[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"article-body-supplier-failure-has-a-lead-time":3},"\nAsk a category manager about their last serious supplier failure\nand you will hear a story with a sudden ending. The letter\narrived, the insolvency was announced, the shipment simply did not\ncome. Then ask what the data looked like for the six months\nbefore. Deliveries had been slipping by growing margins. Quality\nrejects were trending up. Invoices were being chased harder. Two\nnamed contacts had left in a quarter. The failure was announced\nsuddenly; it happened slowly, in public, in the buyer's own data.\n\nThat gap between observable and official is the entire economics\nof supplier continuity, and the standard tooling is not built to\nprice it.\n\n## Scores answer the wrong question\n\nThe supplier risk market sells scores. Each supplier gets a\ncomposite number, refreshed periodically, that says how risky.\nScores answer a static question, and continuity is a timing\nproblem. You do not need to hear that a supplier is amber. You\nneed to hear that at the current rate of deterioration you have\nabout ten weeks, and what that estimate is based on.\n\nInterventions are priced in time. With twelve weeks of warning,\nqualifying a second source is a project. With two weeks, it is air\nfreight and overtime. With none, it is a stopped line and a\ncommercial apology. The same failure, caught at different\nhorizons, differs in cost by an order of magnitude. The detection\nhorizon is the product.\n\nTo our knowledge, no published study quantifies supplier-failure\nlead time across signal types. That silence says something about a\nmarket that has sold detection for a decade; the vendors hold the\ndata and have not shown the horizon.\n\n*Buy lead time, not scores.*\n\n## Reading the telegraph\n\nThe signals mostly sit in your own systems already. Delivery\nperformance decays against its own history, quality drifts by part\nfamily, order acknowledgements stretch, and payment behaviour\nchanges where you can see it. Ownership and management churn gets\n[its own compliance urgency from the fifty percent rule](/insights/the-fifty-percent-rule-countdown).\nAnd the failure often starts one tier down, among\n[your supplier's own suppliers](/insights/your-suppliers-supplier-is-your-problem).\n\nEach signal alone is noisy. Together, on one timeline per\nsupplier, they turn the vague question of risk into a concrete\none. How long until this becomes your problem, with what\nconfidence, and which signals drive the estimate?\n\n## What changes when timing is the product\n\nA stated detection horizon can be backtested against the failures\nthat actually happened, so \"we would have seen it eight weeks out\"\nbecomes checkable rather than atmospheric. Precision matters,\nbecause false alarms burn the category team's attention and real\nmisses burn the line. And every warning needs an owner and a\ndecision attached, or the countdown is just another dashboard\nwidget.\n\nProphesee treats supplier failure as exactly this problem. Timing\nmodels run on the supplier graph, backtested against history, and\neach warning routes as an exception with an owner, a horizon and\nthe intervention economics attached. The backtested horizon is the\nmoat, because it cannot be marketed into existence.\n[Start here](/contact).\n",1786984937235]