How to Use Claude for Market Research: The Serena Skill Guide

Market research is easy to fake and hard to do well. Ask a chatbot to size a market and it will hand you a confident figure to two decimal places — with no source, no method, and no way to know whether it is real. When people use Claude for market research without structure, that is the risk: numbers that look authoritative and fall apart the moment an investor asks where they came from. A real AI market research process values method and honesty about sources over a tidy figure.

Serena is that process. She is a research persona you load once into Claude, ChatGPT, or any AI chat, and she does two things a bare prompt does not: she structures the research properly, and she is honest about what she does and does not know. Where a precise figure would need a primary source, she says so and tells you what to verify, rather than inventing one.

Why generic AI market research misleads

An unguided model fills gaps with plausible fiction. It produces market sizes, growth rates, and competitor revenues that sound credible and may be entirely made up. It has no framework, so the “research” is really a paragraph of impressions. And because it states everything with the same confidence, you cannot tell the well-grounded reasoning from the guesswork — which is exactly the distinction that matters when real decisions ride on it.

What changes with the Serena skill

Serena works through frameworks rather than impressions. She sizes a market from the top down and the bottom up, and shows the logic so you can challenge the assumptions. She breaks a market into segments with real characteristics, builds competitor matrices that compare on the dimensions that matter, and turns findings into customer personas you can act on. Crucially, she labels every figure that needs verification and points you to the kind of source that would confirm it — treating unverified numbers as questions, not facts.

What it actually produces

A market sizing framework with stated assumptions, segment profiles, a competitor comparison matrix, a SWOT for your position, draft customer personas, and a research plan listing the figures to verify and where to find them. It gives you the structure of a research report and is candid about which cells still need a primary source filled in.

How to get the most out of it

Give it everything you already know — your market, your competitors, any real figures you have — so it reasons from facts rather than filling space. Ask it explicitly to flag every number that needs verifying, and use it to build the structure and the questions rather than to source hard statistics it cannot confirm. The output is a strong research scaffold; you complete it with primary data.

Who this is for

Founders writing a business plan or pitch, product teams sizing an opportunity, and marketers mapping a competitive landscape. It works with Claude, ChatGPT, or any AI chat that accepts a system prompt. For the rest of the go-to-market workflow, the marketing skills collection covers positioning, messaging, and campaign planning — each a focused assistant rather than a general chatbot.