Data Core · Performance Analysis · Module

Acting on it

What changed as a result, and whether the change worked — without which the analysis was a document rather than an input.

The idea

How it works

An analysis that is presented, agreed with and changes nothing has cost the same as one that changed something. The difference is what happens in the week afterwards.

Recording which analyses produced action is uncomfortable and informative: a function whose work changes nothing has a positioning problem rather than a quality problem.

Working with it

In practice

  1. 01

    Name the action before the presentation

    What the recommendation is, who would do it, and by when.

  2. 02

    Check whether it happened

    A month later. The rate at which recommendations are implemented is a measure of the analysis function.

  3. 03

    Measure the outcome

    Did the change produce what the analysis predicted? This is how analytical judgement improves.

  4. 04

    Record the analyses that changed nothing

    A pattern of them says something about scoping rather than about execution.

One level in

The components of acting on it

A component is something that exists afterwards which did not exist before — a deliverable or a mechanism, not an intention.

  1. The recommendation

    What should be done, by whom, by when.

    Learn
  2. The implementation

    Whether it was actually done.

    Learn
  3. The outcome

    Whether the change produced what was predicted.

    Learn

Track how many analyses produce an implemented change. It measures the function better than any quality review.