What an honest report of a trial says

What the report would have said

At week twenty-six, counting in absolute weeks from the start:

Which produces, accurately:

Thirty-eight procedures were removed. Seventeen were restored in some form, most at reduced cost. Two gaps were identified and closed. Half the removals have produced no consequence of any kind. Output and service levels were maintained throughout.

Nothing is shaded. The trial was run honestly, measured properly and reported accurately.

What it cannot say

Three of the five problems had not yet happened, and at week twenty-six they were sitting inside the nineteen entries that had produced nothing — which is to say, they were counted on the positive side.

"Half the removals have produced no consequence of any kind" included the shadowing, the refusal log and the reconciliation, which between them went on to cost three departures, four duplicated pieces of work and a scramble through an audit.

The nineteen are therefore not a neutral blank in the report. They are carrying three future problems on its credit side.

Three responses that do not work

Be more cautious. Hedging every sentence produces a document that says less and is read less, and the window is still twenty-six weeks.

Wait longer. The decision cannot wait eighteen months. People have arranged their lives around the answer.

Measure more things. The metrics were not the problem. Output, satisfaction and retention were measured correctly and reported honestly. Four more of the same kind are four more things that are flat.

The four sentences that do work

1. The window. "This report covers twenty-six weeks from the start of the transition."
2. What sits outside it. "Thirty-eight procedures were removed. Nineteen have produced no event of any kind, and we cannot distinguish those that were unnecessary from those whose consequences have not arrived."
3. The shape. "Events so far have arrived between one and twenty-four weeks after the removal. On records of this kind, events continue to arrive well beyond the reporting window."
4. Who looks next. "[Name] will review the list at twelve and eighteen months."

No claim is weakened. Nothing is delayed. Nineteen silent entries stop being evidence and become an open question with an owner, which is what they are.

Why the fourth gets dropped

The first three cost nothing and are descriptive. The fourth commits a named person to doing something eighteen months later, at a point when the project has closed and the people have moved on.

There was no fourth sentence on this record and no report of this kind. The list survived because one person kept it in their own files and happened to still be there.

The report nobody writes

The document that would help most is the eighteen-month note: two pages saying what was removed, what became of each, which turned out to matter, how long each took to find out, and what would be done differently.

It has no audience. By eighteen months the arrangement is permanent, nobody is asking, and the only parts that are novel are the embarrassing ones.

The report that would help is written after everybody has stopped needing it. That is not a failure of diligence. It is a mismatch between when information arrives and when anybody wants it.

What this site is

That note, written out at length, for a transition about which nobody has any further decisions to make.

Its value is not to the organisation it describes, which absorbed all of it the hard way. Its value, if it has any, is to somebody at week two of their own transition, with thirty-eight decisions in front of them and an afternoon in which to write them down.

The version for the organisation that will not write it

Most organisations will not produce the four sentences, for the ordinary reason that nobody is asking for them and the report has a deadline.

The fallback is smaller and it survives better, because it is not part of the report at all.

Attach the list to the report as an appendix, unedited, with the outcome column as it stands on the day. Nineteen rows will be blank. Nobody will read it.

That costs nothing, requires no judgement and no additional sentence, and it leaves the one artefact that makes a later read-back possible inside a document that gets archived rather than in somebody's personal folder.

On this record the difference between a list that survived and a list that did not was a single person's filing habits. An appendix is a worse place for it than a named owner with a calendar entry, and it is considerably better than nothing, which is what the alternative turned out to be.