THE RESEARCH LAB
Researching How Brands Win in AI Search
I’m documenting real-world experiments around ChatGPT visibility, GEO, AI share of voice, citations, and contextual advertising.
01 / RESEARCH PIPELINE
Questions worth testing.
Upcoming projects and working protocols. Planned studies are not completed findings; dates will be added as each project begins.
Planned
Date: To be scheduled
Can You Build Authority in ChatGPT in 90 Days?
A proposed 90-day study of brand mentions, citations, and recommendation context as an expert builds an attributable public presence.
Read Research
Planned
Date: To be scheduled
How ChatGPT Chooses Sources
An investigation into which sources appear in answers across different question types and conversation contexts.
Read Research
Planned
Date: To be scheduled
Measuring AI Share of Voice
A working measurement framework for brand representation within a defined set of AI answers.
Read Research
Planned
Date: To be scheduled
Third-Party Citations and AI Visibility
A study of the relationship between relevant external coverage, source attribution, and brand visibility in AI answers.
Read Research
Planned
Date: To be scheduled
ChatGPT Ads vs Google Ads
A planned comparison of conversational advertising and traditional search advertising, where platform access permits.
Read Research
Planned
Date: To be scheduled
AI Search Visibility Benchmark
A proposed repeatable benchmark for tracking mentions, citations, and recommendations across AI platforms.
Read Research
02 / CURRENT EXPERIMENT
Building Jeet Banerjee’s AI Search Authority From Day Zero
A documented baseline for understanding how an independent expert becomes visible in AI answers. The fields below are awaiting measurement; they are not zero values or reported results.
Starting AI Mention Rate
Baseline pending
Share of tracked responses that mention Jeet.
AI Citation Rate
Baseline pending
Share of tracked responses citing an attributable source.
Commercial Share of Voice
Baseline pending
Representation within a defined commercial-intent prompt set.
Number of tracked prompts
Count pending
Final prompt set and sampling protocol to be documented.
External authority placements
Count pending
Verified third-party placements, with source links.
Measurement notes: record the platform, prompt set, collection date, and repeat count alongside each baseline. Distinguish brand mentions from citations and commercial recommendations. Publish the method and limitations with any future findings.