Research workflow
Turn open-ended research into a decision-ready action plan
An illustrative workflow for moving from a crowded research question to a bounded decision, a short evidence pass, and one concrete next action.
For product builders, operators, and independent researchers who have gathered plenty of material but still need to decide what the work should change.
Starting point
Make the situation specific before making it efficient.
You have a broad question, a folder of links, and several competing interpretations. More reading is possible, but the decision that the research needs to inform is still vague. The risk is producing a polished summary that never changes what happens next.
A reviewable path
Keep the sequence small enough to inspect.
- 1
Name the decision
State who will use the research, what they must decide, and when the answer becomes useful. Keep open questions visible instead of hiding them in a general topic label.
- 2
Set an evidence floor
Choose the source types, recency, and contradictory evidence that must be reviewed before a recommendation is credible. Stop when that floor is met, not when every possible link is exhausted.
- 3
Separate fact from inference
Record what sources directly establish, what you infer across sources, and what remains unknown. This makes the recommendation easier to review and revise.
- 4
Convert the finding
Create a bounded next action with an owner, a success signal, and a check-in date. If the evidence is weak, choose a reversible test rather than presenting certainty.
Review point
Use what happened to shape what comes next.
At the review point, compare the decision made with the original evidence floor. Capture what changed, what the action revealed, and whether the remaining uncertainty deserves another research pass.
Boundaries
This example is an educational workflow, not a claim that Wizzo or AI-generated research is complete or correct. Source verification and consequential decisions remain with the user.