Question Diagnosis
Not all questions are equal. Question diagnosis breaks down every AI response — per question, per model — so you see exactly where your brand appears, where it doesn't, and what AI says when it mentions you.
Your overall AI visibility score is an average. Question diagnosis goes beneath that average to show you the individual responses. For each of the 30–100 strategic questions, you see how ChatGPT, Claude, and Perplexity responded — whether your brand was mentioned, in what position, with what sentiment, alongside which competitors.
Every question tells a different story
You might score 70 overall but have 10 questions where you're completely invisible. Those 10 questions might be the highest-intent purchase queries in your industry. Without per-question diagnosis, you'd never know.
Question diagnosis reveals the distribution: which questions you dominate, which you're absent from, and which show mixed results across models.
This granularity transforms your optimization strategy from "improve overall visibility" to "fix these 5 specific questions where we lose to Competitor X on ChatGPT."
Aggregate Scores vs Per-Question Intelligence
A single number hides critical gaps
You see "Score: 65" and assume moderate visibility. But that 65 hides: 20 questions at 90+ (strong) and 10 questions at 10 or below (invisible). The invisible questions are exactly where your competitors are winning customers.
Surgical precision per query
Every question has its own scorecard: mention presence, position, accuracy, sentiment, competitor co-mentions. You know exactly which queries to prioritize and what changes will improve each one.
Cross-model comparison per question
ChatGPT might recommend you for "best CRM" but Claude might not. Per-question, cross-model analysis reveals these discrepancies and helps you understand why each model responds differently.
Some questions show consistent results across all models — those are your strengths. Others show wild variance — those are your opportunities.
Use cross-model variance as your strategy guide. High variance questions (strong on one model, weak on another) respond to model-specific optimizations. Low variance strong questions are your anchors — maintain them. Low variance weak questions are your biggest opportunity — a single well-targeted content piece can lift all models simultaneously.
Turn diagnoses into a continuous optimization loop
Question diagnosis is not a one-time report — it's a living intelligence feed that evolves with every rescan.
AI models update their training data on different schedules. A question you dominated 6 months ago may now show weaker results — a competitor published better content, or a news event shifted how AI frames your industry.
Question diagnosis gives you a historical view: your score on this question over your last 5 scans. Declining questions trigger targeted content updates. Improving questions validate your optimization strategy.
The diagnosis grid also highlights questions where your score dropped despite no action on your part — early signals of competitor activity or AI model updates that require response.
Pair question diagnosis with the priority filter to create a maintenance schedule: rescan Critical questions monthly, High questions quarterly, and Low questions every 6 months. This systematic approach keeps your AI visibility stable as the AI landscape evolves.
Every question diagnosis links directly to the Generate Content action — creating targeted content for that specific question, based on exactly what the AI responses revealed is missing from your brand coverage.
How Question Diagnosis Works
Every AI response is captured, analyzed, and presented in a structured diagnostic view — making complex data actionable.
1. Response capture
Each question generates responses from all AI models in both web search and native modes. Every response is stored in full.
2. Multi-dimension analysis
Each response is scored on mention presence, position (first vs last), accuracy, sentiment (positive/neutral/negative), and competitor citations.
3. Diagnostic dashboard
You see every question on a filterable grid: sort by score, model, category, or competitor presence. Drill into any question for the full response and analysis.
Every response analyzed across ChatGPT, Claude, and Perplexity — web search and native modes — for every strategic question.
Diagnose Every AI Response About Your Brand
Stop treating your AI visibility as a single number. See exactly what each model says for every question. Free scan.
Diagnose Questions →