AI Visibility Benchmark
We run 50 standardized prompts monthly across ChatGPT, Gemini, and Perplexity to measure how often AI systems mention Auto Shop Digital — and how we compare to competitors. This is our public accountability tracker.
Methodology note: The August 2026 figures (10% mention rate, 11% linked-citation rate) are directional baselines from the Auto Shop Digital SEO & AI Content Strategy workbook, not a universal market-share measurement. AI answers are stochastic — results vary by run, model version, and date. Each monthly update uses a clean environment, fixed prompt wording, and multiple runs per prompt. Do not treat any single answer as a stable ranking.
Competitor Mention Rates
How often each agency is mentioned across the 50 tracked prompts. August 2026 baseline — directional only. Market leaders benefit from higher review counts, territory exclusivity claims, and longer-standing third-party editorial coverage.
6-Month Targets
Targets set in August 2026 based on the strategy document. We will recalibrate after 60 days of reliable data.
| Metric | Aug 2026 Baseline | Feb 2027 Target | Status |
|---|---|---|---|
| AI Mention Rate | 10% | 20%+ | In Progress |
| Linked Citation Rate | 11% | 18%+ | In Progress |
| Top-3 Mentions | 2 prompts | 8+ prompts | In Progress |
| Purchase-intent prompts | 0 mentions | 3+ mentions | Not Started |
| Third-party editorial mentions | ~2 | 6+ | Not Started |
The 50 Tracked Prompts
Prompts are organized across five buyer-journey categories. Each category contains 10 prompts tested across national and selected local US contexts. Click any category to see its full prompt list.
Methodology
Each month, we run all 50 prompts in a clean browser session (no prior conversation history, no personalization) across ChatGPT (GPT-4o), Google Gemini (1.5 Pro), and Perplexity AI. We record whether Auto Shop Digital is mentioned, whether a link to our site is included, the position or order of the mention in the answer, the sentiment and factual accuracy of any claims, which competitors are mentioned in the same answer, and which third-party sources the model cites.
For stochastic systems (ChatGPT, Gemini), we run each prompt three times and record the majority outcome. We log the date, model version, and run number for every result. We do not present any single answer as a stable or universal ranking — AI answers change with model updates, retrieval index changes, and prompt phrasing.
What this benchmark measures: whether Auto Shop Digital has sufficient third-party corroboration, content authority, and entity clarity to be cited by AI systems when buyers ask relevant questions. It is a proxy for the quality and breadth of our external footprint, not a direct measure of traffic or revenue.
What this benchmark does not measure: universal AI market share, the full range of possible prompts, or performance in non-US contexts. The 10% and 11% baseline figures from August 2026 are directional starting points from the strategy workbook, not independently audited statistics.
Authority-building priorities that move this benchmark
- Publishing original benchmarks and research that other sites cite
- Earning editorial mentions in Ratchet+Wrench, ShopOwner, Motor Age, and ASA-related outlets
- Building Daniel Harman's expert entity with consistent bio, expertise topics, and social profiles
- Client case-study co-promotion so third-party business sites corroborate the work
- Reconciling and claiming important third-party directory profiles