How to Optimize Content for Featured Snippets (and Whether It Still Pays Off)

Featured snippets showed on 15.41% of US desktop results in January 2025. By June they showed on 5.53%. Ahrefs, which tracked one million results, says AI Overviews very likely took their place on many queries. So should you still optimize for them? Yes, on the right queries, and not with the old playbook.

The plan is simple. Learn what Google says you can and cannot control, run the numbers before you chase a box, pick the queries where a snippet pays and write answers Google can lift.

Then measure clicks, not trophies. The examples are hypothetical, and no tactic here can promise you a snippet.

Where Google Draws the Line on Control

Start with the rule most guides bury. Google's documentation says you cannot mark a page as a featured snippet.

Its systems decide whether a page makes a good snippet for a search and, if so, elevate it. No markup, tag or plugin switches it on. What you control is quality and format.

Google's help page says snippets are chosen by how well they answer the question and how helpful they are, and that they appear at the top of results, in People Also Ask and beside Knowledge Graph panels. Your job is to be the clearest short answer.

Clicks behave in a useful way. Google says clicking a snippet takes the user straight to the quoted section of your page, scrolling there automatically.

So the passage that gets quoted should sit under a clear heading, and the rest of that section should give people a reason to stay.

You can also opt out. The nosnippet rule blocks all snippets, data-nosnippet hides marked text, and a lower max-snippet value makes a featured snippet less likely, though Google says it may not work every time. Use these only if a box costs you more than it earns.

Do They Still Pay? The Numbers Before and After AI Overviews

Two datasets matter, and they come from different eras. Ahrefs' original study of two million snippets, last updated in December 2025, found featured snippets on about 12.29% of queries in its US data.

It also found that 99.58% of featured pages already ranked in the top 10, and only 30.9% sat at the very top.

The click data is old and worth reading carefully. When a snippet sat at the top, it got about 8.6% of clicks, the result below it got about 19.6%, and a regular number one with no snippet above it got about 26%. Ahrefs itself told readers to treat these averages with caution.

Take 1,000 monthly searches. A plain number one earns about 260 clicks. With a snippet above it, that result earns about 196, a loss of 64, and the snippet owner earns about 86.

A snippet is not free traffic. It moves clicks around, and it can lower the total.

Then AI Overviews arrived. In a later Ahrefs analysis of one million US desktop results, featured snippets fell from 15.41% of results in January 2025 to 5.53% in June, a 64% drop, while AI Overviews grew sharply.

Ahrefs wrote that Overviews seem very likely to have superseded snippets on many queries. Read that with care. It is one tool's sample, one device and one six-month window, and other trackers report different levels.

But the direction is clear. The classic box is rarer, so chase it selectively, query by query, and judge it by clicks.

Also Read: AI Overviews Now Reach 2.5 Billion People: What It Means for Your Blog Traffic

Long Tails Trigger Boxes More Than Wording Does

Most people assume question words trigger snippets. Ahrefs' data says otherwise.

The majority of featured snippets in its sample came from long-tail keywords and the specific search queries they represent, and most of the queries that triggered one did not fall into the question, comparison or preposition groups.

Examples included plain phrases such as a software version and a compatibility check. That finding changes the tactic. You cannot target snippets by adding a word like how or what to a heading.

You win them by covering a topic so thoroughly that a short, quotable answer exists for many specific questions. Depth and clarity produce snippets, while wording tricks mostly do not.

It also explains why one good page can own many boxes. Ahrefs found single pages featured for thousands of queries.

A page that answers a cluster of related questions cleanly earns more passages worth lifting than a thin page built around one phrase.

Pick the Queries a Snippet Could Move

Chase a snippet only when three things line up. You already rank on page one, because 99.58% of featured pages did in Ahrefs' data. The query is a clear question or a snippet already exists. And the click value justifies the work.

Anything else is a hobby. Use your own Search Console data to estimate the upside, applying the same measurement discipline as tracking your Google rankings without breaking the bank.

For each candidate query, take monthly searches times the gap between a snippet's old average share of 8.6% and your current click rate.

It is rough, because 8.6% is an old average, but it ranks your list. A hypothetical list shows the method. The expected figure applies a 25% chance of winning, which is my own assumption:

• Query C: 3,000 searches and a 1.5% click rate gives an upside of about 213 clicks a month, or about 53 expected.
• Query A: 1,200 searches and a 3% click rate gives about 67, or about 17 expected.
• Query B: 400 searches and a 5% click rate gives about 14, or about 4 expected.
• Query E: 150 searches and a 2% click rate gives about 10, or about 2 expected.
• Query D: 900 searches and a 9% click rate gives about minus 4, so skip it.

Notice the pattern. The best targets are queries with high volume where you get few clicks today. If you already own most of the clicks at number one, a snippet may only reshuffle them.

Skip those, and spend your effort where the click rate is low. Some queries are poor targets even when a box exists.

Skip anything where the answer is so short that nobody needs to click, such as a single fact. Skip topics where you cannot add anything new. And be careful with health, money and safety questions, where an error costs a reader more than it earns you.

Also Read: AI Research Tools: How to Find, Summarize and Verify Information

Write Short Passages Worth Lifting

A snippet is a short passage lifted from a page, so write one on purpose. Put the question in a heading, then answer it in the first one or two sentences beneath.

Google gives no exact length, and its own page says the minimum varies, so I use 40 to 60 words for definitions as a working rule.

Take a hypothetical example for the query what is a canonical tag. The weak version spends its first two sentences on the history of duplicate content. The strong version answers at once, in under 50 words:

• The answer block (48 words): A canonical tag is a line of HTML in a page's head that tells search engines which URL is the main version when several URLs show the same or very similar content. It consolidates ranking signals on one address and reduces the confusion that duplicate pages can cause.

A how-to query needs a different shape. Consider the query how to add a canonical tag. The block below is a short intro line followed by an ordered list, written so each step stands alone:

1. Choose the preferred URL: pick the one version of the page you want to appear in results.
2. Add the tag: place a link element with the rel canonical value in the head of each duplicate page.
3. Point to one address: use the full preferred URL so every duplicate names the same page.
4. Check it: inspect the page in Search Console and confirm Google picked your chosen URL.

Use real ordered-list markup for steps, not numbers typed into a paragraph, so the sequence is clear to readers and to software.

That is my own practice, not something Google specifies, and it costs nothing. Keep the intro line short, and put any caveats after the list. Match the format to the question.

A question that asks for a definition suits a short paragraph. A question that asks how to do something suits a numbered list in order.

A question that compares options suits a small table or a tight list. Check the snippet now on show and mirror its shape.

Then beat it. Make your answer clearer, more complete or more current than the one on display, and keep every fact accurate.

Ahrefs' author advised matching the existing format and giving Google a better snippet than the one listed, and that advice holds even as the box gets rarer.

Also Read: How to Actually Rank Your Blog on Google: No Tricks, No Shortcuts

Shape Each Page Around One Question per Section

One question per section works best. A page that asks and answers a dozen related questions under clear headings gives Google many passages to choose from, and it gives readers a page they can scan. Keep each answer near the top of its section, followed by detail. Do not count on markup.

Google says there is none for snippets, and in an old GetStat study that Ahrefs cited, schema markup appeared on 15.7% of featured URLs versus 19.6% of regular results. Structured data has other uses, but it is not a snippet switch. Do not count on backlinks either.

In Ahrefs' data, the backlink metrics of featured pages were on par with the rest of the top 10, so Google was not just picking the strongest page. It picked a page that answered the question well, which is the part you can improve this week.

Protect the basics. You still need to rank in the top 10 first, so a snippet plan sits on top of a solid page, not in place of one. If a page is not ranking yet, fix that before you work on snippets.

Also Read: Why Search Everywhere Optimization Is Replacing Single-Channel SEO for Small Businesses

People Also Ask: The Side Door to the Same Box

Featured snippets are not only the box at the top. Google's documentation says they can also appear within the related questions group, known as People Also Ask.

Its help page adds that snippets are helpful when you use a phone or talk to your device, so these small answers travel further than one results page.

Use that as a research tool. Search your target query, note the questions Google lists, and answer the relevant ones under their own headings, with a short block first.

Each answer is one more chance for a passage to be lifted, and each one helps readers who arrive with a follow-up question.

Keep it relevant. Answer the questions your reader would plausibly ask next, not every question Google offers.

A page stuffed with thin answers invites the weak, repetitive look that hurts quality, so pick the five or six that match what the page is for.

Policy Rules That Can Remove a Box

Google's policies apply here. Its help page says its systems prevent snippets that violate its content policies, and a feature-specific rule says snippets on public-interest topics, such as civic, historical, medical and scientific subjects, should not contradict well-established or expert consensus. There is a site-level consequence.

Google says that if a site has many snippets that break the rules, it stops showing featured snippets from that site.

If you write in sensitive areas, source your claims carefully, keep them current and never stretch a statement to fit a box.

Keep a source note for every factual claim in an answer block. Record where the fact came from and when you last checked it, and review the note whenever the topic changes.

It takes a minute per claim and it makes updates quick, which matters when a snippet can be removed or replaced.

Quality beats tricks. Pages with real expertise, clear sourcing and accurate facts are the ones worth quoting, and the same habits protect you if a snippet disappears. A snippet is a lease, not a deed, so build every page to earn clicks without one.

Also Read: E-E-A-T Explained: How to Build Topical Authority

Measure Clicks, Not Trophies

Success is not owning a box. It is more clicks and more customers from the queries you worked on. Before you change a page, record its impressions, clicks and click rate for the target query from Search Console for the last 28 days. Then change one thing at a time.

Check the live result by hand as well. Search the query in a private window, note whether a snippet appears, who owns it and in what format, and write it down.

Do this weekly for your ten target queries, because snippets change often and tools can lag. Judge each change after four to six weeks, which is my own estimate, against your baseline.

If clicks rose, keep the format. If the box appeared and clicks fell, treat it as the old data warned, and decide whether that box is worth keeping. Plan to maintain what you win.

Snippets change over time, so a box you hold today may move to a competitor next month. Review your target pages quarterly, keep the facts current and update the answer when the topic changes. A stale answer is an invitation for someone to replace you.

If a competitor takes a box from you, compare their passage with yours before you rewrite anything. Check whether theirs is clearer, more current or in a better format. Often a small edit to the answer block is enough, and a full rewrite rarely is.

Picture a hypothetical read-out. A treated page gets 120 clicks from 4,000 impressions, a 3.0% click rate. Four weeks after the rewrite it gets 150 clicks from 4,300 impressions, or 3.5%, which is 25% more clicks. A control page moves from 100 to 103 clicks, up 3%.

The net effect is about 22 points, which is the part your edit may explain. Without the control, you might credit the rewrite for a market-wide shift. Four weeks is short and the numbers are small, so repeat the test before you roll a format across the site.

Also Read: Google Just Made 68% of Searches End Without a Click: Here's How to Adapt

A Four-Week Plan You Can Start Today

Keep it small. Ten queries and one change per page is enough to learn whether snippets pay for your site, and it fits around other work. Run it like this:

1. Week one: export your queries from Search Console and pick ten questions where you rank between positions two and ten.
2. Week two: check each live result, log who owns the snippet and its format, and calculate the upside for each query.
3. Week three: rewrite the answer blocks and headings for the five best targets, and leave the other five as a control.
4. Week four: record clicks and impressions, check the live results again and log what changed.

Keeping five pages unchanged gives you a comparison, so you can tell whether a change came from your edit or from the whole market moving. That matters more now, when Google changes result pages often.

Featured snippets are a smaller prize than they were, but they are still a cheap test. Pick the right queries, write the answer first and trust the clicks. If a box comes, take it. If it goes, the page you built still has to earn its place.

Also Read: GEO vs SEO in 2026: Why Being Citeable Now Matters More Than Ranking