Ask a decision question, not a feature request
Define the user and the decision you need to make. Include constraints such as implementation capacity, the next experiment and the information that would change your view. A useful question might be whether a reporting problem should be tested with a review checklist or a formal approval step.
Bring attributable customer or market sources. Separate accepted facts from assumptions. If the evidence pack is missing or its trace is incomplete, complete the evidence work before treating an evaluation as ready for a decision.
Follow the evidence through the decision
Saved context helps you revisit why an option was chosen or rejected. It does not remove the need to verify that older evidence still applies. A new source or a changed constraint may justify revisiting the question without silently rewriting the earlier decision.
- Select the product question and permitted sources.
- Analyze the evidence and inspect its provenance and gaps.
- Compare alternatives against the question and constraints.
- Review the critique, risks and remaining uncertainty.
- Make the required founder decision, then inspect the saved result and next experiment.
Two options that solve different problems
Synthetic example: one account needs a report approver; another wants confidence that its report is correct. Neither has committed to buy an approval feature. The decision below is an educational illustration.
- Option A
- Test a report review checklist with both accounts.
- Option B
- Test a formal approval step with the account that named an approver.
- Trade-off
- A checklist may help verification; an approval step may help ownership. They answer different questions.
- Evidence gap
- No observed trial yet shows which approach changes the user's workflow.
- Founder decision
- Choose one bounded experiment, document the reason and define what would make you reconsider.
A recommendation still needs a decision
Check whether each claim can be traced to evidence. Inspect alternative explanations and whether the critique actually addresses the strongest objection. An attractive recommendation is not proof that a feature is valuable.
Product Decisions does not autonomously set company strategy or authorize implementation, customer communication or external changes. Required human decisions remain attached to the workflow. Use the outcome to define a test you can observe, not to create certainty where the inputs are weak.
Access and review
Closed Beta requires an application, manual approval, invitation and verified login. Applying does not grant immediate access. Six ready-made teams, guarded Custom teams and the Apps catalog are included. No Oqera payment or subscription is required. Bring your own AI provider key and pay the provider directly. External actions require specific permission and review; unattended schedules and triggers are off.