The market research automator.

You have a question about your market and no good way to answer it that does not eat two days. This turns the question into a structured brief with sources you can open, in about an hour, and it tells you plainly where the evidence is thin instead of filling the gap with confident noise.

Guide, 2026, ~4 min read

What it replaces.

The two-day tab spiral. Forty open tabs, six half-read reports, and a document that still says "TBD" where the actual answer was supposed to go.
The deck built on a number nobody sourced. A statistic that entered a slide once and has been repeated ever since, and nobody now remembers where it came from or whether it was ever true.
Asking an AI and getting fluent nonsense. A clean-looking summary with invented figures and citations that do not resolve, which is worse than no answer because it reads like one.

The starter prompt.

Use a model with web access for this one. Write your real question into the first line, in the words you would use talking to a colleague, then run it. Expect to go one round of follow-ups before the brief is worth keeping.

You are producing a market brief for me. My question is:

[write your actual question here, as specifically as you can]

Work in this order and show me each step.

First, restate my question as the three sub-questions that actually have to be answered. If my question is too vague to answer well, tell me so and give me the sharper version before continuing.

Then answer each sub-question. For every claim you make, give the source and the date. If a number is older than two years, say so next to it. If you cannot source a claim, write "unsourced" beside it rather than dropping it or dressing it up.

Then give me a section called "What I could not establish", listing what is genuinely unknown or contested, and what it would take to find out.

Finish with the three things I should do differently based on this, and the one thing that would change your answer if it turned out to be false.

Do not smooth over gaps. A short brief with honest holes is more useful to me than a complete-looking one I cannot trust.

Make it stick.

Open three sources before you believe any of it. The section that earns the trust is the citations. Spot-check three at random every time. A model that has learned you check stops inventing, and one that gets away with it will do it again on the brief that matters.
Keep the question, not just the answer. Save the exact question and the date beside the brief. Markets move, and in six months the useful thing is knowing precisely what you asked so you can ask it again and see what changed.
Let the gaps drive the next step. "What I could not establish" is the most valuable section on the page. It is your list of what to ask a customer, a partner or an analyst, and it is usually shorter than you feared.

The brief is the input to something else: a deck, a positioning call, a decision about where to spend next quarter. Hand the finished brief to the deck workflow and the slides assemble from evidence rather than from memory.

Not sure this is your first piece? The toolkit page picks for you.

Curious how a piece like this becomes part of how the company runs? How it works.