Enough research. What are we trying next?
A research brief should make the next experiment easier to choose.
Retrospective reconstruction of the week beginning 2025-02-24, written on 16 November 2026. This is not a claim of publication at the time.
Deep research had been available for a few weeks. The tool could help gather material; deciding when the investigation was sufficient still needed a stopping point.
For a tool comparison, that point might be simple. Two options meet the essential requirements. One has a lower setup cost; the other handles an edge case that has not happened yet. Run the common task on the simpler option. Keep the alternative in the notes.
The brief should make that choice easy to see. Put the recommendation first, then the reason, then the detail someone may want to challenge. Do not make the reader scroll through every candidate before discovering which one is worth trying.
If the evidence is weak, choose a small test that could change the answer. A sample import, a short usability session or one representative job can settle something another ten articles cannot.
Leave the unresolved questions visible. Some will matter later. They do not all need to become blockers today. The next experiment should have a result someone can inspect, including a result that sends the project in a different direction.