AI Search Visibility
Audit Competitors' AI Search Visibility in 15 Min
A 15-minute competitor AI search audit shows why they appear in AI answers and you do not. Here's the step-by-step Noma walkthrough for D2C category queries.

Your competitor is being recommended by Perplexity for your product category. ChatGPT names them when someone asks about your product type. Gemini includes them and not you in a category roundup. You want to know why. This audit tells you, in 15 minutes, with enough specificity to build an action plan from the results.

Minutes 0 to 3
Define Your Competitor Set and Query Set
Pick your top two competitors, the ones you see cited most frequently when you manually test AI engine queries. Pick 10 queries that represent your core category:
3 direct product queries
3 problem queries
2 comparison queries naming both brands
2 brand-specific queries about each competitor
In Noma, add your competitors' domains to your competitor tracking. Noma asks your query set daily across ChatGPT, Gemini, Perplexity and Claude on every plan (the rest of its 13 tracked engines are add-ons) and tracks citation by brand automatically. If running manually, open four windows, Perplexity, ChatGPT, Gemini, Google AI Overview, and run each query.
Minutes 3 to 8
Run the Queries and Record Results
For each query on each engine, record: is your competitor cited (Y/N), at what position (first mentioned / subsequent), with what language (positive / neutral / qualified), and which source does the AI cite in footnotes for this brand.
The source column is the most important data point in the audit. In Perplexity especially, footnote citations show you exactly which piece of content the AI is drawing from when it mentions your competitor. This is your gap map, the specific URLs that are driving their AI visibility.
Minutes 8 to 12
Categorise the Sources
Open each source URL from your competitor's AI citations. Categorise each:
Source Type | If Competitor Has This and You Don't | Your Gap |
|---|---|---|
Editorial mentions, DA 50+ publications | 6 of 10 competitor citations | Earned media gap, need similar coverage |
Structured product page content | 7 of 10 competitor citations | On-page GEO gap, content and schema |
Reddit / Quora community discussions | 5 of 10 competitor citations | Community presence gap |
Amazon / marketplace reviews | 4 of 10 competitor citations | Review volume and quality gap |
YouTube review videos | 3 of 10 competitor citations | Creator content gap |

Minutes 12 to 15
Build the Three-Action Gap Closure Plan
Action 1 (this week): the gap you can address fastest. If it is product page structure, rewrite your top three product pages today. If it is robots.txt blocking AI crawlers, fix that in 20 minutes.
Action 2 (this month): the earned media gap. Identify the specific publications your competitor has coverage in that you do not. Draft pitches for the top two.
Action 3 (this quarter): the community and review gap. If your competitor has 200 Amazon reviews with specific outcome language and you have 18, the review-building programme starts now.
Running the Audit Without Noma
The Manual Version, Still Worth Doing
If you are running the audit manually without Noma, the source identification step is the hardest part. Perplexity shows footnote citations clearly, click through every footnote when your competitor is cited and record the URL. ChatGPT with browsing sometimes shows sources, sometimes does not. Gemini surfaces source links inconsistently. Google AI Overview shows the sources in the expandable panel below the overview.
For the queries where no source is visible, use the source URL you already found from other queries as a starting hypothesis, AI engines tend to draw from consistent sources for the same brand across related queries. If Perplexity cites one industry-publication article about your competitor in 6 of 10 queries, that article is almost certainly your competitor's primary citation driver across all four engines.
One additional manual step worth taking: after identifying your competitor's top citation sources, run those URLs through a backlink checker (Ahrefs or Semrush, both offer limited free lookups). Check whether the article links back to your competitor's website. Backlinks from high-DA publications are a combined SEO and GEO signal, they improve both Google ranking and AI engine confidence in the cited source. If your competitor has 3 backlinks from DA-65 publications and you have zero, that is a measurable, addressable gap.
Building the Gap Closure Plan
Prioritising Which Gaps to Close First
After running the 15-minute audit, you will typically find 3 to 5 gaps between your AI search presence and your competitor's. Not all of them are equal. Here is how to sequence the closure plan so you see citation improvement in the shortest time.
Fix technical gaps first, always. If your competitor has editorial coverage and you do not, that is a 6 to 8 week gap closure. If your robots.txt is blocking a crawler, that is a 10-minute fix with a 2-week citation improvement. Technical issues have the fastest return on time invested. Fix them before doing anything else.
Prioritise Perplexity gaps over Gemini gaps. Perplexity is built around research-style questions answered with cited sources, so a gap there directly impacts buyers who are deliberately using AI for considered purchase decisions. Gemini gaps are worth closing but the Perplexity gap is more commercially urgent.
Address review gaps with a 90-day timeline, not urgency. If your competitor has 180 Amazon reviews with outcome-specific language and you have 20, this is a real gap but it cannot be closed in a week. Set up your post-purchase review request sequence this week, run it consistently for 90 days, and revisit the gap at that point. Trying to accelerate review acquisition through non-genuine means risks the opposite of the intended effect.
Track the closure. Run the same 10-query competitor comparison monthly. The gap should narrow over 3 to 4 months if your actions are correct. If the gap is not closing, the action targeting the wrong gap, not the effort being insufficient.
"The gap between you and your competitor in AI search is not a mystery. It is a list of specific URLs they have and you do not. Find the list. Build the URLs."

FAQ
Frequently Asked Questions
How do I audit my competitor's AI search visibility?
Run 10 category queries across Perplexity, ChatGPT, and Gemini. Record every citation of your competitor including position and exact language. In Perplexity specifically, expand footnotes to see which URLs the AI is citing for your competitor. Those URLs are the content gap map: the specific pages your competitor has that you do not.
Why does my competitor appear in Perplexity but not me?
In most cases, your competitor has one or more of: editorial coverage from a DA-50+ publication that links to their product pages, structured FAQ content on their product pages that matches the query intent, or authentic Reddit and Quora mentions from real users. The Perplexity footnotes on your competitor's citations will show you exactly which content type is driving their visibility.
How long does it take to close a competitor AI search gap?
Technical gaps like robots.txt fixes close within 2 weeks. Content gaps like product page restructuring show citation improvement in 4 to 8 weeks. Earned media gaps take 6 to 12 weeks from pitch to publication to AI indexing. Review volume gaps take 90 days minimum with a consistent post-purchase email sequence. Track monthly using the same 10-query set.
How do I track competitor AI visibility over time?
Run the same 10-query competitor comparison monthly. Record competitor citation rate, citation position, and sentiment for each engine. Noma automates this tracking and records the page behind each citation, so when a competitor's share of voice jumps you can see which piece of content drove it and identify your gap rather than guessing at the cause.
Sources and References
Promptwatch – Competitive GEO analysis: methodology and tool guide
Perplexity AI – Source attribution in AI-generated category responses
Search Engine Land – GEO competitive analysis: a practical framework
Nurdd / Noma – Competitor AI visibility audit: D2C tool walkthrough