The system lists what slipped. The person says amber.
One in construction, one in IT rollout, one running an internal transformation programme. All three describe the job the same way when pressed: "I chase people and then I write down that I chased them."
Friday afternoon, mostly. The status report, assembled by hand from four systems and six conversations, for an audience that will read the RAG colour and nothing else. Plus the daily version: meetings that generate actions, actions that live in minutes, minutes nobody rereads, and therefore actions nobody does — discovered three weeks later at the worst moment.
- Turning meeting recordings or notes into a structured action list with owner and date
- Chasing those owners on schedule
- Detecting what has moved since the last report — the diff, not the state
- Assembling the status report from the plan, the tracker and the risk log
- Keeping the RAID log current and surfacing risks whose review date has passed
- Watching dependencies: if A slipped, listing everything downstream of A
- Producing the audience-appropriate version — the same week told three ways
- Whether the project is actually red
- What to escalate and when
- Which risk is real
- Who needs to be told what, and in what order
- The meeting-to-actions pass
- Recording or notes in, action list out: owner, deadline, dependency, open question, filed to the tracker. This closes the single biggest leak in project management, which is that actions decay between the room and the system.
- The chaser
- Owners get a message before their date, not after it. Drafted in your own register and sent by you, so it stays a human asking a human. All three described chasing as the most draining part of the week, precisely because it is repetitive and slightly awkward.
- The diff engine
- The status report is not a summary of the project; it is a summary of what changed. Machines are good at diffs and humans are bad at them. It produces: tasks moved, dates slipped, risks opened or closed, actions overdue. You write the two sentences that say what it means.
- The dependency tracer
- When something slips, an immediate list of what is downstream. Doing this by hand in a large plan takes an hour and therefore does not get done at the moment it matters.
- The three audiences
- One factual base, three registers — steering committee, delivery team, client. Same facts. Nothing invented for any of them.
It does not set the RAG status and it does not decide what to escalate.
The colour on a status report is a political and professional act. A machine that sets it either becomes a rubber stamp you override constantly — in which case it is noise — or it becomes something you defer to, which is worse.
It also never sends upward. Reports to a steering committee or client are drafted and always leave under a human signature.
Actions from meetings that land in the tracker: from about 60% to near all. Hours to produce the weekly report — typically four down to under one. Actions overdue without anyone noticing. Time from a slip to knowing its downstream consequences.
When a programme is failing for reasons reporting cannot touch — no mandate, no resources, a sponsor who will not decide. Better reporting there documents a failure more clearly, and occasionally people want exactly that, but they should want it knowingly.
Recording meetings you already attend, and two sentences of interpretation a week.
My job description could have been “I chase people and then I write down that I chased them.” Actions from meetings now land in the tracker instead of dying in the minutes, which was the single biggest leak I had. The status report is built from what changed rather than from the whole plan, so it takes me under an hour and I spend that hour on the two sentences that say what it means. It does not set the RAG colour, and it should not.
The dependency tracer is the one I would keep if I could keep one thing. When something slips I get the downstream list immediately, and doing that by hand took an hour so it never happened at the moment it mattered. Three audiences from one factual base also stopped me writing the same week three times on a Friday afternoon. Nothing goes to the steering committee without me sending it, which was a condition of mine and turned out to be a condition of his as well.
Tell me what your week looks like.
I'll tell you honestly which parts you can hand over — and which you can't.