
For your team
Team Practice
Module · Who decides, and who just weighs in
The Team Decision Hygiene Check
Most teams do not have a decision problem, they have a decision-hygiene problem: nobody is quite sure who owns the call, input gets mistaken for a vote, and the same question gets reopened for the third time. AI makes this worse, because a model will happily produce a confident recommendation that everyone treats as a decision nobody actually made. This module checks the five habits that keep decisions clean: clear ownership, telling a decision from an input, the ratio of meetings to actual calls, whether closed decisions stay closed, and whether anyone can reconstruct why you decided what you did.
What the five levels look like
Every dimension in this assessment is scored 1 to 5. This is what the levels mean, dimension by dimension. The graded report diagnoses where your own answers land and what to do about it.
Owners are clear
- 1Nobody owns
- 2Owner unclear
- 3Owner implied
- 4Owner named
- 5Owner named, known
At the low end: When nobody owns a decision, it gets made by accident or not at all, and no one can be held to it. Name a single owner for your next significant call and say the name out loud. What good looks like: Named, known owners mean decisions have someone behind them and someone accountable for them. Keep it current as roles shift; yesterday's owner may be today's bystander.
Decision versus input
- 1Fully blurred
- 2Often confused
- 3Sometimes clear
- 4Usually distinguished
- 5Named up front
At the low end: When input and decision are the same word, everyone consulted feels entitled to veto and nothing closes. Start every real decision by naming who decides and who merely advises. What good looks like: Naming decision versus input up front is a small discipline that prevents a large class of resentment and rework. Extend it to AI output too: a confident recommendation is still just input.
Meetings yield decisions
- 1Almost none
- 2Rarely
- 3About half
- 4Most do
- 5Each has a decision
At the low end: If your meetings almost never end in a decision, they are costing real hours to produce a feeling of motion. Require every meeting to name the decision it exists to make, or cancel it. What good looks like: Meetings that reliably produce decisions respect your team's time and attention. Keep guarding the calendar; the meeting that decides nothing always tries to creep back in.
Closed stays closed
- 1Constantly reopened
- 2Often revisited
- 3Sometimes reopened
- 4Reopened on evidence
- 5Closed, revisited by rule
At the low end: Decisions that get constantly reopened teach your team that committing is pointless, so they stop. Make closed mean closed, and require new evidence, not just new doubt, to reopen anything. What good looks like: Decisions that stay closed unless real evidence arrives let your team build on settled ground. Keep the reopening bar explicit so revisiting stays a discipline, not a habit.
Decisions are recorded
- 1No record
- 2In someone's memory
- 3Buried in chat
- 4Recorded somewhere
- 5Recorded, with rationale
At the low end: A decision that lives only in memory cannot be learned from or defended, and gets reopened the moment memories diverge. Start writing a two-line record for every significant call: what, and why. What good looks like: Decision records with rationale are how a team learns from its own history instead of repeating it. Keep them short and keep them current; a record nobody maintains is one nobody trusts.