A featured snippet is the boxed answer at the top of a results page, pulled from a page Google already ranks. It has been called position zero for years, which is accurate about placement and misleading about value.
The uncomfortable part: a snippet that fully answers a query often means the searcher does not click. You occupy the most prominent position on the page and receive less traffic than the site below you. That trade is worth understanding before optimising for it.
Four Formats and What Triggers Each
Paragraph snippets are the most common and answer definitional or explanatory queries. Roughly 40 to 60 words of direct answer.
List snippets answer process or ranking queries, pulled from ordered or unordered lists. Numbered for sequences, bulleted for non sequential sets.
Table snippets answer comparison or data queries and are pulled from genuine HTML table markup, which is why they are covered in the data tables guide. A grid built from styled divs cannot produce one.
Video snippets surface a timestamped segment, which requires the video platform to support chapter markers.
The format Google chooses follows the query. You cannot force a table snippet on a definitional query, and matching your content format to the query's natural format is most of the work.
You Have to Rank First
Snippets are pulled almost exclusively from pages already ranking on page one, typically within the top handful of results.
This makes snippet optimisation a second stage tactic rather than a shortcut. A page ranking on page three cannot win a snippet regardless of formatting, and effort spent formatting it is wasted until the underlying ranking exists.
The corollary is that snippet opportunities are best found among terms where you already rank fifth to tenth. Those pages are eligible and are losing the snippet to someone formatting better, which is a solvable problem.
The Format That Wins Paragraph Snippets
A question as a heading, immediately followed by a direct answer of roughly 40 to 60 words, then the elaboration.
That structure works because it gives Google a clean extractable block that answers the query without requiring it to assemble something from scattered sentences. It is the same principle covered in the BLUF guide, applied to a specific SERP feature.
The failure mode is burying the answer. A section that opens with context and delivers the answer in the fourth sentence gives Google nothing clean to lift, and it will take a competitor's cleaner block instead.
Answer length matters. Substantially under 40 words often reads as incomplete. Substantially over 60 gets truncated mid thought, which looks worse than not appearing.
List Snippets Require Actual List Markup
Ordered lists for sequential processes, unordered for non sequential sets, using real ol and ul elements rather than paragraphs with dashes.
Heading structure matters here for the same reasons covered in the on page SEO guide. Google frequently assembles list snippets from H2 or H3 headings across a page rather than from a single list element, which is worth knowing because it means your heading structure is itself a snippet candidate. Descriptive headings that read as steps produce better assembled snippets than generic ones.
Keep list items short enough to display. Google typically shows a limited number of items and truncates long ones, so front loading the meaningful words in each item matters.
Finding Opportunities Without Guessing
Search Console gives you the raw material. Filter queries to those where your average position sits roughly between two and ten, then check which of those queries currently display a snippet.
Those are eligible pages losing a feature they could win. That list is usually short and immediately actionable, which is the same method covered in the Search Console guide for finding underperforming queries generally.
Question format queries trigger snippets disproportionately, so filtering your query list for terms containing what, how, why, and similar surfaces candidates quickly.
Snippet Ownership Is Unstable
Snippets change hands frequently and sometimes without any change on either page. Google tests alternatives, adjusts what it considers the best answer, and occasionally removes the snippet entirely for a query.
That instability means treating snippet ownership as a durable asset is a mistake. It is worth winning and not worth building a forecast around.
It also means monitoring matters. A snippet lost without a corresponding ranking change usually indicates a competitor reformatted, which is diagnosable by looking at what now occupies it.
The Traffic Question, Honestly
Older research on whether snippets increase or decrease clicks reached mixed conclusions, and the landscape has changed enough that most of it is now stale.
The defensible position: snippets on queries with a complete short answer reduce clicks, because the searcher got what they needed. Snippets on queries where the answer is a starting point increase clicks, because prominence attracts and the answer is insufficient.
Which means the tactical decision is query specific. Winning a snippet for a definitional query about a term probably costs you traffic. Winning one for a query where your answer creates further questions probably gains you traffic and brand exposure.
I would still pursue both, because the alternative is a competitor occupying that space and being the source a searcher remembers.
How AI Overviews Changed the Calculation
On many queries, an AI Overview now sits where a featured snippet would, or alongside it, drawing from multiple sources rather than one.
The formatting that wins snippets is largely the same formatting that gets content extracted into AI answers, which is the overlap covered throughout this blog. Clean question and answer structure, genuine list markup, real tables, and direct answers near the top of a section serve both.
That overlap is the strongest current argument for snippet optimisation. Even where the snippet itself becomes less valuable, the structural work transfers to a surface that is growing rather than shrinking.
Opting Out Is Possible and Rarely Correct
The nosnippet and max-snippet directives let you prevent or limit snippet display.
Publishers occasionally use these to force clicks by withholding the answer. It usually backfires, because Google displays a competitor's answer instead and you lose both the snippet and the prominence.
The legitimate use is preventing sensitive or paywalled content from being excerpted, which is a genuine reason and a narrow one.
People Also Ask Boxes Are the Adjacent Opportunity
The expandable question boxes that appear on most results pages draw from the same kind of content as paragraph snippets and are considerably easier to win, because there are several slots rather than one.
They also expand dynamically. Clicking one generates further related questions, which means a single results page can surface a large number of question and answer opportunities across a session.
The practical value is twofold. Each box is a visibility slot, and the questions themselves are a free research source showing what Google associates with your target query. That is a better input for content planning than a keyword tool, for the reasons covered in the keyword research guide.
Winning them requires the same structure: a question as a heading, a direct answer immediately below, kept short enough to display whole.
Schema Does Not Win Snippets
Worth stating because it is a persistent misconception. Featured snippets are extracted from page content, not from structured data. FAQPage markup produces its own display treatment where supported and does not cause a featured snippet.
Schema remains worth implementing for the reasons covered in the schema guide. Snippet capture is not among them.
Tracking Both Surfaces
Search Console's search appearance filter isolates impressions where your result appeared as a rich result, which is the closest native measure available.
The NotionCue Prompt Tracker covers the parallel question of whether the same content gets cited in AI generated answers, which is increasingly where the same formatting work pays off.
Start your free NotionCue trial and check whether the pages winning your snippets are also the pages being cited. They frequently diverge, and the divergence tells you something about which formatting each surface prefers.
Fastest snippet audit available: search your top ten target queries and note which currently show a snippet and who owns it. Where a competitor owns one and you rank nearby, open their page and look at how the answer is formatted. The difference is usually structural rather than qualitative.
Common Questions
Can a page win multiple snippets?
Yes, for different queries. A comprehensive page with several well formatted question and answer sections can hold snippets for multiple related terms simultaneously.
Does winning a snippet remove your normal listing?
Generally yes. The snippet typically replaces rather than supplements your organic listing for that query, which is part of why the traffic effect is ambiguous.
How long does it take to win one after reformatting?
Once the page is recrawled, which can be days to weeks depending on crawl frequency. Snippet selection updates faster than ranking generally, so a well formatted answer on an already ranking page can appear relatively quickly.