading snippet copy. A section that opens "In this section we'll look at" gives Google nothing to show. A section that opens with the answer gives it a snippet.

Before and after. Invented pages:

nosnippet, max-snippet and data-nosnippet can remove or shorten what's shown, but they're a veto, not a way to choose. Featured snippets are a separate, extracted surface, covered in how featured snippets are chosen.

What Google shows. A byline date is "the date that Google estimates that the web page was updated or published", and it appears only when Google thinks it's useful. Google says it doesn't rely on one signal: "Google doesn't depend on a single date factor because all factors can be prone to issues."

What you can do. The byline dates documentation lists it. "Add a user-visible date to the page and feature it prominently", with a label such as Published or Last updated. Add datePublished and dateModified to your Article or BlogPosting markup. Make the visible and structured dates match. "Don't specify future dates, or the date of the action described on the page." And if Google keeps picking the wrong date: "Minimize the presence of other dates on the page". Even then, the documentation notes that "Google doesn't guarantee that a byline date, whether visible or in structured data, will be shown in search results".

What makes a searcher choose it. On queries where things change, like prices, software steps or rules, a recent date is a reason to click and an old one a reason to skip.

Before and after. Invented page:

Change the date only when the page has actually changed. A bumped date on an unchanged page is the same mismatch as an exaggerated title.

What Google shows. Rich attributes are "one or more lines of additional information about the web page, such as review stars and recipe information", usually powered by structured data. Google's structured data gallery lists review snippets for books, recipes, movies, products, software apps and local businesses, plus product details such as price and availability, events, videos and more.

What's actually available to you. Less than most plans assume.

When Google leaves them out. The structured data policies say "Google does not guarantee that your structured data will show up in search results, even if your page is marked up correctly according to the Rich Results Test." They also say "Don't mark up irrelevant or misleading content, such as fake reviews or content unrelated to the focus of a page." Misleading markup can also bring a structured data manual action, which removes rich result eligibility.

Before and after. Invented product page: before, a plain result with no price. After, Product markup with the real price and stock status shown on the page, and review markup only for genuine reviews the page displays.

Every figure below is an invented example, from an imaginary site called Example Tile Shop. None of it is SparkCliks data, a study result or a benchmark. The method is real and runs on your own property.

Step 1. Export the candidates. In Google Search Console, open Performance, then Search results. Set Search type to Web and the date range to Last 3 months. Add a Query filter, Custom (regex), set to Doesn't match, with your brand name and its misspellings. Open the Queries tab, turn on Clicks, Impressions, CTR and Position, and export.

Step 2. Mark low CTR for the position. There's no position filter in the interface, so this part happens in the spreadsheet. Band each row by position, take the median CTR of your own rows in each band, and divide each query's CTR by its band median. Drop anything under an impression floor you set in advance (500 impressions over three months for a site this size). The full method, including why you use the median and keep brand queries out, is in how to set a realistic organic CTR benchmark.

Step 3. Audit the displayed result. For each low row, filter Search Console to the ranking page to confirm which URL it is. Then run the query in a private window, set to the country and device that make up most of its impressions. Record every piece in a sheet with one column for what you wrote, one for what Google displayed, and one for which documented source it appears to come from. That third column tells you where the fix goes.

The last row matters. A query can have low CTR because the result did its job: the searcher got the answer on the results page. Rewriting that snippet to hold back the answer would be the withholding tactic Google's Discover guidance warns about. Leave it.

Step 4. Rewrite one class of change. For the comparison page: new title "Porcelain vs Ceramic Floor Tiles: Cost, Wear and Water", a page-specific description, a first paragraph that opens with the answer, and data-nosnippet on the cookie notice so it stops competing for the snippet. For the ideas page: a visible "Last updated" date, a matching dateModified, and comment dates moved out of the main content. Keep "how to cut porcelain tile" untouched. At an index of 0.92 it's performing normally, and pages like it make up the control group.

Step 5. Measure without fooling yourself. Start the change window on the day the new text is displayed, which the title link documentation says can take "a few days to a few weeks", not on the publish day. Compare against untouched pages in the same position band, at steady position. At about 2,000 impressions per 28 days, one query can only reveal a large change, so pool similar pages. The design, impression floors and control arithmetic are in how to run a title tag test and read CTR honestly.

• It can't guarantee clicks. Every Google page quoted above says the same thing: no guarantee that a title, snippet, site name, favicon, date or rich result will be shown the way you asked.
• It can't stop Google rewriting. Title links and snippets are generated. You can remove the issues Google names, and it can still pick other text for some queries.
• It can't beat position. A result sitting below an AI answer, a map pack and ads gets seen by fewer people, however well it's written. Rewriting helps you reach your position's ceiling. It doesn't raise the ceiling.
• It isn't a ranking tactic. Google has said publicly that clicks aren't a direct ranking factor. Microsoft's documentation on how Bing delivers results takes a different line: "Bing also considers how users interact with search results." Neither statement turns a CTR gain on your page into a promised ranking change, so judge a rewrite by the visits it earns.
• It can go down when you fix it. Take the overpromise out of a title and CTR may fall while the visitors who do arrive are the right ones. That's a correct result, not a failed one.

About our own product: SERP Clicks sends paid people who search your keyword and click your listing. Those clicks count in Search Console, so a page receiving them doesn't belong in a rewrite test or its control group. And bought clicks can't make a result more appealing to the real searchers who see it. If nobody chooses your listing, fix the listing first.

Run this for a page before you publish or rewrite it.

1. Home page WebSite markup names the site the way the home page does, with an alternateName fallback.
2. Favicon is square, larger than 48 pixels, in a listed format, crawlable, and at a URL that won't change.
3. URL path and breadcrumb names describe the page type in plain words, and the markup matches the visible trail.
4. The </code> leads with the searcher's object and situation, differs from the other results for the main query, and isn't counting on a brand suffix being shown.</li> <li>The title agrees with the page's <code><h1></code>, and nothing in it overstates the page.</li> <li>The meta description is unique to the page, confirms the title with concrete detail, and isn't a keyword list.</li> <li>The first sentence under every heading would work as a snippet for a query that section answers.</li> <li>A visible, labeled date matches <code>dateModified</code>, isn't in the future, and changed only because the page did.</li> <li>Any structured data describes things visible on the page, with no invented reviews and no FAQ or HowTo markup added in the hope of a display.</li> <li>The displayed result was recorded before the change, and will be recorded again before any CTR is read.</li> </ol>