Turns 'they are always late' into what it cost you
Every operations team knows which supplier is the problem. What they lack is evidence, and without it the quarterly review is two people exchanging impressions while the supplier's account manager wins on confidence.
The data exists — receipts, delivery notes, quality rejections, invoices — but it sits across systems and nobody assembles it, so the conversation runs on anecdote. Anecdote is biased toward whatever went wrong most recently rather than what costs most.
The number that changes the conversation is not on-time delivery. It is what the failures cost: the line that stopped, the expedited freight to cover a late delivery, the rework on rejected material, the second order placed because the first was short. A supplier who is cheapest per unit and causes three line stops a quarter is the most expensive supplier you have, and no price comparison will ever show it.
This agent assembles that picture per supplier, prices the failures where the records allow, tracks price drift against the agreed rate, and separately reports concentration — which suppliers you could not replace quickly. That last one is invisible until the day it matters, and on that day it is the only thing that does.
It contacts no supplier and ends no relationship. It gives you the evidence to hold a real review.
Rather than impressions against a confident account manager.
Which frequently reverses the ranking price alone produces.
While there is still time to qualify an alternative.
Expedited freight, rework, a stopped line. The cheapest supplier per unit is often the most expensive one you have.
A delivery that arrived on the day and short is not on time in any sense that helps you.
Small increases that never triggered a review because each one was small.
Which suppliers you could not replace quickly. Invisible until the day it is the only thing that matters.
Every finding traces to receipts and records, so the review stops running on anecdote.
Supplier relationships are commercial and long-term. It arms the conversation; it does not have it.
Why Procurement in particular. On-time delivery is the wrong number. What matters is what the failures cost — expediting, rework, a stopped line — and that routinely reverses the ranking that unit price produces.
Runs unattended
Started by you or by an event, and it finishes on its own. Nothing waits for someone to be at a desk.
The same standard every time
The two-hundredth item is held to the bar the first one was. Consistency is the part people cannot sustain.
It cannot act on its own
Supplier Performance Agent has no path to sending, spending or committing. That limit is why its output is safe to act on.
This agent runs server-side through the PROMIVO runtime. Each run is logged step by step and every tool call is permission-checked before it executes.
Read-only by design. This agent has no path to sending, spending, publishing or committing anything. Where that limit is the product, removing it would remove the reason to trust the output.
Demo dataIllustrative sample output, abridged.
{
"orders": [],
"period": "Q3 2026",
"currency": "USD",
"concentrationThresholdPercent": 40
}{
"escalate": true,
"suppliers": [
{
"spend": 480000,
"evidence": [
"14 short deliveries against PO-4412 through PO-4680.",
"Three expedited freight charges totalling 21,400 raised to cover them."
],
"supplier": "Supplier A",
"failureCost": 96000,
"inFullPercent": 74,
"onTimePercent": 91,
"rejectionRate": 2.1,
"totalCostRank": 3,
"priceDriftPercent": 11.4,
"failureCostQuantified": true,
"onTimeAndInFullPercent": 69
}
],
"disclaimer": "An analysis of your own records. No supplier has been contacted, no order changed, no termination recommended, and failure costs are only reported where a record supports them — the rest is reported as frequency.",
"concentration": [
{
"category": "Machined housings",
"supplier": "Supplier B",
"whatStops": "Assembly stops within 9 days of a supply failure. No second source is qualified.",
"sharePercent": 78,
"alternativeQualified": false
}
],
"performingWell": [
"Supplier C: 98% on time and in full, no rejections, no price drift. Worth more volume, and nobody has ever said so in a review."
],
"couldNotQuantify": [
"Line stoppage cost for 6 events — the duration was recorded but no cost rate exists, so frequency is reported without a figure."
],
"escalationReason": "Supplier A's failures cost 96,000 against 480,000 of spend, and Supplier B holds 78% of a critical category with no qualified alternative."
}No integrations required.
The numbers and the evidence, prepared.
Performance priced, not described.
Which to keep on total cost rather than unit price.
Where a single failure would stop you.
$349/month
Billed monthly through your PROMIVO subscription. Cancel at any time.
Runs consume your plan allowance for agent executions and tokens. See plan limits.
No. It contacts nobody and recommends no termination. Switching supplier carries qualification, tooling and continuity risk this cannot see.
Then it reports frequency and says the cost could not be quantified, rather than estimating. An invented cost figure would be the most quotable and least defensible thing in the report.
It ranks them on measured total cost where the data supports it, and says where it does not. A single score across incomparable categories would be decoration.
No reviews yet. Reviews open once customers have run this agent.
Tell us what to change and our team will scope a customised version for your business.
Customize this agent