Home › Vocabulary › Prompt Visibility Testing
Advanced Metric · AI Commerce Intelligence

Prompt Visibility
Testing

The most direct measurement of AI Commerce visibility. You send real buyer prompts to AI systems. You track whether your brand appears, at what position, and who shows up instead of you.

Get My Free AI Commerce Score Learn About Recommendation Share
What a test looks like

Real buyer prompts. Real results.

This is what a Prompt Visibility Test output looks like. Each row is a buyer prompt submitted to ChatGPT, Perplexity, and Google AI Mode. Your brand either appeared or it did not.

Prompt Visibility Test results · Skincare niche · 20 of 50 prompts shown
Recommendation Share: 18%
"best organic skincare for sensitive skin under $60"
1st pick
appeared
ChatGPT
"fragrance-free moisturizer for rosacea"
not found
missed
ChatGPT
"best clean beauty brands dermatologist approved"
2nd mention
appeared
Perplexity
"sustainable skincare brands similar to Aesop"
not found
missed
Perplexity
"vitamin C serum for hyperpigmentation under $50"
1st pick
appeared
Google AI
"recommend a complete skincare routine for beginners"
not found
missed
Google AI
"what skincare brand is good for oily acne-prone skin"
not found
missed
ChatGPT

A Recommendation Share of 18% means the brand appeared in 9 out of 50 prompts. The test also shows which prompt types it is missing from, which competitors appear in the gaps, and what the Recommendation Position was for each appearance.

Building your prompt set

The five prompt types that matter most

TYPE 01
Buyer Prompts
"Best protein powder for women runners"
Direct purchase intent. The highest-value prompt type. Appearing here drives real purchase decisions. Cover price ranges, use cases, and certifications.
TYPE 02
Comparison Prompts
"Best alternatives to Gymshark for women"
Category-level comparisons. Being included when buyers compare options builds Recommendation Density across the niche.
TYPE 03
Problem Prompts
"Skincare for acne-prone sensitive skin"
Tests intent matching. If you sell for this use case but do not appear here, you have an Intent Misalignment problem. The fix is specific content, not more content.
TYPE 04
Discovery Prompts
"Eco-friendly activewear brands like Patagonia"
Tests your position in the AI Recommendation Graph. Appearing in discovery means AI understands your brand characteristics and relationships.
The fifth type: AI Shopping Prompts. "Recommend me a complete skincare routine under $100." These simulate the exact moment a buyer hands buying decisions to AI. Appearing here is the most commercially direct form of Prompt Visibility.
Where to run tests

Three platforms, three different results

Your Recommendation Share will be different on each platform because they work completely differently. Running tests on only one platform gives you a partial picture.

🤖
ChatGPT
Uses training data. Results reflect what ChatGPT learned about your brand months ago. Slower to update but highest buyer intent.
Key driver: brand entity
🔍
Perplexity
Crawls in real time. Reflects your current page content and external citations. Fastest to respond to improvements.
Key driver: page content
🌎
Google AI Mode
Blends organic search signals with AI evaluation. Schema markup and structured data matter most here alongside traditional SEO.
Key driver: schema + SEO
Reading your results

What your test results tell you to fix

A
Low appearance rate across all prompt types
Your brand has low entity presence in AI training data. Priority: build external mentions through Reddit in your niche, get press coverage, add to Trustpilot. These changes take 4 to 12 weeks to filter into training data but are the foundation of everything else.
B
Appears on branded queries but not buyer prompts
You have brand entity but poor intent matching. This is Intent Misalignment. Rewrite content around actual buyer query language. Create FAQ sections that answer the prompts you are failing. Update llms.txt with buyer-intent-specific language.
C
Appears but always as a supporting mention, never first
You are in the Recommendation Graph but your Recommendation Position is weak. Fix Commerce Accuracy (prices in HTML, return policy visible) and strengthen trust signals. Getting from second mention to first pick can double AI-referred revenue.
D
Strong on ChatGPT, weak on Perplexity
Your brand entity is good but your current page content is weak. Perplexity crawls in real time. Create more specific, query-answering content and build external citations that Perplexity can find and cite in real time.
FAQ

Prompt Visibility Testing: common questions

What is Prompt Visibility Testing?▾
Prompt Visibility Testing is the process of submitting real buyer prompts to AI shopping systems and measuring whether a brand appears in the responses, at what position, and with what recommendation confidence. It is the most direct measurement of AI Commerce visibility because it tests actual recommendation behavior rather than inferring it from structural signals.
How many prompts do I need to test?▾
A minimum viable test is 50 prompts covering all five prompt types: buyer prompts, comparison prompts, problem prompts, discovery prompts, and AI shopping prompts. A comprehensive test uses 100 or more prompts. The more prompts you test, the more reliable your Recommendation Share calculation.
Which AI systems should I test with?▾
Test with at least ChatGPT, Perplexity, and Google AI Mode. These three operate differently. ChatGPT relies on training data. Perplexity crawls in real time. Google AI Mode blends organic search signals with AI evaluation. Testing all three gives a complete picture of your AI Commerce visibility.
How often should I run Prompt Visibility Tests?▾
Monthly is the right cadence for most stores. Running the same prompt set every month gives you Recommendation Velocity, the rate of change in your Recommendation Share over time. Positive velocity confirms your AI Commerce improvements are working. Negative velocity means a competitor is gaining ground.

Start with a free AI Commerce Score

The score predicts your Prompt Visibility results. The Paid Audit includes full Prompt Visibility Testing with 50 buyer prompts across ChatGPT, Perplexity, and Google AI Mode.

Get My Free AI Commerce Score Learn About Recommendation Share
Free. No credit card. Results in 30 seconds.
Related
📊
Recommendation ShareThe metric this test measures
 🎓
Learn guideFull Prompt Visibility Testing guide
 🤖
ChatGPT VisibilityHow ChatGPT evaluates stores
 👁️
Human View vs AI ViewWhat AI sees on your store
 📚
VocabularyEvery AI Commerce term defined
 📐
MethodologyThe full 8-factor scoring model
Home  ·  Prompt Visibility Testing  ·  Recommendation Share  ·  Methodology  ·  Privacy
© 2026 Atom Foundry · The AI Commerce Graph™ · atomfoundry.dev
Find → Prove → Fix → Protect
The AI Commerce Graph™