← blog / SEO & GEO

Query Fan-Out Keyword Research: One Search, Ten Queries

·updated ·16 min readseogeoai-modekeyword-researchcontent-strategy
Abstract Magic Share brand graphic: a field of dots with a highlighted cluster in green and teal on a near-black grid, with the Magic Share name and the label "seo".

When someone asks Google's AI Mode a question, Google does not run that question. It breaks it into subtopics and fires a batch of searches on the user's behalf — an average of 10.7 per prompt in Seer Interactive's study of 501 Gemini 3 prompts, and 11.7 for software topics in Nectiv's larger sample of roughly 9,000 prompts.

Here is the part that breaks the way most of us plan posts: in Seer's data, 95% of the queries Gemini generated had zero global search volume. They will never appear in your keyword tool, at any volume threshold, ever. So query fan-out keyword research starts one level up from the volume column. You choose a topic, map the eight-to-fifteen sub-questions it fans into, and then decide which of them a single post can answer in full.

That is a planning change, not a trick. But if you are still sorting a keyword tool by volume and assigning one head term per post, you are optimising for the query a human types. The thing that retrieves your page is a machine typing ten queries you never saw.

In short:

  • One AI Mode prompt becomes roughly ten searches, and almost none of them show up in keyword tools.
  • The strings never repeat, but their shapes do: priced, dated, brand-named, compared and reviewed.
  • Map those shapes into one thorough post with self-contained sections, not one thin post per sub-query.

What query fan-out actually is

Google described the mechanism itself. In its own words, "AI Mode uses our query fan-out technique, breaking down your question into subtopics and issuing a multitude of queries simultaneously on your behalf." Deep Search uses the same technique "taken to the next level," and can issue hundreds of searches for one prompt.

Search Central confirms this is not an AI Mode exclusive. Both AI Overviews and AI Mode "may use a 'query fan-out' technique — issuing multiple related searches across subtopics and data sources — to develop a response." That, Google says, is how it can show a wider and more diverse set of links. Its newer guide to optimizing for generative AI search gives a concrete example: a query like "how to fix a lawn that's full of weeds" might fan out into "best herbicides for lawns", "remove weeds without chemicals", and "how to prevent weeds in lawn".

The sub-queries are not random rewordings. A Google patent often linked to fan-out, US11663201B2, Generating query variants using a trained generative model, lists eight variant types: equivalent, follow-up, generalization, specification, canonicalization, language translation, entailment and clarification. A separate "Thematic Search" patent, filed December 2024, describes generating narrower sub-themes and cascading into sub-sub-themes — expand outward, then drill down.

What happens to the results matters as much as the expansion. Mike King's teardown of AI Mode documents retrieval at the passage level over dense vector embeddings rather than at page level. In his words, "every query, subquery, document, and passage is converted into a vector embedding," and the model fans out across a hidden web of synthetic queries, looking for ideas that can complete a "reasoning chain." So a single section of your post, not the post as a whole, is the unit that gets retrieved. Hold that thought. It is what makes one thorough post a legitimate answer to a fan-out rather than a compromise.

And the audience is real. Sundar Pichai told investors in July 2026 that AI Mode had "surpassed 1 billion monthly active users". Search Senior Engineering Director Dounia Berrada sums up the mechanism as "basically doing a dozen searches for you in the time it takes to do one."

The real numbers: one search becomes ten, sometimes nineteen

You will see "8–12 sub-queries" repeated everywhere as settled fact. Treat it as folklore until someone publishes the sample. The measured studies are more useful anyway, and they broadly agree with each other.

Seer tracked 501 prompts through Gemini 3 with forced grounding and found 10.7 fan-out queries on average, ranging from 3 to 28 — a 78% jump from Gemini 2.5, which averaged 6.01. Nectiv's larger run, roughly 9,000 prompts through the Gemini 3 API producing over 60,000 fan-out queries, landed at 9.06, with 5 the most common count. That distribution is the number to plan around: 59% of prompts fired between 5 and 11 sub-queries, 24% fired between 12 and 19, and the maximum was 28.

Sixteen is real, then, but it is upper-quartile behaviour rather than the average — roughly one prompt in four fans out that far. Vertical matters too, and it matters in your favour if you sell software: Nectiv clocked software topics at 11.7 sub-queries per prompt, travel at 10.8, careers at 9.8 and local at 3.79. Someone researching a tool triggers three times the machine-side research that someone looking for a nearby plumber does.

Engines differ as much as verticals. EMGI's July 2026 pilot on SaaS buying prompts recorded ChatGPT averaging 6.3 sub-queries per searching prompt, well under Gemini 3's 10.7. It found the bigger variable is whether the model searches at all. On those prompts Perplexity searched 100% of the time, Claude 81%, ChatGPT 42% and Gemini's assistant just 6%, answering 15 of 16 prompts from memory. EMGI calls this a small pilot, so read the percentages as directional. When a model answers from memory, nothing you publish can influence that answer; only what it already absorbed about you counts. That is the boundary around everything below.

Why the volume column has gone blind

Two findings from the Seer data do the real damage. The first is the 95% with zero global search volume. The second is quieter and worse: there was "only 1% overlap across the full fan-out dataset, so almost every query Gemini generated was unique," in the words of Seer R&D lead Nick Haigler. Even if you harvested ten thousand real fan-out queries, you would not be holding a keyword list. You would be holding samples from a distribution that never repeats exactly.

None of this is new, which should be reassuring. Ahrefs finds keywords with fewer than ten monthly searches "account for almost 93% of our U.S. keyword database", 2.3 billion of them. The same piece notes that roughly 15% of daily Google searches have never been searched before. And one page serving hundreds of queries was already normal: Ahrefs' study of three million searches found the average #1 ranking page also ranks in the top 10 for nearly 1,000 other keywords, median around 400. That study is from 2017. "One post, one keyword" was a bookkeeping convention long before AI Mode existed.

Human behaviour drifted the same way. Robby Stein, VP of Product for Google Search, wrote in August 2025 that early AI Mode users ask questions two or three times the length of traditional search queries. Longer human questions, expanded into more machine questions, resolved against passages rather than pages — and your volume column can see only the first layer of that.

The strings are noise; the shapes are the signal

Here is the good news, and it is the whole method. You cannot target individual fan-out strings. But the shapes those strings take are strikingly consistent across four independent datasets and two model families.

Seer and Nectiv both measured an average fan-out query length of 6.7 words, with Nectiv finding 77% between five and eight words. Nectiv's most common n-grams were year references ("2024/2025," 6.26%), "reviews" (2.14%), "vs." (1.41%), "free" (1.05%) and "top" (1.05%); Seer found 21.3% of queries contained a year and 26.4% named a brand. Centium's much larger sample covers 322,485 fan-out searches across over 130 brands, captured from ChatGPT and Gemini over three months in 2026. It reports that "four in five of the prompts AI researches (80%) trigger at least one search for the best or top option." More than a third (38%) trigger a review search, about 15% trigger a guide search, and 46% include a year-specific search. On SaaS buying prompts specifically, EMGI's 127 captured sub-queries were 86% year-stamped, 69% brand-named and 51% about pricing.

One conclusion, then: the machine researching your category asks priced, dated, brand-named, compared and reviewed questions in five to eight words. That is not a keyword list. It is an outline.

Fan-out shape Evidence What it wants on the page
Priced ("X pricing 2026") 51% of SaaS sub-queries (EMGI) A real number or named range, and what changes it
Year-stamped 21.3% (Seer), 86% of SaaS sub-queries (EMGI) A visible date and current figures, not "recently"
Brand-named 26.4% (Seer), 69% (EMGI) Named competitors and integrations, described fairly
Comparative ("vs") 1.41% of n-grams (Nectiv) A side-by-side with an honest "choose them if…"
Best / top 80% of researched prompts (Centium) A shortlist with stated selection criteria
Reviews Over a third of prompts (Centium) Attributed third-party evidence, not self-praise
How-to / guide ~15% of prompts (Centium) Numbered steps that stand alone as a passage

EMGI's captured example makes it concrete. Run "what is the best CRM for a small B2B sales team?" through GPT-5.2 with live search and it generated eight sub-queries, including "HubSpot Sales Hub pricing 2026 Starter Professional" and "Pipedrive pricing 2026 plans." Nobody types those. A machine does, on its way to answering the question somebody did type.

A keyword-selection method that survives fan-out

Cyrus Shepard's five-step fan-out framework is the closest thing the field has to a practitioner standard, and it adapts cleanly to a one-person content operation.

1. Start from a query you already rank for. Not a volume winner, but an existing foothold you can find in Search Console. Fan-out rewards pages that are already retrievable, and Shepard's blunt version is worth taping to the wall: "You really want to create content that's worthy of ranking in the top 10-20 of Google's search results."

2. Harvest observed fan-outs before simulated ones. Most fan-out generators are a model guessing at what Google might ask; two sources give you queries a model really issued. Bing Webmaster Tools' AI Performance report, launched in public preview on February 10, 2026, includes a Grounding Queries metric. Microsoft defines it as "the key phrases the AI used when retrieving content that was referenced in AI-generated answers", and it is free and about your own pages. Second, the Gemini API returns the searches it ran in a google_search_call object. Grounding gives you 5,000 free search requests a month on Gemini 3 and newer models, then $14 per 1,000, billed per search. At roughly ten searches per prompt, that is about 500 prompts a month of observed data.

3. Use simulators as a fallback, knowingly. iPullRank's Qforia, Wellows' Query Fan-Out Generator, Otterly's fan-out analysis and Locomotive's in-house fan-out coverage analysis are useful for structure. Qforia classifies its output by query type, user intent and recommended format, from tables to checklists. DEJAN's Query Fan-Out tool goes a step further and shows the searches a grounded model runs, which is closer to observation, though still not Google's own expansion. Log all of these as hypothesis, not observation.

4. Cluster into eight to fifteen sub-questions, then cut. Group the harvested strings by the shapes in the table above and de-duplicate ruthlessly. Then apply the honesty test: which of these can this post answer in full, with a number or a procedure, from evidence you already have? A sub-question you can only wave at belongs to a different asset, or nowhere.

5. Write each surviving sub-question as a self-contained section. Since retrieval happens at passage level, every section needs its claim, its number and its scope in the same place. Aim for a paragraph that still makes sense with the rest of the post deleted. That is about completeness, not length: Google's guide says there is no requirement to break content into tiny pieces. iPullRank's worked example of "What's the best way to save for retirement?" fans into "what are the different types of retirement accounts?", "how much money should I save for retirement?" and "common retirement savings mistakes to avoid". In its analysis, one is served by a table, one by a calculator, one by a list. Match the format to the sub-question instead of flattening all of them into prose; our guide to the on-page patterns that earn AI citations goes deeper on shaping those paragraphs.

Steps 4 and 5 are where a solo operator runs out of hours. Answering twelve sub-questions properly, each with its own sourced evidence, is a research job wearing a writing job's clothes. That is the shape of work Magic Share's drafting and verification pipeline was built around: drafts sized to the query, from 600 to 4,000 words, with every cited URL checked by the agent, a verification report attached, and the draft waiting for a human to decide which sub-questions the post is entitled to answer.

If you need practical methods for gathering primary data without customers, see our guide on original research for a SaaS blog with no customers.

When to split, and when not to

The obvious wrong turn is to read a twelve-query fan-out as a twelve-post content calendar. Three reasons not to.

First, the policy language. Google now addresses fan-out directly. Its generative AI optimization guide warns that it "might be tempting to create separate content for every possible variation of how people might search," fan-out queries included. Doing that primarily to manipulate rankings or AI responses, it says, "violates Google's scaled content abuse spam policy." The older doorway rule points the same way: doorway abuse is when "sites or pages are created to rank for specific, similar search queries", a fair description of one thin post per sub-query. Our closer read of what Google's scaled content abuse policy actually bans covers where that line sits; the short version is that value per page is the test, and method is irrelevant.

Second, the correlation data. Ahrefs' study across 75,000 brands found the number of site pages among the weakest correlates of AI visibility, at roughly 0.194, while YouTube mentions (~0.737) and branded web mentions (0.656–0.709) sat far higher — with the authors' own correlation-isn't-causation caveat attached. Publishing more pages is the lever with the least visible relationship to the outcome you want.

Third, what worked in practice was consolidation. Resource Guru, a SaaS in resource management, spent Q1 2026 refreshing existing posts across three clusters: project time tracking, agency software and resource capacity planning, 44 keywords in total. The team updated angle and UX and added expert insight, entity markup, FAQ blocks and schema. Share of voice on those keywords went from 1.82% in January to 9.84% in March, ten of them earned AI Overview citations, and organic traffic moved from 17,614 to 22,546 over the same period. It is an agency-published case study, so treat it as directional, but the direction matches Shepard's advice to add fan-out coverage to pages that already exist.

The rule that falls out: sixteen sub-queries is an outline, not a calendar. Split only when a sub-question needs a different format or serves a different stage of intent — a pricing page, a calculator, a migration guide. If two candidate posts would answer overlapping sub-questions, you have invented a cannibalisation problem instead of solving a coverage one. One seed, many shoots; you don't plant sixteen seedlings in one pot.

Which fan-out branches a SaaS can actually win

Not every sub-query is winnable, and pretending otherwise wastes the scarcest resource you have. The "best [category] software" branch is largely spoken for. One June 2026 analysis logged 1,259 AI Overview citations across 100 commercial "best software" searches in 20 B2B SaaS categories. Third-party best-of lists took 63% of citations while the recommended product's own site took just 12%. It is an agency study, so directional, but it is consistent with Centium finding that best/top lists make up 44.6% of Gemini's cited URLs. Semrush's cross-platform work adds that Reddit dominated across AI Mode, AI Overviews, ChatGPT and Perplexity. Writing your own "best CRM tools" listicle to win that branch means beating every review site on its home turf.

The branches vendors win are the ones only vendors can answer. EMGI found 56% of ChatGPT's citations went to vendor-owned pages, mostly pricing and comparison content — which lines up with 51% of its captured sub-queries asking about pricing. So the instruction is specific: spend your effort on published pricing with the conditions that change it, fair head-to-head comparisons, integration and migration questions, limits and specs, and implementation timelines. Those sub-queries fan out constantly, and your own documentation is the most authoritative surface on the web for them.

Since comparison content is where vendors tend to win these citations, the page has to earn trust, and this guide to writing a fair SaaS competitor comparison page walks through how to compare rivals without tilting the scales.

There is a structural reason this is a better bet than it used to be. In a study published in March 2026, Ahrefs found that 38% of AI Overview citations come from pages ranking in the top 10, across 863,000 keyword SERPs and 4 million AI Overview URLs. That is down from about 76% in their July 2025 study of 1.9 million citations. The author reads the change as AI Overviews "relying less on the direct search results and more on the sources showing up in fan out query SERPs." Semrush's 5,000-keyword comparison points the same way: AI Mode has the loosest relationship with Google's top 10, at roughly 54% domain and 35% URL overlap. Ranking first for the head term matters less than it did; being the best available answer to one specific sub-query matters more. For a small site, that arithmetic is a gift.

How to check whether you were right

Measurement is the weak link, and it is asymmetric. Pages appearing in Google's AI features are "included in the overall search traffic in Search Console," reported within the "Web" search type. Google now also offers Search generative AI performance reports, launched for a subset of sites in June 2026 and rolled out to all websites on August 31, 2026. They show impressions, pages, countries, devices and dates for features such as AI Overviews and AI Mode, but no queries, so there is still no fan-out view.

Microsoft hands you grounding queries for free, though Bing's report has real limits. Its launch post says citation counts show how often a page is cited, not where it sits in the answer or how important it is, and that the grounding queries are a sample. On June 16, 2026 Microsoft began rolling out intents, topics, citation share and a compare view in preview. For a solo marketer, the smaller engine is currently the better instrument for fan-out. Use it as a proxy: if your page is retrieved for six of the twelve sub-questions you mapped, the missing six are your next editing session.

Then the question underneath all of this: is any of it worth the hours? Pew's tracking of 900 U.S. adults across 68,879 searches in March 2025 found users clicked a traditional result in 8% of visits to pages with an AI summary versus 15% without, and clicked links inside the summary in just 1% of visits. Fewer clicks, plainly. Semrush argues the survivors are worth more: in its July 2025 study, the average AI search visitor is 4.4 times as valuable as a traditional organic visit, based on conversion rate. That figure is contested. Amsive's study of 54 sites found LLM referrals converting at 4.87% against 4.60% for organic, a difference that was not statistically significant. Don't build a business case on 4.4x. Our take on what a blog is still for when most searches end without a click sets out the version we would defend: the job shifted from capturing traffic to being the source the answer is assembled from.

The last word belongs to Google. Its Search Liaison, Danny Sullivan, has argued that SEO for AI is still SEO, and that optimising narrowly for one AI system leaves you forever catching up as those systems change. Search Central says something similar: there are "no additional requirements to appear in AI Overviews or AI Mode, nor other special optimizations necessary." Its AI optimization guide adds that you don't have to worry about whether you have "captured every variation of how someone might seek content like yours." Read alongside the fan-out data, that is not a contradiction. Fan-out doesn't change what wins; it changes how many distinct questions one page has to satisfy to keep being found. Mapping durable question shapes instead of chasing this month's model-specific strings is exactly the non-narrow approach Sullivan is recommending.

FAQ

How many sub-queries does one search become? Measured averages cluster between 9 and 11: Seer recorded 10.7 per Gemini 3 prompt and Nectiv 9.06, with software topics higher at 11.7. Sixteen or more happens in roughly the upper quartile, but it isn't typical.

Should I still use keyword volume at all? Yes, for the entry query your title and framing target. Use volume to pick the door and fan-out shapes to furnish the rooms, since 95% of fan-out queries had zero search volume in Seer's data.

Are fan-out simulators trustworthy? They are useful for structure, but none of them shows Google's own expansion. For observed data, use Bing Webmaster Tools' grounding queries and the google_search_call object the Gemini API returns.

Is it better to write one thorough post or several short ones per fan-out? Consolidate by default: Google's guide says separate content for every fan-out variation, made to manipulate rankings, breaches its scaled content abuse policy. Split only when a sub-question needs a different format, such as a pricing page, a calculator or a migration guide.

The short version

One search became ten, most of the ten are invisible to your tools, and the shapes they take (priced, dated, brand-named, compared, reviewed) are stable enough to plan around. Pick your entry query the way you always did, then spend the planning time mapping the eight-to-fifteen sub-questions behind it and cutting the ones this post cannot fully answer. That map is your keyword research now.

If the bottleneck is doing that research every week rather than knowing how, that is the gap Magic Share was built for: the agent studies your niche, drafts the post with every cited link checked and a verification report attached, then leaves it as a draft until you approve it. Your first three posts are free, and no credit card is needed.

If you publish many cited links, run a periodic link rot audit: your blog's dead outbound links to find and fix broken or redirected references that erode your post's evidence and credibility.

More on SEO & GEO →

by email

Get new posts in your inbox

A short digest when new posts go up. Confirm once, unsubscribe from any email.

We use your address only for this digest. Privacy

Keep reading

  1. ·20 min read

    Your First 20 Blog Posts: A SaaS Blog Topic Map

    What to blog about when your SaaS blog has zero traffic: a 20-post topic map, ordered so each post makes the ones published before it easier to rank.

    seocontent-strategycontent-planningkeyword-researchinternal-linking

  2. ·16 min read

    How to Write a Fair SaaS Competitor Comparison Page

    How to write a competitor comparison page for SaaS that names rivals fairly, sources every claim, and converts — with the evidence and update cadence.

    content-strategyseosaas-marketingcomparison-pagesgeo

  3. ·21 min read

    Topic Clusters for a 30-Post Blog: Internal Linking

    How to shape a 30-post blog into topic clusters with bidirectional internal links, and why anchor variety beats page count for earning AI citations.

    seointernal-linkingcontent-strategycontent-planninggeo

Want posts like this for your site?

Magic Share researches, writes and fact-checks posts like this for any site — point it at your URL and review your first draft today.

Plant your first post