You are funding somebody else’s promise
The MRO manufacturer sits behind a commitment they did not write. A distributor wins a national account by guaranteeing a savings number; the support that makes that number achievable comes out of your programme. The commitment is measured on a basket, and the basket moves — substitution, private label, changing consumption — while your funding stays priced against what it looked like at signature.
Nobody is being dishonest. It is simply that a catalogue of hundreds of thousands of items cannot be governed by attention, and attention is the only control most programmes have.
The long tail is where drift lives
The top hundred items get reviewed. The remaining tens of thousands do not, and that is precisely where prices go stale, where cross-references rot, and where the difference between the contracted price and the transacted price accumulates quietly for a year until an annual review finds it.
Detecting drift inside the period it occurs is the whole game, and it is only possible if contract prices carry effective dates and item identity is resolved across every coding scheme in play.
What good looks like
- Above 99% of lines with a verifiable live price on the date they transacted.
- Item identity resolved above 97% — below that, validation covers the head and misses the tail, which is the opposite of what is needed here.
- Drift detected within 30 days, not at the annual review.
How RevUpra runs this
Contract price lists are held with enforced effective dating, so what was live on a given day is a query rather than an archaeology exercise. Item identity is cross-referenced across your part numbers, distributor SKUs and customer codes, which is what makes long-tail coverage possible. Claims are validated against reported end-customer volume at the contract’s own basis, drift is surfaced continuously rather than annually, and programme profitability by national account is a materialised read available days after period end.