A continuous discipline of executive foresight.
Predictive planning is not a software category and not a forecasting upgrade. It is a leadership discipline — a five-principle, four-phase loop that converts a stream of weak signals into sized strategic commitments, then steers those commitments as the operating environment changes underneath them.
Anchors that hold across every phase of the loop.
The phases describe what the team does. The principles describe how the team does it. Every revolution of the loop is held to all five.
- 01Continuous beats episodic
Quarterly planning was built for an operating environment that no longer exists. The discipline runs at the cadence of the world, not the cadence of the calendar.
- 02Decisions, not decks
The output of the loop is a commitment, not a presentation. A team that cannot move from scenarios to a stake decision is not yet running the discipline.
- 03Range, not point
Every forecast is a range. Every commitment is sized to the range, not to the midpoint. Confidence intervals belong on the front page, not in the appendix.
- 04Judgment over machine
Predictive analytics is a layer of the discipline, not the discipline itself. Tacit knowledge held by senior operators is a first-class input, not noise.
- 05Reversibility is currency
The cost of being wrong is the dimension that distinguishes a Probe from a Bet. A firm that prices reversibility into every commitment carries less structural risk than one that doesn't.
The Lattice Spine
Sources on the left, decisions on the right, the lattice in the middle — the single math contract every signal class shares. Every layer above the spine reads from it; every layer below feeds it.
The diagnostic question.
Every Stake decision passes through one filter before it leaves the room.
If we are wrong about this commitment, is the cost of being wrong symmetric or asymmetric?
Symmetric losses are recoverable. The team learns, re-bases, and the firm keeps its operating posture. Asymmetric losses are structural — they remove options the firm cannot rebuild on a second attempt. The discipline does not refuse asymmetric bets. It refuses to size them as if they were symmetric ones.
A commitment that fails this test is not killed. It is resized, hedged, or staged so the cost of being wrong stops being a one-way door. Reversibility is currency. The asymmetry test is how a Stake meeting prices it.
Three categories the discipline is consistently confused with.
Forecasting produces a number. Predictive planning produces a commitment. A team that asks the model for next quarter's point estimate has already failed the range principle. The forecast is an input — what the team does with the range is the discipline.
Scenario planning was the right idea for a slower operating environment. It produced three named worlds and stopped there. Predictive planning runs the same scenario logic on a continuous cadence, ties every scenario to an indicator set, and forces a sized commitment at the end of every revolution.
Predictive analytics is a math problem. Predictive planning is a judgment problem with math as an input. Confusing the two is how organizations end up with sophisticated dashboards and static strategies — instrumented, observant, and unable to move.
The discipline names a vocation. The Certified Predictionist credential operationalizes it.
Read the full discipline →