AI Commerce Metric · Atom Foundry
Recommendation
Share™
The percentage of high-intent buying prompts where your brand appears in AI recommendations. The AI-era equivalent of search market share. The single most commercially direct metric in AI Commerce Intelligence.
Average Recommendation Share across scanned stores
18
% of buyer prompts
Top brands in each niche exceed 40%. Most brands hover below 15%.
Definition
What is Recommendation Share?
Recommendation Share™
Recommendation Share is the percentage of high-intent buying prompts relevant to a brand's niche in which that brand appears in AI recommendations. If your brand appears in 18 out of 100 tested buyer prompts, your Recommendation Share is 18%.
Where SEO tracked keyword rankings, AI Commerce Intelligence tracks Recommendation Share. The shift from position to presence, from ranking to recommendation, changes what you optimize for, how you measure success, and what gaps cost you revenue.
Recommendation Share is not a vanity metric. Every point of Recommendation Share represents a percentage of AI-referred purchase decisions in your category. AI-referred orders convert at 50% higher rates with 14% higher average order value. Every point of Recommendation Share you gain is directly tied to revenue.
The shift
From keyword rankings to Recommendation Share
These are not the same metric in a new format. They measure fundamentally different things and require completely different strategies to improve.
SEO Era (pre-2025)
What you tracked
Keyword position #3
How visibility worked
10 results per page
Who decided
Human clicked
What you optimized
Keywords and backlinks
→
AI Commerce Era (2025+)
What you track
Recommendation Share %
How visibility works
2 to 3 recommendations
Who decides
AI recommends
What you optimize
Trust, clarity, intent match
Winner takes most. AI returns 2 to 3 recommendations, not 10 pages of results. The stakes of appearing vs not appearing are dramatically higher in AI commerce than in search. Recommendation Share compression means the difference between appearing and not appearing is often the difference between getting the sale and not.
The commercial reality
Why Recommendation Share matters more than ranking
13x
Growth in AI-referred orders year over year (Shopify Q1 2026)
50%
Higher conversion rate from AI-referred traffic vs organic
14%
Higher average order value from AI-referred purchases
Every buyer prompt that AI answers without mentioning your brand is a potential sale going to a competitor. Recommendation Share quantifies exactly how much of your category's AI traffic you are capturing versus missing. It is the most commercially direct metric in AI Commerce Intelligence.
Quality matters too
Recommendation Share and Recommendation Position
Appearing in AI responses is not enough. Where you appear, and how confidently AI recommends you, determines the commercial outcome. Recommendation Share measures quantity. Recommendation Position measures quality.
Recommendation Position and estimated conversion impact
First recommendation
High conversion
Supporting mention
Medium conversion
Comparison mention
Lower conversion
Weak citation
Minimal conversion
A brand with 60% Recommendation Share but always as a weak citation converts very differently from a brand with 35% Recommendation Share always as the first recommendation. Both Recommendation Share and Recommendation Position matter for commercial outcomes.
Industry benchmarks
Recommendation Share by niche
Recommendation Share concentration varies significantly by niche. Some categories are dominated by one or two brands. Others are more distributed. Understanding your niche benchmark sets the right target.
💕 Beauty
48/100 avg AI score
📸 Fashion
42/100 avg AI score
💪 Health
51/100 avg AI score
🏛 Home
44/100 avg AI score
🏉 Sports
47/100 avg AI score
🐾 Pets
49/100 avg AI score
🍎 Food
46/100 avg AI score
Benchmarks based on Prompt Visibility Testing across top-scoring stores per niche in the Atom Foundry database. May 2026.
How to measure it
Running your first Recommendation Share test
Recommendation Share is measured through Prompt Visibility Testing. The process is straightforward. The discipline is in building a representative prompt set and running it consistently over time.
Type 1
Buyer Prompts
"Best protein powder for women runners"
Direct purchase intent. The highest-value prompt type. Appearing here drives direct sales.
Type 2
Comparison Prompts
"Best alternatives to Gymshark"
Category-level comparisons. Appearing builds Recommendation Density across the niche.
Type 3
Problem Prompts
"Skincare for acne-prone sensitive skin"
Tests intent matching. If you sell for this use case but do not appear, you have Intent Misalignment.
Type 4
Discovery Prompts
"Eco-friendly activewear brands like Patagonia"
Tests Recommendation Graph position. Appearing in discovery builds long-term category visibility.
1
Build 50 to 100 prompts for your niche
Cover all four prompt types above. Use the actual language buyers use in your category. Include price qualifiers, use-case specifics, and comparison terms relevant to your competitors.
2
Submit to ChatGPT, Perplexity, and Google AI Mode
Run each prompt through all three systems separately. Record which prompts your brand appears in, at what position, and who appears in the ones where you do not appear.
3
Calculate Recommendation Share per platform
Appearances divided by total prompts tested. Do this separately for each platform. Your ChatGPT Recommendation Share and Perplexity Recommendation Share will differ significantly because the two systems work differently.
4
Track Recommendation Velocity monthly
Run the same prompt set every month. The change over time is your Recommendation Velocity. Positive velocity confirms your AI Commerce improvements are working. Negative velocity means competitors are gaining ground faster than you are.
How to improve it
What moves Recommendation Share up
Recommendation Share is the output. The inputs are your AI Commerce Score factors, your external authority presence, and how well your content matches actual buyer prompt language. These are the highest-leverage levers.
A
Fix your AI Commerce Score first
A low AI Commerce Score predicts low Recommendation Share. Fix Commerce Accuracy (prices in HTML), AI Trust Confidence (return policy in crawlable text), and Semantic Clarity (specific positioning) before anything else. These are the structural foundations that enable visibility.
B
Build your AI Trust Graph
External authority is the most under-leveraged driver of Recommendation Share. Reddit presence in your niche, press coverage, Trustpilot or Google reviews, LinkedIn mentions. These signals feed into ChatGPT training data and Perplexity real-time citation. Each external mention is a vote for your brand in the AI Recommendation Graph.
C
Align content with buyer prompt language
Run your current content against your prompt test set. Where your content language does not match buyer prompt language, rewrite it. This is Intent Misalignment and it is fixable. Create FAQ sections that answer the exact prompts you are failing. Update llms.txt with buyer-intent-specific language and use-case descriptions.
D
Increase Recommendation Density
Recommendation Share measures presence in your primary category. Recommendation Density measures how far that presence extends across adjacent buyer intents. Create content that answers related problems, comparison queries, and adjacent use cases. Broader coverage means more buyer prompts where you appear.
FAQ
Recommendation Share: common questions
What is Recommendation Share?▾
Recommendation Share is the percentage of high-intent buying prompts relevant to a brand's niche in which that brand appears in AI recommendations. If your brand appears in 18 out of 100 tested buyer prompts, your Recommendation Share is 18%. It is the AI-era equivalent of search market share.
How is Recommendation Share different from keyword rankings?▾
Keyword rankings measure where you appear in Google search results for specific queries. Recommendation Share measures whether you appear in AI recommendations when buyers ask conversational shopping questions. AI returns 2 to 3 recommendations, not 10 pages of results. The stakes are higher and the metric is more directly tied to revenue.
How do I measure my Recommendation Share?▾
Build a set of 50 to 100 buyer prompts for your niche covering purchase-intent queries, comparison queries, discovery queries, and problem-solving queries. Submit each to ChatGPT, Perplexity, and Google AI Mode. Record which prompts your brand appears in. Divide appearances by total prompts to get your Recommendation Share percentage.
What is a good Recommendation Share score?▾
In most niches, the top two or three brands capture 60 to 80% of AI Recommendation Share combined. A new brand might start at 5 to 15%. A well-optimized mid-size DTC brand typically reaches 20 to 35% with focused effort. Category leaders in some niches exceed 50%.
Is Recommendation Share the same as Recommendation Position?▾
No. Recommendation Share measures how often your brand appears across buyer prompts. Recommendation Position measures where in the response you appear when you do appear. A brand with high Recommendation Share but always as a weak citation converts differently from a brand with lower share but consistent first-recommendation positioning. Both metrics matter.
What is Recommendation Velocity?▾
Recommendation Velocity is the rate of change in your Recommendation Share over time. If your Recommendation Share was 14% in March and 19% in May, your Recommendation Velocity is plus 2.5 percentage points per month. Positive velocity confirms your AI Commerce improvements are working. Tracking velocity over time is more valuable than any single measurement.
Start measuring your Recommendation Share
Get your free AI Commerce Score first. The score predicts your Recommendation Share. Then the Paid Audit includes full Prompt Visibility Testing with 50 buyer prompts across ChatGPT, Perplexity, and Google AI Mode.
Free. No credit card. Results in 30 seconds.