Measurement
AI share of recommendation: the metric beyond share of voice
AI Share of Recommendation measures how often an answer engine recommends a brand across a fixed set of buyer prompts. It goes beyond share of voice, which usually measures whether a brand is mentioned. A brand can be mentioned frequently but still lose if answer engines recommend competitors more clearly.
Updated 2026-05-04
Questions this guide answers
- What is AI Share of Recommendation?
- How is Share of Recommendation different from share of voice?
- What AEO metrics should marketing teams track?
Direct answer
AI Share of Recommendation measures how often an answer engine recommends a brand across a fixed set of buyer prompts. It goes beyond AI search share of voice, which usually measures whether a brand is mentioned. A brand can be mentioned frequently but still lose if answer engines recommend competitors more clearly, more often, or with stronger evidence.
Why mentions are not enough
AI search visibility is not a single metric. A brand can appear in an answer and still fail to win the buyer.
- It is mentioned only as an afterthought.
- It is listed but not recommended.
- It is described with vague or outdated positioning.
- It is cited less often than competitors.
- It is framed as a weaker fit for the buyer's use case.
What AI Share of Recommendation measures
AI Share of Recommendation measures the percentage of relevant prompts where a brand is recommended as a good option, best fit, or preferred choice.
Basic formula: AI Share of Recommendation = prompts where brand is recommended / total relevant prompts tested.
Example: if a brand is recommended in 18 out of 60 category and comparison prompts, its Share of Recommendation is 30 percent for that prompt set. This should be measured by prompt group, answer engine, region, persona, and time period. A blended number is useful for executives, but the operational value comes from the breakdown.
Share of voice vs share of recommendation
A mature AEO program should track all five metrics below, but Share of Recommendation is often the clearest executive metric because it maps to shortlist influence.
| Metric | What it answers | Limitation |
|---|---|---|
| Mention rate | Did the answer name the brand? | Does not show whether the brand was favored |
| Citation rate | Did the answer cite the brand or its sources? | Does not show whether the citation helped positioning |
| Share of voice | How much of the answer space does the brand occupy vs competitors? | Can overvalue shallow mentions |
| Share of recommendation | How often is the brand recommended for relevant prompts? | Requires clear scoring rules |
| Answer accuracy | Is the brand described correctly? | Does not show competitive strength alone |
How to score recommendations
Use a simple, consistent scoring model. The 0-5 scale below is intentionally simple so a team can run a first measurement cycle by hand. SolCrys's production scoring uses richer signals - prompt popularity, engine-specific recommendation cues, position within the answer, and time-weighted recency - calibrated against human-reviewed ground truth.
| Score | Meaning |
|---|---|
| 0 | Brand is absent |
| 1 | Brand is mentioned but not explained |
| 2 | Brand is explained but not recommended |
| 3 | Brand is recommended as one option |
| 4 | Brand is strongly recommended for the prompt's use case |
| 5 | Brand is recommended, cited, and differentiated against alternatives |
Measure by prompt intent
Share of Recommendation should not be averaged across every possible prompt. Segment it by intent.
Category prompts
Show whether the brand appears in initial discovery. Example: 'What are the best platforms for answer engine optimization?'
Use-case prompts
Show whether the brand is connected to specific buyer needs. Example: 'What should a B2B SaaS company use to monitor and improve ChatGPT visibility?'
Comparison prompts
Show whether the brand wins against known alternatives. Example: 'Compare AI visibility dashboards with AEO execution platforms.'
Risk prompts
Show whether the brand is trusted in sensitive moments. Example: 'How can companies prevent AI tools from hallucinating brand claims?'
Retail prompts
Show whether shopping assistants recommend products. Example: 'What is the best protein bar for a low-sugar snack after workouts?'
What improves Share of Recommendation
Share of Recommendation improves when answer engines can clearly retrieve and verify why a brand is a strong fit for a prompt. The point is not to manipulate answers. The point is to make the strongest truthful answer easier to find and cite.
- Strengthen category pages with direct answer blocks.
- Add comparison pages for high-intent competitor prompts.
- Publish use-case pages tied to personas and jobs-to-be-done.
- Add proof sections, tables, FAQs, and current facts.
- Improve product and marketplace listing clarity.
- Align third-party source strategy with recurring answer gaps.
- Fix inaccurate or outdated brand descriptions across owned pages.
How SolCrys tracks it
SolCrys measures prompt-level visibility, citations, competitors, answer accuracy, and recommendation quality across answer engines. The platform helps teams identify where the brand is merely mentioned versus where it is actually recommended, then maps weak recommendation patterns to content and source actions. That makes Share of Recommendation a working metric, not just an executive chart.
FAQ
What is AI Share of Recommendation?
AI Share of Recommendation is the percentage of relevant AI search prompts where an answer engine recommends a brand or product as a good fit.
How is it different from AI search share of voice?
Share of voice measures visibility or presence. Share of Recommendation measures whether the brand is actually recommended in the answer.
Why can a brand have high mention rate but low recommendation share?
Because answer engines may mention the brand in a list but recommend competitors more strongly, cite competitors more often, or frame the brand as a weaker fit.
What prompts should be included?
Include category, comparison, alternative, risk, implementation, persona, and brand-specific prompts. For ecommerce, include use-case and shopping recommendation prompts.
How can SolCrys help improve Share of Recommendation?
SolCrys identifies prompts where the brand is not recommended, diagnoses the likely reason, and helps teams execute content or source actions grounded in approved Corporate Context.
Related guides
Measurement
AI Search Share of Voice
Measure how often a brand appears versus competitors in AI-generated answers, citations, and recommendations.
Prompt Intelligence
AI Search Prompt Set
A practical guide to building an AI search prompt set across category, comparison, risk, implementation, competitor, and brand-specific prompts.
Measurement
AI Brand Visibility Monitoring
A practical guide to measuring brand mentions, citations, sentiment, and competitive position across AI answer engines.
AI visibility audit
Find out where your brand is missing, miscited, or misrepresented.
SolCrys maps high-intent prompts to mentions, citations, answer accuracy, and content gaps so your team can prioritize the next pages to ship.