Will this work there? And what have we overlooked?
Any firm, including product and strategy teams, can use Market Evidence for testing a business thesis, beyond a market-entry decision. Market Evidence tests your thesis claim by claim against the evidence, in every market you name.
An example
A US company wants to check whether a new product feature works in a set of states, and which requirements, regulatory, development or local, it would miss. Market Evidence runs the same fixed question set against each state, surfaces the requirements the team had not planned for, and names a gap wherever the evidence does not exist yet. The same approach tests any claim your team brings, not only a go-to-market decision.
Sub-markets from a fixed catalogue
Sub-markets, such as US states, German Länder or Swiss cantons, come from a fixed catalogue, which you can group into named sets for comparison. There are no counties and no custom boundaries, so a claim about a region has to be stated as the states or Länder that make it up. See which markets are covered and how the research runs.
Research for a whole country, including Germany, Austria and Switzerland, is told to search and cite in that country’s own official language as well as English.
State-by-state requirements, named individually
A claim like “this feature is compliant in the US” usually means compliant in some states and not others, and a single answer hides which. Testing a business thesis state by state means running the same claim against every state you name, from the fixed catalogue of all 50 states plus DC, so a requirement that holds in California and not in Texas comes back as two separate, cited answers instead of one averaged guess.
States can also be grouped into a named set when the decision is really about a region, without losing the state-by-state requirements behind the group. Nothing is rounded up where the evidence runs out: a state with no public answer yet comes back as a named gap, with a reason, not a blank, so your team knows exactly which jurisdictions still need a human call before the thesis goes further.
Why the usual check falls short
The usual check is a prompt to a general chatbot, or an ad hoc call with someone who knows the market. Both are a legitimate first read, but neither is repeatable: the questions change from one prompt or one conversation to the next, nothing records which claim was tested or against what, and a different person asking the same question later gets a different, uncited answer. For one claim in a market your team already knows well, that may be enough. For a thesis with several claims, across several markets, it stops being something your team can rely on, or hand to someone else to redo, which is why testing a business thesis needs a repeatable method rather than another ad hoc conversation.
What testing a business thesis covers
Testing a business thesis properly means one thing: you bring the claims, and Market Evidence turns each one into a question its research agents test against the evidence, market by market. A claim such as “this market needs a local certification before launch” or “pricing here follows the same logic as our home market” is tested the same way in every market you name, using the same fixed question set, so a difference between two markets is a real difference, not a difference in who asked the question.
Every claim comes back with one of five statuses: Established, where the evidence supports it; Inferred, where it is likely but not directly stated; Absent, where the evidence points the other way; No evidence, where nothing turned up; or Not addressable today, where the research cannot reach it yet. The three gap statuses always carry a reason, so a missing answer is never shown as a quiet zero, and your team can see exactly which claims held, which did not, and which still need a human call.
The research agents work from live web sources and primary records, the same way they do for a market-entry case, and cite the source behind every status. Run a thesis with ten claims across five markets, and you get fifty tested answers back, each one traceable, rather than fifty separate searches your team would otherwise have to run and remember one by one.
A worked example
Say a vendor from the construction space wants to test the claim that a new market lets it sell direct, without a local partner. Market Evidence runs that claim, alongside the rest of the thesis, against the market’s regulatory and commercial evidence, and returns a status your team can act on, not a guess dressed up as confidence.
In this illustrative case, the claim does not survive as stated: the evidence names a licensing requirement that works like a local-presence requirement in practice, so a direct motion without any local party would run into it. That is a useful result in its own right. Knowing a claim is wrong before the team commits budget to it is worth as much as knowing which claims are right, and the gap sits next to the claim, with its source, rather than inside someone’s notes.
The same walk-through runs for every claim in the thesis, in every market you name, so the result is a claim-by-claim record your whole team can read, rather than one person’s notes from one market.
Claim: “This market lets us sell direct, without a local partner.”
Result: the evidence names a licensing requirement that functions as a local-presence requirement, so the claim does not hold as stated.
Absent
Official statistic
What your team provides
Your product or strategy team supplies the claims: the beliefs the thesis depends on, stated as things that could turn out to be wrong. The more specific the claim, the sharper the research: “this segment buys on price” tests better than “this market is competitive.” Input quality follows the same rule as any other case: a claim stated once, by the person who holds it, reaches the research agents faster than one relayed through several people first. Running the same question set against every market and naming every gap is the research agents’ job, not yours.
How it fits into a case
Testing a business thesis inside a case is rarely about one claim; it is a short list, each one testable on its own. Market Evidence tests each claim against the evidence for every market in the case, so the result is a claim-by-claim scorecard your team can read across markets, not one blended verdict that hides which claim is doing the work.
Claim 1
Stated as something that could be wrong, not a hope.
Claim 2
Tested against the evidence for every market in the case.
Claim 3
Returned with a status and the source behind it.
What this does not do
Market Evidence tests claims against public evidence; it does not run experiments, and it does not talk to your customers. It does not promise the thesis will hold, and it does not stop at the first market that looks good: every claim you bring is tested market by market, and a gap is named instead of guessed at. Where the real test is what a handful of prospects think, the most direct route is to ask them yourself; that conversation sits outside what this product does.
Related reading
- Which markets are covered
- How the research runs
- The method
- Testing a business thesis works the same way whether it is a market-entry decision or not.
- Market Evidence vs. self research
Have a thesis to test against a set of markets?
Get a quote for your markets.