Negotiation and Debate - Best Practise¶
Pattern: the Position-to-Decision Loop¶
- Frame the choice. Write the decision, deadline, decision owner, and cost of waiting.
- List positions and interests separately. A position is "ship Friday"; an interest is "give the customer credible proof before the renewal meeting."
- Agree on criteria before defending options. Use measures that can prove or disprove an option, such as a load-test result, rollback time, customer date, or support impact.
- Write options before judging them. Include an option that changes scope, exposure, timing, or sequencing rather than only arguing for the two opening demands.
- Get focused challenge. Ask people to comment on evidence, missing risks, and the criteria. Use silent writing first when hierarchy or strong personalities could shape the room.
- Record the decision. State the option, why it won, the people who own the next steps, and the evidence that would reopen it.
- Commit after the call. Challenge a decision respectfully before it is made. Afterward, support the work unless new material evidence appears.
Start with this template¶
Decision: [The exact choice]
Decision owner and deadline: [Who decides and when]
Cost of waiting: [What happens if we do not decide]
Positions: [What each side says it wants]
Interests: [Need, risk, or constraint behind each position]
Criteria: [Facts or measures the option must meet]
Options: [At least two distinct paths]
Decision: [Chosen option and why]
Actions: [Action | owner | due time | proof of done]
Review trigger: [New evidence that would reopen the choice]
Use it in the export-launch example¶
- Frame: Decide what customers can safely see by Friday, with the product lead as decision owner.
- Interests: Product needs proof for the renewal meeting; engineering needs a proven stop-and-recover path; support needs a clear customer promise.
- Criteria: Load test passes, rollback completes within 15 minutes, and the customer sees the real workflow.
- Options: Full release, staging demo, or one-customer feature-flag pilot.
- Decision: Pilot only if both safety checks pass by Thursday; otherwise use the demo and move the pilot to next week.
- Commit: Product prepares the customer message, engineering owns the checks and flag, support owns the pilot contact, and the group reviews the evidence Thursday.
Make it a habit¶
- Before a difficult meeting, write the decision in one sentence and ask the decision owner to confirm it.
- When someone states a demand, ask: "What problem does that solve for you?" Then repeat the answer until they agree you understood it.
- Bring one piece of evidence that could change your own mind, not only evidence that supports your preferred option.
- Before speaking in a high-stakes debate, write the strongest version of the other side's argument. If they would not recognize it, ask more questions.
- End every decision meeting with the choice, owner, actions, and review trigger in the shared record.
- Run a short retrospective after a hard decision: did the criteria predict the result, did people raise concerns early, and did the group follow through?
Self-check¶
- Is the decision clear, with one owner and a deadline?
- Did we separate stated positions from underlying interests?
- Did we agree on evidence and limits before choosing?
- Did we create more than one meaningful option?
- Did people have a safe way to raise a concern?
- Is the decision, reasoning, action ownership, and review trigger written down?
- Are we committing to the chosen option unless new material evidence appears?