October 23, 2026
A Complete Guide to the AEOTrack Questions Page: Track Every Question That Matters to Your Brand
Introduction
Your overall AI Visibility Score tells you how you're doing. The Questions page tells you why. This is where AEOTrack breaks visibility down to its most granular, actionable level: the individual questions real customers might ask an AI assistant, and whether your brand shows up in the answer.
This guide walks through every card, tab, table column, and chart on the AEOTrack Questions page, using a fictional example company, Acme Coffee Roasters (acmecoffee.example), so you can understand exactly what you're looking at before you ever open your own dashboard. Acme tracks coffee questions against fictional rivals Northwind Roasters, Bluebird Coffee, Harbor Beans and Ridgeline Coffee.
What Is the AEOTrack Questions Page?
The Questions page (found under Questions in the sidebar) is where you manage and review the specific questions AEOTrack checks on your behalf. Each one is a real, natural-language question a prospective customer might type into an AI assistant, and AEOTrack repeatedly asks AI models that exact question to see whether your brand gets named, cited, and recommended.
Where the main dashboard gives you a bird's-eye view of your AI visibility, the Questions page answers more specific questions:
- Which exact questions is my brand winning or losing?
- Is my position on a given question improving or getting worse?
- Which AI engines were checked, and did any of them have trouble answering?
- Is my website actually being cited as a source?
- Which competitors are named instead of me on a given question?
- How is my overall pool of tracked questions shifting over time — toward visibility, or away from it?
The Questions Tab: Summary Metrics
At the top of the Questions tab, four summary cards give you a snapshot of your question-tracking performance.
Questions Score
What it means: How AI answers the pool of questions tracked on this page, on two cards. Named in AI answers counts the latest answers that name you, out of the answers that could be read, with the sample line under it: the ± band, how many answers and engines it covers, and the date of the check. Visibility score is the 0 to 100 score from the same answers. Before the first answers arrive, the score card reads AI visibility and Answers analysed stands in for the count.
Example: Acme Coffee Roasters is named in 32 of 60 answers, and its Visibility score is 38/100. The sample line reads "±19 · 45 answers in the score · 5 engines · checked 29 Sep 2026": the score leaves out its two questions with its name in them and one advice-only question, which the count above it includes.
Why it matters: This is the question-level equivalent of your dashboard's overall AI Visibility Score — it tells you how you're doing across the exact set of questions you've chosen to monitor.
What you should do: Treat this as your working number for question-level strategy. If it's low, the table below will show you exactly which questions are dragging it down.
Branded
What it means: Your score on the questions that contain your brand name (branded and near-brand questions), scored apart. Since scoring version 9 they are left out of the Visibility score, so this card shows whether AI knows you when it is asked about you by name, not whether it recommends you to buyers who have not heard of you. Its hint says how many answers it is from. It shows no number while none of your questions contains your name. When every question does, the Visibility score counts them too, and the line under it says so. See Branded and near-brand questions.
Example: Acme Coffee Roasters scores 60/100 on the card labelled Branded, from its two questions with its name in them; its Visibility score, from its other questions, is 38/100.
Why it matters: It keeps reputation apart from discovery: an engine asked about you by name names you, whatever your visibility to buyers who do not know you yet.
What you should do: If Branded is high and the Visibility score low, the engines know you but do not yet recommend you. Work on the questions buyers ask before they know your name.
Questions Tracked
What it means: The number of questions currently being actively monitored, out of your plan's total allowance.
Example: Acme Coffee Roasters tracks 12 out of 75 available question slots.
Why it matters: This tells you how much of your available tracking capacity you're using — and, indirectly, how broad or narrow your current visibility picture is.
What you should do: If you're tracking far fewer questions than your plan allows, consider adding more questions (see "Track a New Question" below) to get a fuller picture of your visibility across your category.
Answers Analysed
What it means: The total number of individual AI answers AEOTrack has collected and analyzed across all your tracked questions and engines.
Example: Acme Coffee Roasters shows 45 answers analysed.
Why it matters: More analyzed answers generally means a more statistically reliable picture of your visibility, since it smooths out day-to-day variation in how AI models respond.
Engines Behind This Score
Just below the summary cards, AEOTrack shows which AI engines contributed to the current score, along with any notes about how the check ran.
What it is: A row of engine icons (for example, Perplexity, ChatGPT, Gemini, Claude, and Grok) representing every AI platform queried for the current set of results. An engine that is not configured on the platform is shown as not measured: it is absent from the score, never counted as a 0. An answer that was cut off before it named anyone is not counted either.
Why it matters: Not every engine responds cleanly every time. AEOTrack transparently flags when something unusual happened during a check.
Example (fictional): A note might read: "Grok needed retries on this run but answered in the end, so your score still covers the full panel." This tells you the score is still complete and trustworthy, even though one engine had a rough run.
When an engine's newest attempt failed outright, its last good answer stands in, shown with its own date and a note that the newest attempt failed. A failed answer is never counted as "not named". See Engine chips.
What you should do: If an engine is consistently missing or failing to answer (rather than just needing a retry), that's worth investigating separately — it may mean that engine can't currently be evaluated for your brand at all.
The Questions Table
The core of the Questions page is a detailed table listing every tracked question, with a column for each key data point.
Question
What it is: The actual tracked question, word for word as it is asked of each AI engine. Example: "best coffee subscription for home brewing". Some questions are labeled BRANDED (the question names you) or NEAR-BRAND (it contains your name once the spaces are removed); both are scored apart from the headline rather than in it. During onboarding after a free audit, the audit's question is pinned to the top of the table and labelled From your free audit.
Score
What it is: A per-question visibility score. How to interpret it: A score of 0 means AI is not currently naming your brand at all on this specific question — this is your most direct, granular signal of where a content gap exists. A row that reads Advice-only · not in your score is different: no engine named any business for that question, so it is counted but not scored. Example: Acme Coffee Roasters scores 83 on "best coffee subscription for home brewing", while "best decaf coffee beans" reads Not yet named, with Harbor Beans named instead, and "how to store coffee beans to keep them fresh" is advice-only. The advice-only row carries a Replace action so you can swap it for a question buyers ask by name; the new question's first check is free.
Mentions
What it is: A fraction showing how many engines named you in the latest check, out of the engines that answered (e.g., "0/5" means none of the five engines named you; each engine gives one answer per check). How to interpret it: it is the per-engine picture behind the score, and it can move between checks because AI answers vary from one run to the next.
Trend
What it is: A small sparkline showing the recent trajectory of this question's score. How to interpret it: A downward red line signals declining performance on that specific question; a flat gray line signals no meaningful change; an upward line signals improvement. What to do: Prioritize investigating questions with a declining trend before they fully drop out of contention.
Sentiment
What it is: Where available, an indicator of the overall tone of the AI's answer when your brand is mentioned (positive, neutral, or negative). Why it matters: Being named isn't automatically a good outcome — sentiment tells you whether the mention is likely to help or hurt.
Engines
What it is: Icons showing which specific AI platforms were checked for this question, with visual emphasis on which ones actually answered.
Citation
What it is: A count showing how many total citations (source references) were logged for this question, often with a link icon. Example: Acme Coffee Roasters' question "best coffee subscription for home brewing" shows a citation count of 216, meaning that, across all checks and engines, your website (or others) were cited a combined 216 times in answering this question. Why it matters: A high citation count with a low mention score can indicate the topic is well-covered by sources in general — but not yet by you specifically.
Competitors
What it is: A small row of colored initials representing which competitor brands were named in answers to this question. What to do: Click into a question to see exactly which competitors are winning it, and review their relevant content for gaps you can fill.
Recheck
What it is: A manual refresh control letting you re-run the check for a specific question immediately, rather than waiting for the next scheduled check — useful right after you've made a content change.
The Questions table for Acme Coffee Roasters: a score, mentions out of 5, trend, sentiment, engines, citations and competitors for each question, with "Not yet named" and who was named instead on the two questions no engine named Acme for.
The Other Tabs: Topics, SERP monitor, Responses, Add questions
Alongside the Questions tab, the Questions page includes four additional tabs:
- Topics — groups your tracked questions into broader subject clusters, useful for spotting which categories of questions you're strong or weak in overall, rather than reviewing one question at a time.
- SERP monitor — where your website actually ranks in classic Google search results for a keyword, so you can compare organic rankings with the AI answers on the Questions tab (Pro and up).
- Responses — lets you review the raw AI-generated answers behind each score, so you can read exactly what an AI model said in response to a given question.
- Add questions — where you submit new questions for AEOTrack to begin tracking.
"How Your Questions Have Moved": The Question Status Chart
Below the table, AEOTrack shows a stacked bar chart tracking how your entire pool of tracked questions is distributed across four possible states, at each check-in date.
The four question states
- Stronghold (green): AI names your brand and cites your website. Your strongest possible position on that question.
- Named, not cited (purple): AI names your brand but doesn't reference your website as a source.
- Read, not credited (orange): AI reads your website's content as a source but doesn't actually name your brand in its answer.
- Invisible (gray): Neither named nor cited — you have no presence at all on that question.
How to read the chart
X-axis: Check-in dates.
Y-axis: Number of tracked questions.
Each bar: A snapshot of your entire tracked question pool on that date, segmented by state.
The AEOTrack question status chart, showing stronghold, named, read, and invisible questions over time.
Example (fictional): Acme Coffee Roasters' chart covers 12 checks from 19 Aug to 28 Sep 2026, with 13 questions in every bar. On 19 Aug it had no strongholds, 3 questions named but not cited, and 10 invisible. By 28 Sep it had 6 strongholds, 2 named but not cited, and 5 invisible. A single orange "read, not credited" question appeared on 2 Sep and was gone by the next check.
The opposite pattern is the warning sign: if Acme had added more questions and the gray segment had grown with them while the green stayed flat, the new questions would be outpacing its visibility.
What this tells you
A growing invisible segment, especially if it's growing faster than your stronghold segment, suggests that new tracked questions are outpacing your ability to earn visibility on them. A shrinking invisible segment alongside a growing stronghold segment is the clearest possible sign of real progress.
What action to take
Use this chart as your single best "big picture" view of whether your overall effort is working. Then use the questions table above it to identify exactly which questions are stuck in the invisible or read-not-credited states, and prioritize those: open Sources for each and work through the pages the engines read for it (Get named: work the off-site plan).
Track a New Question
Beneath the chart, AEOTrack provides a direct way to expand your tracked question set.
- "Add questions" lets you manually submit new questions you want monitored — questions you know your customers are likely asking an AI assistant.
- "Suggest questions" uses AI to brainstorm relevant questions worth tracking for your specific site and category, which is useful if you're not sure where to start or want to make sure you're not missing an obvious gap. Suggestions are built from the business you confirmed and only ask for names: how-to and informational questions, whose answers name nobody, are left out. A service business gets questions that ask for providers, never for software.
- "Manage everything" links out to a fuller question-management view for reviewing, editing, or archiving your full tracked question list.
What action to take: If your Questions Tracked count is far below your plan's limit, use "Suggest questions" to quickly expand coverage with relevant, high-value questions rather than guessing.
How to Read the Questions Page: Step-by-Step
- Check your Questions score and the Branded figure for the current snapshot.
- Review the questions table, sorting or scanning for the lowest scores and declining trends.
- Open a specific question to see engine-by-engine responses, citations, and named competitors. The drawer has an Open opportunity report button that takes you to the full-page opportunity report for that question, or generates one if none exists yet.
- Check the question status chart to see your overall stronghold vs. invisible balance over time.
- Identify gaps — prioritize "read, not credited" and declining-trend questions, since discovery is already happening there.
- Add or suggest new questions if your tracked pool doesn't yet reflect the full range of things customers ask.
- Use "Recheck" after making a content change to see the new answers straight away, rather than waiting for the next scheduled check.
Frequently Asked Questions
What is a "NEAR-BRAND" question?
A tracked question that does not name you but contains your name once the spaces are removed ("best AEO tracking software" contains "aeotrack"). Like a branded question, it is scored apart from the headline, not in it; see Branded and near-brand questions.
AEOTrack says my questions may be out of date. What does that mean?
The questions AEOTrack suggested were written for the business as it understood it. After you change your brand name or aliases, or confirm a different business, it marks them and can propose a new set, and shows how many suggested questions it would replace before anything changes. Only suggested questions are replaced, never ones you added after setup, and replaced questions keep their history. On a website set up before late October 2026, a question typed during setup can count as suggested. See Confirm your business.
What does "This question doesn't fit" mean?
The question has been checked at least twice without naming you, and its answers can never name a business like yours: they name no business at all, they answer a how-to or informational question with steps, or they list software when you sell a service. It shows under Check these questions at the top of this page and in Check your setup on the Dashboard, and offers Replace with "..." when a buyer-style version is ready, Suggest buyer questions, Stop tracking this question and Keep this question. A replaced question keeps its history. See A question that doesn't fit.
What does it mean if my score is 0 on a question?
No engine named your brand in the latest check for that question; the row says who was named instead. If no engine named any business at all, the question is advice-only: the row reads Advice-only · not in your score, it is left out of the headline score rather than scored 0, and Replace swaps it for a better question. Either way a 0 is the usual first result, and the fix is to get onto the pages the engines read for that question; see Get named: work the off-site plan. See Why AI does not name you yet.
An advice-only row is left out of the score instead of counting as 0.
What's the difference between "Named, not cited" and "Read, not credited"?
"Named, not cited" means AI mentions your brand but doesn't reference your site. "Read, not credited" is the reverse — your site was used as a source, but your brand wasn't actually named in the final answer.
Why does an engine need a "retry"?
Some AI platforms occasionally fail to respond on the first attempt. AEOTrack automatically retries so your score still reflects the full panel of engines rather than a partial result.
How do I add new questions to track?
Use "Add questions" to submit your own, or "Suggest questions" to have AI recommend relevant questions for your site.
What should I do with a declining-trend question?
Investigate what changed — a competitor may have published new content, or an AI model's underlying answer may have simply shifted. Consider updating or expanding your own relevant page.
Conclusion
If the main AEOTrack dashboard is your bird's-eye view, the Questions page is where the real work happens. It shows you, question by question, exactly where you stand — and which questions name a competitor and not you. Make it a habit to review your lowest-scoring and declining-trend questions regularly, expand your tracked pool as your category evolves, and use the "Recheck" option to see whether the answers changed after your content updates.