ASSESSMENT
Find out whether people on this team can say the difficult thing.
Seven statements, answered in about three minutes, produce a single team score and one of four bands. It measures whether admitting a mistake, raising a risk or disagreeing carries a cost here — the condition that decides whether every other measure you take is answered honestly.
You receive
There is one thing being measured here, not a profile of several — and it is measured because each of these four behaviours has a price attached when it goes missing. The seven statements approach it from these four directions.
Whether an error gets named while it is still cheap to fix, or absorbed quietly and discovered downstream — where the same error costs many times more to put right.
Whether flagging a risk is read as useful or as obstruction. Where the person who flags it pays for having done so, they stop, and you lose your earliest warning system.
Whether asking is ordinary practice or evidence that someone is not up to the work. Teams that cannot ask do not stop being stuck — they just get stuck privately, for longer.
Whether a dissenting view survives the meeting it is raised in. Unanimous agreement is often reported as alignment when it is actually the absence of anyone willing to test the plan.
Low psychological safety almost never shows up on a report labelled as such. It shows up as a defect found three stations downstream that somebody noticed at station one. As a project slipping in the final fortnight after months of green status. As the resignation that arrives with no warning, from the person who had in fact given several.
Each of those is expensive, and each is normally investigated as a process failure. Sometimes it is. Often the process worked exactly as designed and the information simply never entered it, because the person holding it had learned what raising it would cost them.
Not a diagnosis of everything, and not a project. One reading that tells you whether the problem you are chasing is a process problem or a candour problem — because the two have completely different remedies and are routinely confused.
Every survey, retrospective and skip-level you run assumes people will say the true thing. Where that assumption does not hold, the rest of your instruments return confident, well-formatted noise. This is the one that checks the assumption.
One total and one of four bands, each with a written evaluation of what it means for the team. The number shown against each band below is the top of its range.
When issues are consistently discovered downstream, the question is usually not detection. It is whether anyone felt able to say so upstream.
Three minutes to establish whether the longer instrument you are about to run will be answered candidly at all.
A first reading taken early, and the same reading taken again two quarters later, is the cheapest evidence of whether anything changed.
Rigour
This is a team climate reading, not a clinical instrument and not an individual assessment. It describes how a group reports its own conditions at one point in time. No result from it should be used as an input to a hiring, promotion or disciplinary decision, and a single administration is a starting point for a conversation rather than a conclusion.
Psychological Safety