Why static city-pair pricing leaves money on the runway
Category-leading marketplaces ship one price per city pair, refreshed weekly. The clearing price moves hour by hour. Here's the gap, and what we did about it.
DEMO · figures referenced in this post are illustrative until Round 3 wires live BigQuery exports.
The clearing price for a one-way charter is not the city-pair median. It’s the price at which a specific tail, on a specific morning, with a specific weather window, will accept the leg. The two diverge by tens of percent.
This is a placeholder Field Note. The Round 3 editorial pass replaces this paragraph with the actual analysis: what we measured, how we measured it, what the model returned, and where we were wrong.
What we tracked
For Q1 2026, we instrumented every quote-to-accept transition in the Avinode federated supply we have visibility into, and compared the accepted clearing price against the platform’s published median for the same city pair on the same week. The variance distribution is wider than category narratives suggest.
Sample numbers will be re-published from a real BigQuery export in Round 3 — the figures shown today are illustrative.
What changed
The Vertex Price model now produces a triplet — floor, list, settle — for every quote, refreshed at quote time, not weekly. The settle column is what reconciles to the wallet at wheels-up.
Operators see the same triplet. So do brokers, with operator authorization.
What’s next
The next Field Note covers the empty-leg sell-through curve and what changed when we shipped the SmartRoute v2 reinforcement-learning policy update.