We work on what AI says about your brand
Buyers now ask ChatGPT, Perplexity, Gemini, and Google's AI Overviews before they ask a search box. Those answers are assembled from a handful of sources, and being one of them is not the same problem as ranking. Geology measures which sources an AI answer is built from, prompt by prompt, and then changes them.
Ranking and citation are different problems
Search “best health insurance india” and the top organic result is a large carrier. The AI Overview printed above it cites a different set of pages entirely, including an aggregator sitting several positions lower and two sites almost nobody has heard of. The number one result supplies no source material to the answer above it.
That gap is our whole reason for existing. Traditional SEO tools measure position. We measure whether the model actually used you.
Founders
Mehul works at the intersection of AI, SEO, and generative engine optimisation, and runs an AI services company helping startups and enterprises deploy AI systems and automation. He writes most of Geology's research on insurance and financial services, and leads the India market.
Sankalp is a product designer and entrepreneur based in New York. He spent four years at Postman across product and design before co-founding Geology, and focuses on the measurement side: what brands can actually observe about their presence in AI answers, and what they can change.
Memberships and events
A not-for-profit body promoting technology across the Indian insurance value chain, with members spanning carriers, reinsurers, brokers, health-tech, and technology providers.
Held at the Jio World Centre and Trident BKC in Mumbai, organised by the Payments Council of India, NPCI, and the Fintech Convergence Council, under the patronage of the RBI, SEBI, IFSCA, and PFRDA.
Announcements are collected on the press page.
We publish our own research
Most of what gets written about AI search is recycled advice. We run the prompts, record the citations, and publish what the data shows, including when it contradicts us.
- Insurance AI Visibility 2026100 prompts across 4 AI engines, 15 insurers.
- GEO for B2B SaaS3,352 citations across 881 domains, 12 hypotheses tested.
- How ChatGPT Decides What to Recommend83 real shopping conversations, 500+ product mentions.
Where we focus
Insurance and financial services, more than any other sector. Both are heavily regulated, which pushes brand content toward hedged, compliance-safe phrasing that AI systems extract badly. Both are also categories where an AI answer carries real consequences for the person reading it. Writing something quotable that still clears a compliance review is a narrow skill, and it is the one we have built the most of.
See insurance and financial services, or the insurance case study.
Ready to grow your AI visibility?
Run a Live Audit and see how your brand performs across AI platforms like ChatGPT, Perplexity, and Google AI Overviews. Full report in your inbox in under 15 minutes.