Supplier scorecard template for comparable performance

A supplier scorecard template should let you explain a result before assigning a grade. Record one supplier site, review period, and metric per row, with the eligible population, numerator, denominator, weight, source, and owner. Keep exceptions and the next action beside the calculation so a favorable average cannot make a disputed delivery disappear.

Copy the table below into Excel or Google Sheets and replace the bracketed cells. The worked example uses delivery, quality, and supplier-document completeness, with declared weights. Those choices illustrate the calculation; they are not a recommended grading policy for every category or evidence of any supplier’s performance.

Copy the template and define its review scope

Use a stable supplier and site identifier, not the display name alone. State the review dates, reporting cutoff, and definition version above the sheet. Retain one row per metric even when its result is unknown; deleting the row would remove evidence that the review is incomplete.

Calculate each rate as numerator divided by denominator, multiplied by one hundred. Add a result column to your working sheet, with an explicit unknown state when evidence is insufficient. Weights must sum to one hundred percent before the full composite is calculated. Infor’s supplier-evaluation documentation distinguishes categories, criteria, and category weights; this template uses directly measured rates instead of its questionnaire rating scale.

The source cell should identify an export, report, or record set you can reopen, with its cutoff. Name the person responsible for correcting the input and put the next action’s due date in the final column. Preserve closed actions in a history tab instead of overwriting the evidence behind an earlier review.

Set delivery, quality, and evidence rules before scoring

Delivery: define on-time, in-full (OTIF) at the PO-line level for this template. An eligible line passes only if its required accepted quantity is available within the agreed arrival window. Specify whether the baseline is the requested date or supplier-confirmed date, the permitted early and late tolerances, the location, and the calendar. Freeze the applicable baseline so a later promise does not rewrite an earlier failure.

Microsoft’s supply-risk configuration separates requested-date acceptance, on-time, in-full, and OTIF rates. It lets customers choose risk thresholds, which is a reason to state your policy rather than copy a software default. The OTIF calculation covers the delivery definition in more depth.

For a partial delivery, aggregate eligible receipts against the line before deciding whether the required quantity arrived within the window. Record physical receipt and inspection-release dates separately if acceptance can lag arrival. A late remainder cannot turn the frozen window into a pass. Include overdue, undelivered lines in the due-line population unless an approved exclusion applies.

Quality: use accepted inspected units divided by all inspected units in the declared population. Keep held or disputed units visible and state whether the result is provisional while disposition is unresolved. Do not describe an inspection sample as the quality rate for every delivered unit without the sampling basis needed to support that conclusion.

Complete evidence: define the supplier-controlled packet and deadline, such as an agreed packing record and lot certificate. Count a packet as complete only when every required element is available by that deadline. This measures a declared supplier obligation; a missing internal upload or broken connector belongs in data coverage, not a supplier failure.

Calculate an illustrative weighted score

Assume a hypothetical supplier site, S-017, reviewed for October with complete underlying data. Eighteen of twenty eligible due PO lines pass OTIF, giving 90%. Quality accepts 980 of 1,000 inspected units, giving 98%; all quality dispositions are final. Sixteen of twenty required supplier packets are complete by the agreed deadline, giving 80%. The remaining packets are confirmed supplier misses, not unknown internal records.

Declare delivery and quality weights of 40% each, with complete evidence weighted at 20%. Weighted score = 90 × 0.40 + 98 × 0.40 + 80 × 0.20 = 36 + 39.2 + 16 = 91.2 out of 100. The percentages measure different populations, so combine their normalized rates, not their raw numerators or denominators.

No pass grade follows from that number without an approved policy. Retain the two missed delivery lines, twenty rejected inspected units, and four incomplete packets for review. A critical quality or compliance failure can require a separate decision gate even when the composite is high; agree that gate before reviewing results.

Keep unknown data and unlike populations visible

If the receiving feed is incomplete, leave the affected metric unknown and show coverage beside it. Do not assign zero, assume a pass, or silently redistribute its weight. Under this template’s policy, withhold the full composite until required metrics are resolved; a partial result must state which weights and evidence it covers.

Keep denominators beside percentages. Two successful deliveries out of two and two hundred out of two hundred both produce a perfect measured rate, but describe different amounts of evidence. Set a review rule for low-volume suppliers and retain the counts when combining periods. Do not average monthly percentages without accounting for their populations.

Compare suppliers within relevant product, site, and delivery-lane groups before interpreting a rank. An emergency-order mix and a scheduled replenishment mix impose different demands. The fill-rate calculation explains another comparison trap: units, complete lines, and complete orders can yield different percentages from the same deliveries.

If you add price variance, compare the billed unit price with the applicable contract price for that item, currency, quantity tier, and effective date. Keep freight, tax, and approved surcharges separate. Price variance is not automatically a higher-is-better percentage, so define a scoring transformation and revise the weights before adding it to the composite.

Keep metric-level detail available even when the meeting starts with a summary. Oracle’s supplier-scorecard documentation separates overall and section trends and lets users drill into evaluation details. For choosing the system that maintains those records, use the supplier management software comparison.

Where Permute fits when a scorecard needs current evidence

Permute sells connected business data, custom applications, and governed context for AI tools. We can scope a supplier review around permitted NetSuite records, Excel assumptions, and supporting documents in Google Drive. Supplier and item identities need agreed matching rules, with source references retained when records are reconciled.

We use declared definitions so the same due-line population and acceptance rules can support repeated reviews. A buyer can ask which records changed a score and inspect the evidence through a scoped application or an AI tool such as Claude or ChatGPT. The template’s static copy (frozen at its export cutoff) needs a new reconciliation when receipts, dispositions, or supplier records change.

Keep grading policy and supplier decisions with their owners

We do not select your weights, approve supplier qualification, or forecast future delivery performance. Verify source access, refresh behavior, and the completeness of required evidence before relying on a connected scorecard. The vendor scorecard software comparison addresses the separate selection decision when a static sheet no longer fits the workflow.

Use the template to preserve each metric’s population, calculation, and source beside its owner and action. Apply declared weights only when the required evidence supports comparable rates. Keep unknowns separate from confirmed failures and review critical exceptions outside the average. A supplier conversation should end with an assigned action and a due date that the next review can check.

Questions about supplier scorecards