Implementation

Small, cheap, and easy to abandon.

This framework needs no license, software, training programme or budget line. It is a reasoning order you can run on a sheet of paper. That is deliberate — anything heavier tends to become theatre.

Six steps

The whole practice.

01

Pick one recurring friction

Not a culture initiative. One thing that keeps happening: the same handoff, the same hour, the same kind of error. A narrow subject is what makes the exercise honest.

02

Write only the facts first

Observable, checkable statements. No adjectives about people, no inferred motives. This step is harder than it sounds and it does most of the work.

03

List variables without choosing one

Skill, clarity, workload, tooling, authority, information, conditions, recovery. Write them all down before you have a favourite.

04

Rule out what the facts rule out

State the reason each explanation was eliminated. An explanation removed without a reason is not removed; it is just currently unpopular.

05

Name what stays uncertain

Write the open questions down where others can see them. This is the step that keeps the exercise from becoming a more articulate way of blaming someone.

06

Change one thing you can actually change

Then look again in two weeks. If nothing changed, the diagnosis was probably wrong — which is information, not failure.

How it fails

The predictable ways this goes wrong.

Stated here rather than discovered later.
  • Turning it into a form

    The moment it becomes a mandatory template, people fill it in to finish it rather than to find something.

  • Using it only after incidents

    If it appears only when something has gone wrong, it will be read as an investigation and answered defensively.

  • Using it about people rather than with them

    A diagnosis performed on someone in their absence is a rumour with better structure.

  • Letting it replace accountability

    If expectations quietly disappear, the practice has been misused and will lose credibility fast.

  • Scaling it before it has worked once

    A single honest case is worth more than an organization-wide rollout of something untested.

Conditions for using it well

  • Expectations, standards and consequences stay exactly where they were.
  • The person the case concerns is part of the conversation, not the subject of it.
  • Nothing recorded should be personal health information or anything a person would not want written down.
  • Anything resembling a clinical concern goes to a qualified professional, not into this process.
  • If it stops producing anything useful, stop using it.

Support

If you want a second set of eyes.

Pilots are small and slow on purpose. Honest reports of where this fails are as useful as reports of where it works.