How to read the decision log and check a track record

Decisions in the sidebar holds two tabs: the log of everything recorded, and the Track record of how those decisions turned out. See What are decisions and track records? for the concepts.

What is in the log?

A sortable table: Decision, Affects, Who, When and Verdict.

Affects is a chip naming the record the decision was recorded against. Clicking it opens that record, which is usually where you want to be.

How do you find something?

Sort by decision, person or date by clicking a heading. Search the affected record. Filter by person, by date range, and by verdict state.

Three searches are worth knowing.

By affected record, to answer "what have we decided about this supplier". That is the one you will use most, usually just before a call with them.

By date range, to reconstruct a season. Filtering to last spring produces the reasoning behind a whole buying cycle in one screen.

By verdict, to see everything that missed. That is the most useful filter in the whole product and the one nobody thinks to use.

The log shows the most recent decisions and tells you when it has reached that edge, asking you to narrow the date range rather than quietly showing a partial list.

What is on the Track record tab?

The same decisions grouped by Supplier, Person or Team, showing how many commitments closed, how many were met, missed or could not be measured, and the hit rate.

Only decisions with a committed outcome appear, so this is a smaller set than the log. See How to commit to an outcome and have it measured.

Somebody in two departments counts in both rows of the Team view, and the screen says so rather than leaving you to wonder.

How should you read a hit rate?

With the count next to it, always.

A hit rate over four closed commitments tells you almost nothing. Over forty it is a genuine signal, and it is the sort of signal most brands have never had.

A dash rather than 0% means nothing could be judged, which is a statement about your records rather than about the decisions.

What is this actually for?

Three uses, in rising order of value.

Answering "why is this like this". Somebody finds an odd payment term or an unusual supplier and the log says why.

Preparing for a conversation. Every decision about one supplier, with what was rejected, before you talk to them.

Learning what your business is good at. Filter to missed verdicts across a year and read the reasons. That is the closest thing to a post-mortem most brands ever run, and it costs a minute per decision to have made it possible.

Can you ask instead of reading?

Yes. The analyst can answer what was decided, why, what was rejected and what was measured.

It needs the decisions view permission, and it reports decision and outcome side by side without claiming one caused the other. See How to ask about stock, suppliers and marketing.

Why is the log empty?

Because nothing has been recorded yet. The empty state points at your most recent real purchase order rather than inventing an example, which is the obvious place to start.

Record one decision on your next significant order, per How to record a decision. The log is worth exactly what people put into it.