Compare
Not another reading list.
A well-made research brief is a genuine service: someone reads forty sources so that a team need not, and hands back the shape of a subject in twenty pages. The brief, though, is usually organized around a topic. A decision is organized around a handful of claims, and the brief rarely tells you which of its pages bear on them.
Typical research brief
- organizes its sources by topic
- lets background pass for proof
- seldom names the claim being tested
- can gather support for the favored story
- ends when it feels long enough
Argumont
- starts from the claims the choice rests on
- rates each source for what it proves
- records support, contradiction and gaps
- lets missing evidence stay visibly missing
- can reopen research through Challenge the Verdict
The value of a good brief
Background matters. A brief gives a team a shared vocabulary, a map of who the players are and a sense of which numbers count as normal in a field. Without it, a group argues from whatever each member happened to read last.
Organized by topic, not by claim
Most briefs follow the subject's natural chapters: market size, competitors, regulation, trends. A decision instead turns on perhaps three or four claims, each of which could be true or false, and a forty-source brief may speak to only one of them directly. Readers then mistake volume for coverage.
There is also the pull of the question as posed. A researcher asked to support a plan finds support, because that is where the search terms lead, and the contradicting report never gets typed into the search box.
How careful researchers work
Start with the claim, written as a sentence that could turn out false. Search first for what would refute it, then for what would confirm it. Prefer the primary source (the filing, the statute, the contract, the raw data) over a story that summarizes it, and notice when a figure appears in ten places that all trace back to one origin.
Keep absence of evidence apart from evidence of absence. A search that finds nothing has told you about the search, which is a different thing from learning that the claim is false.
Research in a Decision Room
The room begins with the claims each option rests on, and Scout investigates those claims in public sources rather than the topic at large. Inspector then rates each claim as supported, partly supported, unverified or contradicted, and says what the source actually proves.
Where coverage is thin, the Judge says so in the Verdict instead of filling the gap with confident prose, and the honest result can be No Strong Winner Yet. Challenge the Verdict can reopen the research if the first pass missed a contradiction. What you then choose is the Human Decision, recorded on its own.
Thirty-two pages and two claims
A group of clinics weighs opening a site in a growing town. Its brief runs to 32 pages and 41 sources on demographics, insurers and local employers. In the room, the choice reduces to two claims. The first, that the town is growing about 3% a year, is supported by the county's own published estimates.
The second, that the nearest rival clinic will close when its lease ends, rests on one news story quoting an unnamed source, and Inspector marks it unverified. The plan's forecast of 52 visits a day assumes the rival is gone, while the site breaks even at 38. The Verdict favors opening only if the town supports 38 visits a day with the rival still trading, and asks for that estimate before any lease is signed.
Bring Argumont a Decision worth challenging.
Start with a paid 60 day pilot on your own real Decisions.