[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"article-body-the-plan-is-a-probability-distribution":3},"\nEvery plan of consequence in your company is a single number.\nRevenue will land at 104, the transformation will cost 12.5, go-live\nis the first of October. Everyone experienced in the room believes\ntwo things at once. The number is almost certainly wrong, and the\nyear will be managed against it anyway.\n\nThat is not a failure of skill; it is a failure of form. The future\nhas many ways of arriving, and a form that permits exactly one\nnumber forces everyone to pretend otherwise, then spend the year\nexplaining variances against a fiction. The honest form of a plan is\na distribution. It shows the range of outcomes, each carrying odds,\nupdated as evidence arrives.\n\n## Odds, not lists\n\nMost planning teams believe they already handle uncertainty, because\nthey produce scenarios. Base, upside and downside make three\nspreadsheets and three stories. But when the downside case carries\nno odds, nobody can say whether it deserves a contingency budget or\na shrug, so the decision defaults to temperament. Whoever is most\nsenior settles the argument, and evidence never enters it.\n\nAttach odds and the conversation changes. A 12% chance of missing\nthe covenant is an instruction. A \"downside scenario\" is a mood.\n\n## The meeting, restructured\n\nThe deepest change is in what the monthly meeting argues about.\nAgainst a point plan, the meeting litigates whose number is right,\nwhich cannot be resolved and therefore never ends. Against a\ndistribution, nobody's number is right by construction, so the\nargument moves to thresholds and interventions, where it can be won.\nAt what probability of missing the year do we act? Which move shifts\nthe odds most per unit of cost? Those questions have answers, and\nthe answers can be scored afterwards.\n\nThis is also the only structure in which\n[success stops being a surprise](/insights/success-should-never-be-a-surprise).\nOdds that update daily cannot ambush you in November, because the\ndrift was visible in March.\n\n## Why the market stopped short\n\n\"Probabilistic\" is everywhere in enterprise software now, so look at\nwhere the word points. It almost always describes model behaviour,\noutputs to be bounded and governed. Anaplan's\n[agentic enterprise positioning](https://www.anaplan.com/news/anaplan-introduces-the-agentic-enterprise/)\npairs deterministic calculation with probabilistic AI as a trust\nstory. Nobody publishes a distribution over plan outcomes; nobody\nstates the odds that the year lands, updates them, and scores them\nagainst what happened.\n\nThe absence is not technical. Stated odds can later be\n[checked for calibration](/insights/of-everything-called-70-percent),\nand that check is public and unforgiving. A vendor who says\n\"probabilistic\" commits to nothing. A vendor who states the odds can\nbe proven wrong.\n\nHorizon is built on the second posture. Plans carry distributions,\ncommitments carry rolling odds updated daily, and the odds are\nscored against outcomes, so their calibration is itself on the\nrecord. That is the moat; a marketing team cannot retrofit checkable\nodds. To see your year as a distribution, [start here](/contact).\n",1786984937589]