From not cited to published: finding the sub-questions AI answers and choosing which to write

You check your organic rankings and see your core guide sitting in Google's top 10. Or you open Google Search Console and export a couple hundred long-tail queries where your pages earn impressions. Yet when you test the same topics in ChatGPT, Perplexity, or Google AI Overviews, your company is nowhere in the citations.

That disconnect happens because traditional search rankings reward a page for matching one primary query, while answer engines break a prompt into a cluster of narrower follow-up questions and pull passages from whichever pages answer each slice directly. Trying to reverse-engineer those hidden sub-queries by asking another chatbot to guess them, or spinning up a new article for every phrasing variation in your keyword export, only leaves you with a bloated backlog of thin pages.

This six-step guide shows you how to move from "not cited" to a focused publishing plan. Starting with your underperforming top-10 pages, Search Console queries, and buyer conversations from sales calls, support tickets, Reddit, and LinkedIn, you will map the sub-questions buyers actually care about, filter them by commercial relevance and evidence strength, update existing pages first, and build durable topic clusters.

Step 1: Diagnose why a top-10 ranking doesn't earn the citation

A page that ranks in Google's top 10 for a head term misses AI citations when it fails to answer the specific sub-queries an answer engine runs behind the scenes, nearly all of which show zero search volume in keyword tools.

When a buyer submits a prompt, AI search platforms use query fan-out: a retrieval technique that decomposes one user prompt into multiple parallel sub-queries across subtopics and data sources before synthesizing a single answer, as documented in Google Search Central's guide to AI features. Instead of selecting one page that matches the broad keyword, the system retrieves candidate passages for each sub-query and merges them.

The scale of that expansion explains why a single keyword-focused page gets skipped:

• One prompt can trigger roughly a dozen background searches. In first-hand tests, Ahrefs observed Google AI Mode running 5 to 11 searches for a standard shopping prompt (and ChatGPT Deep Research running 420 searches for "buy red phone case"). Larger API studies from Seer Interactive (501 prompts) and Nectiv (more than 60,000 fan-out queries across nearly 9,000 prompts that skewed commercial and transactional) found an average of 9 to 11 fan-out queries per prompt (10.7 in Seer and 9.06 in Nectiv, rising to 11.7 in software). Because both studies pulled queries from the Gemini 3 API with Google Search grounding forced on every run, they show how the model expands prompts when search is required, not how often live AI Mode or AI Overviews trigger web search naturally versus answering from training data.
• Almost none of those sub-queries show up in keyword tools. In Seer Interactive's dataset, 95% of the fan-out queries Gemini generated had zero monthly search volume, and averaged 6.7 words in length. If your page was built strictly around high-volume keywords, it likely leaves out the comparative, implementation, and risk sub-questions the model searches for.
• Beware of unverified formatting formulas. Agency and plugin blog posts from Bigeye Agency and AIOSEO claim that 89.6% of searches split into sub-questions, that AI engines prefer 40-to-80-word passages, that covering related questions makes a page 161% more likely to be cited, and that FAQ schema raises citation probability by 28% to 40% (or 3.2×). Neither source publishes a sample size or methodology for those numbers, so treat them as unverified claims rather than engineering targets.

To diagnose a top-10 page that isn't getting cited, run its primary topic prompt in ChatGPT, Perplexity, and Google AI Overviews. List every sub-topic, comparison criterion, or constraint included in the synthesized response, record which URLs are cited for each slice, and check which of those slices your ranking page leaves unanswered.

Step 2: Skip simulated LLM fan-out and check why synthetic guesses mislead

Asking a chatbot to predict the 5 to 10 sub-queries an answer engine will run stacks synthetic guesses on top of a retrieval process that changes from one run to the next.

When SEO teams try to map query fan-out, they typically rely on one of two workflows: using enterprise platforms (inspecting logged ChatGPT and Perplexity fan-out queries in Ahrefs Brand Radar, or comparing organic rankings against AI answer visibility in Semrush and Conductor to infer missing sub-topics), or prompting ChatGPT or Claude to act as a retrieval system and guess 5 to 10 micro-informational sub-queries for a topic. While tool logs and visibility audits show what happened on specific runs, LLM simulation mistakes plausible phrasing for stable buyer demand.

As Ahrefs notes in its analysis of fan-out behavior, generated sub-queries are synthetic, inconsistent across models and users, and probabilistic across runs. In Seer Interactive's follow-up study pulling Gemini 3 fan-outs twice daily across 13 runs for 100 tracked prompts (1,300 prompt runs total), the model generated 11,029 unique fan-out queries with 0.1% overall query similarity. Only 8 exact-match queries appeared in all 13 runs, and 99% had zero monthly search volume.

You cannot verify those synthetic sub-queries in Google Trends, either. According to Google's Trends FAQ, Trends normalizes each data point by total regional search volume, injects statistical privacy noise that creates artificial spikes on low-interest terms, records low-volume queries as "0," and explicitly filters out internal searches made by AI Mode and AI Overviews. For a deeper look at why volume and Trends fail here, see our analysis of which demand signal should drive AEO content.

Step 3: Use buyers' follow-up questions from calls and communities as your fan-out map

Instead of chasing exact synthetic strings that mutate on every API call, use the follow-up questions, constraints, and objections real buyers raise in sales calls, Reddit threads, LinkedIn discussions, and support tickets as a grounded map of the sub-questions an answer must cover.

Even when an AI model changes its exact search phrasing between runs, the decision criteria buyers need resolved stay consistent. At Rocksalt, we organize surfaced buyer evidence in a three-level hierarchy: prompt → buyer question → theme. Rather than writing a page to match one raw prompt or copying a specific LinkedIn or Reddit post, teams write at the theme level and structure page H1 and H2 headings to match the underlying buyer questions. (For the extraction workflow, see our guide on how to mine sales calls, Reddit and support tickets for the questions buyers ask AI and our comparison of AnswerThePublic vs AlsoAsked vs Reddit mining tools.)

In an anonymized cybersecurity project documented in Rocksalt's prompt-set guide, eight distinct prompts sat beneath a single buyer need theme, Employee-Driven Vulnerability Remediation.

That theme and the guide's classification framework show how buyer conversations map the sub-questions a topic cluster needs to answer:

• Five Problem Research and three Vendor Recommendation prompts in one theme: Within the eight-prompt Employee-Driven Vulnerability Remediation theme, five prompts reflected Problem Research and three reflected Vendor Recommendation. Separately, the guide illustrates how these two classes differ across a security evaluation: Problem Research covers operational and deployment questions (such as how to get employees to reboot laptops without disrupting their workday, what goes wrong when remediation depends on end users, how disruptive employee-led patching is to deploy, or building an evaluation checklist), while Vendor Recommendation covers tool selection and fit (such as which tools automate outreach to device owners after Wiz finds a vulnerability, which products support employee-led remediation without administrative access, or comparing patch follow-up tools for a mid-market company).
• Cross-source corroboration across seven of the eight prompts: Seven of the eight prompts inside the Employee-Driven Vulnerability Remediation theme recurred across multiple independent sources, including sales calls, confirming the sub-questions reflected recurring buyer needs rather than a one-off post.
• From raw buyer remark to specific sub-question: In the documented Buyer Evidence Record from that theme, the paraphrased buyer remark "patching is invasive and interrupts whatever the employee is doing" became the inferred buyer question "How can employees patch and reboot their devices without disrupting their workday?" and the trackable prompt "How can a company get employees to reboot and patch their laptops without disrupting their workday, including appropriate timing, prompts and self-scheduled reboot windows?" backed by six related questions from call transcripts and Reddit.

Step 4: Decide what to publish with commercial relevance × evidence strength

Once you have a list of Search Console long-tail queries and mined buyer questions where your brand isn't cited, publish where both commercial relevance and buyer evidence strength are high, review low-frequency questions that carry strategic deal value, and drop the rest.

Current AEO publishing advice splits into two conflicting camps. One side, reflected in FrictionAI's buyer-question guide, insists you should publish only questions written verbatim by buyers in Reddit, Quora, transcripts, or tickets and discard marketing keyword lists altogether. The other side, seen in checklists from OutreachBloom and 5WPR, tells teams to filter Ahrefs or Semrush keyword exports for phrases of eight or more words and build pages around those strings. Both sides also repeat unsupported benchmarks, such as FrictionAI's claim that mining four sources yields "~60 candidate questions in one hour" or OutreachBloom's assertion that "web mentions drive 10x more AI visibility than on-site adjustments."

You don't need to choose between ignoring search data and trusting keyword exports blindly. Let Search Console queries and 8+-word keyword lists enter your backlog as candidates, then filter every candidate against three operational checks we use at Rocksalt:

1. Commercial intent and buyer stage: Does the question address a problem your company solves across awareness, research, or decision stages, and could the answer influence a shortlist or deal?
2. Existing content coverage: Does your site already answer this question clearly, or is there a substantive gap?
3. Unique expertise angle: Does an internal subject-matter expert have operational data, implementation lessons, or a counterintuitive point of view that isn't already summarized in the top search results? At Rocksalt, internal experts supply those angles via email or inline comments directly inside the document.

Crossing commercial relevance with evidence strength (how often a need recurs across independent buyers and channels) turns an unmanageable query audit into four clear publishing decisions:

Step 5: Update existing pages before you add new URLs for query variations

Close visibility gaps by adding missing sub-question sections and direct answers to your existing pages before publishing a separate URL for every long-tail query variation.

In Rocksalt's teardown of a 59,515-keyword agency plan for a cybersecurity company, 98 seed terms expanded into 17,463 clusters and 15,326 recommended pieces of content, yet missed the deal objection "We already own Microsoft E5. Why do we need to pay someone else to manage it?" Publishing thousands of overlapping pages doesn't earn AI citations. Google Search Central's generative AI guidance explicitly advises against creating separate pages for every query variation, emphasizing non-commodity insights and first-hand expertise instead.

When triaging a list of missing Search Console queries or buyer questions, run three checks before commissioning a new article:

1. Verify whether AI answers trigger for the query: Test a sample of your missing queries in Google and AI assistants. Google only displays AI Overviews when its systems determine an AI summary adds value beyond classic search results; if a query is purely navigational or local and triggers no AI answer, handle it with standard organic SEO.
2. Consolidate variations into 5 to 10 thematic clusters: Group long-tail queries that ask about the same underlying buyer problem, pricing objection, or integration requirement into a single cluster rather than treating each phrasing as a separate brief.
3. Refactor established pages first: Compare each prioritized cluster against your current website, blog posts, and product documentation. When an existing high-authority page already covers the parent topic, add the missing buyer sub-questions as H2 or H3 headings followed immediately by a direct two-to-three-sentence answer and your expert's supporting detail.

At Rocksalt, once a customer connects their CMS, we check each theme's buyer questions against what's already live on their website to flag where existing coverage is already sufficient versus where an update or new piece is needed. There's no typical ratio of updates to new pages, because it depends on how much a company has already published. A young company with little content will usually need new pages to answer its buyers' questions, while a mature company with a large library will close more of its gaps by updating what's already there.

Step 6: Turn each theme into one pillar page and a handful of interlinked subpages

When a prioritized theme requires new content, publish at the theme level with one pillar page supported by roughly six interlinked subpages, and refresh the cluster monthly as buyer questions persist or new ones appear.

Writing a standalone article for every granular prompt is too narrow to sustain and often reads like a thinly disguised reply to a single social thread. Operating one level higher, at the buyer-question and theme level, gives answer engines both topical breadth and passage-level depth:

• One pillar page per theme: The pillar page addresses the overarching problem or buying decision, establishes your point of view, and links out to each supporting subpage in the cluster (for instance, our pillar guide on how to build an AI visibility prompt set from real buyer questions introduces the four companion guides in this series).
• Roughly six interlinked subpages: Each subpage answers one distinct high-priority buyer question (or tight group of sub-questions) from the theme, using H1 and H2 headings that match the buyer's question and interlinking back to the pillar and sibling pages. In a privileged-access security program, for instance, a theme around just-in-time and break-glass access translates into one pillar page establishing the architecture and six interlinked subpages tackling the specific operational, compliance, and implementation questions beneath it.
• A monthly keep, refine, discard, and add refresh: Buyer questions aren't static. At Rocksalt, theme and question data refreshes monthly to show which questions persist across calls and communities and which new ones have emerged. Apply a keep, refine, discard, and add review to your content cluster just as you would to a prompt set: keep pages that consistently earn citations, refine sections where persisting questions still miss citations, discard or consolidate outdated angles, and add a new subpage only when a distinct buyer need surfaces.

Stop chasing synthetic sub-queries and build around buyer themes

When a page ranks on Google but disappears in ChatGPT, Perplexity, or AI Overviews, no single diagnosis explains every miss: crawler blocks, third-party mentions, and run-to-run model variation all play a role. On the page itself, however, a common gap you can directly control is answering a broad head term while skipping the specific implementation, comparison, and risk sub-questions that buyers and answer engines try to resolve.

Instead of asking an LLM to guess fan-out queries that change on every run, anchor your content plan in the follow-up questions real buyers ask across sales calls, support tickets, Reddit, and LinkedIn. Filter your candidate list by commercial relevance and evidence strength, update your existing pages before spinning up new URLs, and build each priority theme into a tightly interlinked pillar-and-subpage cluster backed by internal expertise that no search summary can replicate on its own.