No-Surprises Forecasting.
Written and narrated by Serguei Poppeleer.
Reporting tells you what happened. Forecasting tells you what’s about to. The gap between them is where operational control begins.
The previous piece named the four signals worth measuring. This one is the weekly system that turns them into action — because a measured signal nobody acts on is just a quieter report.
Most of what gets called “control” is a rear-view mirror. The report lands, and the slip it describes already happened. You’re not steering. You’re narrating.
A controlled programme catches issues in the room. An operational one sees them forming before they’re issues at all.
The forecasting rhythm — one worked week
This is the system, not a sentiment. It runs the same way every week, on the same clock.
No debate about whether to look. No debate about what moved. No debate about who owns the next action.
Decision latency, verified-vs-claimed dates, scope velocity, plan-to-site distance. Numbers, not narration. Fifteen minutes, same format every week.
Not “what happened?” The whole meeting turns on the forward question. A signal that moved is a forecast, not a fact to record.
Decision latency past its limit forces a call. An unverified date forces a verification before it enters the plan. The trigger is decided in advance — so the forecast acts, instead of being noted.
That is the difference between a forecast and an anxious report: the trigger is wired before the pressure, and someone owns the move.
What happened: the programme held the line through the 2017 GCC embargo with no material slip.
Why that’s remarkable: the embargo emptied ports and blocked materials across an entire region overnight — a textbook surprise.
Why it wasn’t a surprise here: the leading signals were read early enough to act while acting was still cheap. New supply routes were qualified and design adapted to available materials before the disruption reached the critical path. The embargo was a surprise to the region. It was not a surprise to the programme.
The distance between reporting and control is where claims are born.
Two moves that make forecasting real
List what the plan quietly depends on — lead times, rates, approvals, routes — and the condition that would break each one. You cannot forecast against assumptions you have never written down.
Choose, before the pressure, which signal forces a re-plan and who owns the call. A forecast nobody is mandated to act on is just a more anxious report.
The objections this raises
“Forecasting is guesswork — we’d rather report facts.”
A forecast built on leading indicators isn’t a guess; it’s an early reading of facts that haven’t finished happening. The alternative is reporting facts that have already cost you the chance to act on them.
“Our environment is too volatile to forecast.”
Volatility is the argument for forecasting, not against it. You don’t forecast to predict the shock. You forecast so the shock reaches a system already watching the signals it moves first — and can bend before it breaks.
Continuous forecasting was historically too expensive — you couldn’t put a person on every assumption every week, so we settled for the annual review. AI changes the economics. It watches the assumption ledger and the leading signals continuously, and flags the one that moved.
It does not replace the judgement of what to do about it. Standardise the inputs — the 5Cs — then let AI accelerate the watching. Architecture before automation, every time.
Forecasting is the difference between managing the programme and being managed by it.
The line worth keeping
Read this way, the 5Cs are a forecasting instrument. CONFIRM turns a promise into a verified date. CLOSE turns decision latency into a number you can watch climb. CONTROL turns the weekly cadence into the place drift shows up first, not last.
By the time something reaches the report, it is already history.
No-surprises delivery was never luck. It’s reading the leading edge while there’s still time to bend it.
Forecast the programme — or be managed by it.
Could your programme see the next surprise coming?
The Delivery Score shows whether you’re forecasting or narrating — and which signals you’d need to watch to stop being surprised. Fifteen Yes/No questions. About four minutes. Instant result.
Get my Delivery Score — free → Read the CONTROL Special Report →