Minutes 0–5: read the FTE's own report
Open the Performance tab and read the Digital FTE's monthly report in full. It is short and honest by design: what it delivered against each target, what it missed and why, what it could not measure, and what it recommends changing. Note anything that surprises you; surprises are what the rest of the review investigates.
Do not skip to the numbers. The report is the FTE's account of its month, and reading it first is what makes the review a management conversation rather than an audit.
Minutes 5–10: check targets against evidence
Open the Targets tab and, for each item, check its status against the evidence behind it rather than against the report's summary. A target marked met should show the record that met it; one marked missed should show the gap and the reason; one marked unmeasured should say what is missing. If the report and the evidence disagree, that is the first finding.
Decide provisionally whether each target is measuring the right thing; the first month usually reveals one that is not.
Minutes 10–20: read the send-back reasons grouped by category
This is the most valuable ten minutes. Read every send-back reason for the month grouped by class of action. A category with a repeated reason is a remit gap or a wrong Brand Brain item; write the fix. A category with scattered reasons is fine. A category with no send-backs over a long run is a promotion candidate, to be judged against the written standard. A category that was stable and is now producing send-backs is drift; find the change.
The approval rate is the number; the reasons are the meaning. Never read one without the other.
- Repeated reason: fix the remit or the memory
- Scattered reasons: note and move on
- Long clean run: candidate for promotion
- New send-backs after stability: drift, find the input that changed
Minutes 20–25: read usage and cost per outcome
Read the role's token usage for the month against its budget and divide by the outcomes recorded in Targets for cost per outcome. In the first month, label the figure first-cycle and do not compare it with the baseline yet, because the month included setup, close reading, and remit tuning. Note it; the second month's figure is the one to compare, and the difference between the two shows how quickly the role settled.
Check any external spend the role controls against its own cap.
Minutes 25–28: decide three things
Decide whether the targets are right, changing, adding, or dropping any, and accepting or declining any additions the FTE suggested. Decide whether the remit needs changing, from the send-back patterns. Decide whether any class of action moves on the autonomy ladder: up, only if its run meets the written standard, which in a first month it rarely does; down, if a promoted class produced a send-back.
Declining to change anything is a decision too, and it is recorded like the others.
Minutes 28–30: write the note
Write five lines with the date and your name: the three decisions and the evidence each rests on, and anything the review found that is not the FTE's to fix, such as a wrong source document or a connection that needs narrowing. Keep it with the FTE's report. Twelve such notes are a year's history of the delegation, readable in twenty minutes by a successor manager.
The note takes two minutes and is what makes the thirty count.
What does done look like?
Done is a review that read the FTE's own account, checked it against the record, found the send-back patterns, labelled the first month's cost honestly, made three decisions, and wrote them down, in thirty minutes. The second review is faster because the patterns are fewer and the note from the first is there to compare against.
A manager who does this monthly has a direct report they understand better than most human ones, because every claim in the review came with its evidence.