Supplier evaluation software scores each supplier at regular intervals on criteria such as on time delivery, quality and price compliance, and turns the result into grades like A, B and C. The short answer: the score should come from records your system already holds, not from a buyer's gut feeling. Delivery dates come from goods receipts, reject rates from incoming inspection and price variance from invoices; the software combines the three and lets data decide who stays on the approved supplier list.
Why is supplier evaluation more than filling in a form?
In many companies supplier evaluation is a form completed once a year: the purchasing lead gives every supplier a mark from 1 to 5, the form is filed and shown at the audit. The problem is that the mark relies on memory. Last month's late delivery is remembered, three defective batches from eight months ago are forgotten, and different people rate the same supplier differently. The evaluation stops being a decision tool and becomes paperwork.
A good supplier score needs no extra data entry. If you already record orders, receipts, inspections and invoices, the score is simply a summary of those records.
Which criteria should supplier evaluation software cover?
- On time delivery rate: deliveries that arrived on or before the promised date divided by all deliveries; partial deliveries count as separate lines.
- Quantity accuracy: the difference between the ordered and delivered quantity.
- Quality: rejected units against inspected units at incoming inspection, with critical defects flagged separately.
- Price compliance: whether the unit price on the invoice matches the price agreed on the purchase order.
- Documents and compliance: expiry dates of certificates, test reports or contracts.
- Weights: the share of each criterion in the total score is configurable, and every change is kept in a history.
- Grades and approved list: A, B or C by score, with an action or removal decision for C suppliers.
How is a supplier score calculated?
Each criterion is first converted to a value between 0 and 100, then multiplied by its weight and added up. You might give delivery and quality 40 percent each and price 20 percent; those numbers are not a recommendation, they are a setting that reflects your priorities. Grade thresholds are settings too. The example below calculates one supplier's score from the records of a period:
type Delivery = { supplierId: string; promisedDate: string; deliveredDate: string };
type Inspection = { supplierId: string; inspectedQty: number; rejectedQty: number };
type PriceLine = { supplierId: string; agreedKurus: number; invoicedKurus: number };
const WEIGHTS = { delivery: 0.4, quality: 0.4, price: 0.2 };
function scoreSupplier(id: string, d: Delivery[], q: Inspection[], p: PriceLine[]) {
const mine = d.filter((x) => x.supplierId === id);
const onTime = mine.filter((x) => x.deliveredDate <= x.promisedDate).length;
const deliveryScore = mine.length ? (onTime / mine.length) * 100 : null;
const insp = q.filter((x) => x.supplierId === id);
const inspected = insp.reduce((s, x) => s + x.inspectedQty, 0);
const rejected = insp.reduce((s, x) => s + x.rejectedQty, 0);
const qualityScore = inspected ? (1 - rejected / inspected) * 100 : null;
const lines = p.filter((x) => x.supplierId === id);
const matching = lines.filter((x) => x.invoicedKurus <= x.agreedKurus).length;
const priceScore = lines.length ? (matching / lines.length) * 100 : null;
if (deliveryScore === null || qualityScore === null || priceScore === null) {
return { id, result: 'insufficient-data' };
}
const total =
deliveryScore * WEIGHTS.delivery + qualityScore * WEIGHTS.quality + priceScore * WEIGHTS.price;
const grade = total >= 85 ? 'A' : total >= 70 ? 'B' : 'C';
return { id, total: Math.round(total * 10) / 10, grade };
}Two rules matter here. First, a criterion with no records does not count as zero; the result stays "insufficient data", because pushing a supplier that delivered nothing into grade C is simply wrong. Second, amounts are stored as integer minor units. When an evaluation period closes, the score is saved together with the weights used, so the result can be reproduced later even if the weights change.
How does it connect to purchasing, quality control and CAPA?
The raw data for the evaluation is born in other modules. Order dates and agreed prices live in your purchasing software, and rejected quantities come from the incoming inspection record in your quality control software. The supplier code must be shared between these systems; otherwise the same company ends up with two names and two separate scores. When a score drops or a critical defect appears, a corrective action is opened for the supplier and followed in your nonconformance and CAPA software; the supplier's response and closing date then show up in the next evaluation.
Off the shelf or custom software?
Most ERP and quality management suites include a basic supplier evaluation screen. If your purchasing, inspection and accounting records all live in that suite, the screen is often enough. If the data is spread across different systems, if you have industry specific criteria (cold chain temperature logs or sample approval time, for example) or if you want a portal where suppliers see their own scores and open actions, custom development fits better. Before deciding, list which system holds the past year of delivery, inspection and invoice records and under which supplier code; that alone shows how big the job really is.
Conclusion
Good supplier evaluation software calculates scores from existing records, never treats missing data as a low score, keeps weights and past results, links poor scores to corrective actions and updates the approved supplier list based on data. To build this on top of your purchasing and quality processes, see our corporate solutions page or describe your needs and request a quote.