Research programme · Quarterly
Platform Comparison Research
How often do ChatGPT, Gemini, Claude, Perplexity and Google AI agree on which local businesses to recommend?
No findings published yet
This page describes a study that has not produced results. There are no figures on it because none have been measured — zero live AI observations have been collected to date.
It is published now so the method, the planned output and the limits are on record before the data exists, and can be compared against the study when it lands.
How it will be measured
Cross-platform agreement analysis on identical prompts run in the same period and geography, reporting overlap in named businesses and in cited sources.
The underlying benchmark is the same one every client measurement uses: 24 fixed buyer-intent prompts across 5 platforms, each run 2 times, for 240 recorded observations per cycle. Prompts are frozen between cycles, which is what makes one period comparable with another.
What the study will contain
Written before collection starts, so the deliverable cannot be reshaped around a convenient result.
- Overlap in the businesses named by each assistant for identical prompts in the same period.
- Overlap in the sources each one cited.
- Which platforms were most and least consistent with themselves across repeated runs.
What it will not be able to tell you
Stated in advance. A study that does not name its own boundaries is asking to be over-read, and research that gets cited is research that was honest about its limits.
- Identical prompts, same period, same geography — real users vary all three.
- Consistency is not accuracy. A platform can be reliably wrong.
The rules this research is published under
- Rankings cannot be purchased. There is no paid placement in any Greenbug index.
- Being a Greenbug client cannot raise a score or a ranking. Client scores are computed by the same formula from the same kind of evidence as everyone else's.
- Where Greenbug has a commercial relationship with a business that appears in published research, that relationship is disclosed on the page.
- If sponsored placement is ever introduced, it will be visually and structurally separated from editorial rankings and labelled as sponsored.
- Greenbug.ai measures itself with the same methodology and publishes its own result, including when it is unflattering.
The rest of the programme
- Research in progress
DMV AI Recommendation Index
Which businesses do AI assistants actually recommend across DMV service categories, and how consistently?
- Planned
State of DMV AI Recommendations
How is AI recommendation behaviour across the DMV changing over a full year?
- Research in progress
AI Citation Studies
Which websites do AI assistants actually rely on when recommending local businesses in this region?
- Planned
Industry Recommendation Benchmarks
What does a typical AI Recommendation Score look like in a given category, so a business can tell whether its own score is good?
- Research in progress
Local Business Recommendation Studies
How does question phrasing — city versus county, 'near me' versus a named place — change which businesses an AI recommends?
Measure your own business
The research above runs on the same benchmark a client measurement uses. Yours starts with a baseline.