See exactly which prompts your competitor wins that you don't.

Your competitor is cited. You are not. AI engines have found a reason to prefer them for this prompt and you have no presence in the answer at all.

→ Generate a brief targeting this prompt directly.

Both you and your competitor are cited in the same AI response. You appear, but so do they. You share the citation rather than owning it.

→ Strengthen the signals that make you the primary recommendation.

You are cited. Your competitor is not. You own this prompt. These are the citation slots you need to defend.

→ No immediate action — monitor for competitor movement.

What does 'prompt-level' competitive intelligence actually mean?

Most AEO tools show you share of voice — a single percentage comparing how often your brand appears vs. a competitor across all tracked prompts. Prompt-level means you see the individual query: 'best project management software for agencies' — they appear, you don't. That specificity is what makes the gap actionable. A share of voice number tells you you're losing. A prompt-level gap table tells you exactly where and why.

How is this different from the SOV comparison already in the AI Visibility Tracker?

The AI Visibility Tracker compares share of voice at an aggregate level — your citation rate vs. their citation rate across the full prompt set. Competitor Gap Intelligence goes prompt by prompt: for each specific query, who appears and who doesn't, with the actual AI-generated snippet shown. The Tracker tells you you're behind. This product tells you on which exact prompts.

Which engines does the scan cover?

ChatGPT (OpenAI), Perplexity, Gemini, and Google AI Mode. The scan runs all four in parallel. Results show which engines cited the competitor and which cited you, per prompt — so you can see if a gap is universal or engine-specific.

How long does the first scan take?

The first scan for a new competitor takes 60–90 seconds — prompts are batched and run in parallel across all four engines. Subsequent loads return instantly from a 7-day cache. When you add a new competitor to your tracking list, the first gap scan runs automatically.

What does the brief generation include?

A brief generated from a gap row includes: the target prompt (the exact query the competitor wins), the competitor's cited snippet (what the AI said about them), a recommended headline and article structure, an answer-first paragraph guide (direct answer in the first 40–60 words), an FAQ section expanded from the prompt, an E-E-A-T checklist, and a schema type recommendation. It also uses your Brand Context profile — so the brief reflects your actual positioning against that specific competitor.

Will this show me gaps on prompts I'm not currently tracking?

The gap scan runs against your monitored prompt set — prompts you've added to your AI Visibility Tracker. It identifies which of those prompts the competitor wins. It does not discover entirely new prompts outside your tracked set. To discover new category-level prompts, use the Topic Explorer in the Page Creation Engine.

What is the Source Gaps tab and how does it work?

The Source Gaps tab (inside the Competitor Gaps panel) surfaces the specific third-party domains AI engines cite when your competitor wins a prompt. It derives these domains directly from the Phase 1 scan cache — no additional AI calls needed. Domains are classified by type (editorial, UGC, reference, competitor's own site) and paired with a suggested action. Upload an Ahrefs referring domain export in Profile → Competitors to add DR scores and sort by authority. If AI citation data is thin for a given competitor, the view falls back to your uploaded dataset filtered to domains where that competitor has a referring link.

What is the Narrative Gaps tab and how is it different from Prompt Gaps?

Prompt Gaps show you which prompts a competitor wins that you don't. Narrative Gaps go deeper — they extract what the AI actually says about your competitor versus what it says about you, for the prompts where both brands appear. For each co-mentioned prompt, the tab shows the specific attribute gap: your competitor is described as 'the enterprise-grade choice' while you're described as 'suitable for small teams.' Five narrative gap types are identified: positioning, authority, specificity, feature, and social proof. Each card includes a suggested action for closing the framing gap.

Can I track multiple competitors at once?

Yes — the panel supports up to 4 competitor tabs simultaneously. Click 'All Competitors' (the last tab) to open the Competitive Exposure Map: for each prompt in your tracked set, it calculates an exposure score — how many competitors appear in that AI response when you don't. Prompts are ranked by exposure level (Critical, High, Moderate) so you can see at a glance where you're most exposed across your full competitive set. Use 'Scan all' to run all competitors in sequence with a live progress counter. The Prompt Gaps, Source Gaps, and Narrative Gaps tabs let you drill into a single competitor for full per-prompt analysis.

What are the Comparison, Use-case, and Broad prompt types?

Every scan generates prompts across three tiers: Broad (category-level queries like 'best tools for X'), Comparison (queries that name your competitor directly — 'HubSpot vs alternatives', 'best HubSpot replacement for agencies'), and Use-case (persona or location-specific queries derived from your uploaded Ahrefs keyword data via AI — 'email marketing software for SaaS startups', 'CRM for remote sales teams'). The Comparison tier produces the highest citation signal because the competitor is named in the query — AI engines have a strong reason to mention them, and these prompts show you the exact framing they win on. Use the Prompt type filter in the gap table to focus on any single tier.