Engine tournaments

A buyer evaluating between SAP IBP, Kinaxis, and o9 today runs a year of dueling vendor POCs. Each vendor demos on their own cherry-picked data. Each claims victory. The buyer is left to adjudicate marketing claims that compress the actual operating question — which solver produces the best decisions for our specific planning problem against our specific risk profile? — into a beauty contest.

VYAN offers something no vendor in that bake-off can: a neutral arena. Feed all three solvers the same iteration set built from the buyer's own past variabilities and future risks. Let each solver run. Build the histograms. Compute PVA for each. The buyer sees, on their own data, with identical inputs, which solver actually produces the best decisions for their specific planning problem. The vendors don't get to cherry-pick their data. The buyer brings their own.

The buyer who runs that tournament has, by definition, just experienced VYAN as the layer that made the comparison possible. Whichever solver wins the tournament, VYAN won the buyer — because VYAN was the intelligence layer, the measurement framework, the neutral ground. The buyer might pick Kinaxis as their solver. They'll keep VYAN as the System of Intelligence above it. Over time, as the business changes and PVA telemetry updates, VYAN tells them when the answer changes, and they switch again — with VYAN unmoved as the layer above.

This is a positioning the planning-stack vendors structurally cannot answer. o9 cannot offer a neutral tournament because o9 is a contestant. Only a System of Intelligence that sits above all solvers and has no stake in which one wins can run the arena. It's structurally exclusive to VYAN's position.

↓ Worked-example tournament numbers — specific PVA outputs from a three-way tournament between named solvers on a representative industrial dataset — unlock at gate 4.