Supplier scorecard metrics that matter most are on-time in-full delivery, defect rate in parts per million, lead-time consistency, total cost to serve, corrective action closure speed, responsiveness, and compliance status. Seven measures, each with an agreed formula and a data source you already own, cover almost everything a manufacturing buyer needs to judge a supplier.
The rest is discipline. A scorecard with thirty rows gets ignored; a scorecard with seven rows and one named decision attached to each one gets used in the quarterly review and changes how a supplier behaves.
Last reviewed October 2026.
Table of Contents
- What Are Supplier Scorecard Metrics?
- Supplier Scorecard Metrics That Matter Most
- Quality and Defect Performance
- Delivery, Lead Time and Supply Reliability
- Cost and Total Cost of Ownership
- Responsiveness, Communication and Service
- Compliance, Sustainability and Risk
- How to Design a Supplier Scorecard
- Supplier Scorecard Metrics That Matter by Supplier Type
- How to Use Scorecard Results for Improvement
- Frequently Asked Questions
- Conclusion
What Are Supplier Scorecard Metrics?
A supplier scorecard metric is a number that rates one dimension of supplier performance over a fixed period. Every metric worth putting on a scorecard has four parts, and if one is missing it is not a metric yet, it is an opinion with a number attached.
The four parts of a real metric
- Definition — one sentence everyone signs off on, written before any data is pulled. “On time” means against the supplier’s own acknowledged promise date, not your requested date.
- Formula — the arithmetic, agreed in writing.
- Data source — the system of record, usually purchase order lines, the receiving log, incoming inspection records or the corrective action log.
- Attached decision — what someone will actually do differently because this number moved. Award, corrective action plan, second-source qualification, exit.
If no decision is attached, cut the row. It is data you are collecting for no reason.
Leading and lagging indicators behave differently
Defect rate and late delivery are lagging indicators. They tell you what already happened, usually after the parts have shipped. Process capability, first-pass yield and corrective action closure time are leading indicators: they move first and give you a chance to intervene before a customer shipment slips.
Use lagging metrics for the score and for supplier conversations. Use leading metrics for the improvement plan, because that is where the corrective work actually happens.
Supplier Scorecard Metrics That Matter Most
These seven cover quality, delivery, cost, service and risk. The formula column is what you put in the scorecard header so no one has to guess how a number was produced.
| Metric | Exact formula | Data source | Decision it drives |
|---|---|---|---|
| On-time in-full (OTIF) | Order lines received complete on the promised date ÷ total order lines due | ERP purchase order lines plus receiving log | Delivery corrective plan, second-source qualification |
| Defect rate in PPM | Defective pieces received ÷ pieces received × 1,000,000 | Incoming inspection records | Containment, CAR, quality improvement plan |
| Lead-time variability | Standard deviation of actual minus quoted lead time, in days | ERP receipt dates against PO dates | Safety stock policy, forecast sharing, planning trust |
| Corrective action closure | Days from CAR issue to verified closure, and percentage closed within due date | Corrective action log | Whether the supplier gets another award |
| Cost to serve | Expedite, rework, extra freight, inspection and admin cost per unit | ERP, freight invoices, receiving log, inspection hours | Negotiation, volume allocation, exit |
| Responsiveness | Percentage of RFx responses, PO acknowledgments and issue responses inside agreed times | Sourcing platform, ERP email trail, buyer log | New-business allocation, award decisions |
| Compliance status | Percentage of required certifications current, plus open audit findings past due | Supplier document portal, audit records | Continuity of supply, compliance hold |
There are no universal targets. A supplier making industrial castings and one supplying printed marketing inserts should not share a defect target, and any scorecard that hands you a number like “95 percent is good” without context is selling something.
Quality and Defect Performance
Measure defects in parts per million rather than as a raw percentage, because percentages flatter small suppliers and punish large ones for the same absolute number of escapes.
Defect PPM with a worked example
A supplier ships 48,000 pieces in a quarter and your inspectors reject 96 of them.
96 ÷ 48,000 = 0.002, and 0.002 × 1,000,000 = 2,000 PPM. Now a second supplier ships 9,000 pieces with 15 rejects: 15 ÷ 9,000 × 1,000,000 = 1,667 PPM. As percentages, 0.2 percent and 0.17 percent look nearly identical. In PPM, one supplier is a fifth worse, and the gap widens as volumes change.
When a low defect rate is bad news
Track your inspection coverage alongside the PPM number. A supplier at 200 PPM inspected on every lot is a different situation from a supplier at 200 PPM whose parts bypass incoming inspection entirely because receiving has been told to trust them. Falling inspection levels show up as an improving defect rate long before they show up as a customer complaint.
Add two more quality rows once the basics are stable: first-pass yield from the supplier’s own process, and process capability against the print, and the percentage of corrective action requests closed by the agreed date.
Delivery, Lead Time and Supply Reliability
On-time delivery is the most quoted supplier metric and the least informative one on its own. It tells you whether a shipment arrived on the date. It says nothing about whether the shipment was complete.
OTIF is stricter than on-time delivery
Take 100 order lines due in a quarter. 95 arrive on or before the promise date, so on-time delivery is 95 percent. Of those 95, ten were short, so in-full is 85 out of 100 lines, or 85 percent. OTIF counts only lines that arrived complete and on time: 85 out of 100, so 85 percent.
OTIF is 85, not 95 × 85 = 80.75. The two results differ because they are counting different populations, and you cannot multiply them because the short shipments and the late shipments are not the same events. A supplier can hit 95 percent on-time delivery with partial shipments all quarter and still have an 85 percent OTIF, which is the number your production schedule actually cares about.
Lead-time variability beats lead-time average
A supplier quoting ten days that delivers between six and twenty-two is harder to plan around than one quoting twelve days that delivers eleven to thirteen. Record the standard deviation of actual minus quoted lead time and track it as its own metric.
Keep an eye on schedule attainment, the percentage of order lines delivered at or before the date the supplier committed to in writing, plus backorder and cancellation counts and the premium freight spend you incurred because of supplier misses.
Cost and Total Cost of Ownership
Unit price is the least useful row on most scorecards because it is the only cost the supplier controls completely. Everything else on this section is cost you absorbed.
Build a cost-to-serve line per shipment: premium freight after a missed date, labor and scrap from defects that escaped inspection, inspection hours above the normal sampling plan, warranty and returns attributable to the part, and buyer administrative time spent chasing missing documents and credit notes.
Track price variance against the agreed contract price or the awarded benchmark, and track invoice accuracy as the percentage of invoices matching purchase order terms on the first pass. Then divide total cost to serve by units purchased and compare it against a second source. A supplier 3 percent cheaper on paper can be well behind once its expedites and inspections are counted, and that gap is the number worth bringing to a negotiation.
Responsiveness, Communication and Service
Service metrics are the cheapest to collect and the most predictive of trouble, because they show behavior before delivery and quality deteriorate.
Count RFx response rate, the percentage of quotes returned inside your stated deadline, and PO acknowledgment, the percentage of orders confirmed within one business day. Then count non-conformance response time, how fast someone answers when you raise a problem, and documentation quality, the percentage of shipments arriving complete with certificates of conformance, material data sheets and test reports attached.
Add time to resolution and how many issues the supplier reopened after being marked closed. A supplier that answers an email quickly but reopens the same issue three times has not saved you anything, and resolution time on first contact is the row that catches it.
Compliance, Sustainability and Risk
For most manufacturers this section starts and ends with certification status: which required certifications are current, which expire within 90 days, and which audit findings are still open past their agreed closure date.
Regulatory documentation is worth its own line if you run under ISO 9001 or FDA Q7A expectations, because a clean twelve-month trail of inspection results and closed corrective actions is audit evidence you already produced for a supplier review.
Then cover material restrictions such as RoHS or REACH declarations, business continuity and disaster recovery plans with a tested recovery time, and any data or cybersecurity controls that apply where supplier systems touch your drawings or forecasts. Finish with the things no supplier controls but you should watch anyway: financial health, single-source exposure for your top parts, and geographic concentration risk when several of your suppliers sit in one region.
How to Design a Supplier Scorecard
Step one: pick five to eight measures
Start with one quality row, one delivery row, one cost row and one service row. Add compliance before you add anything optional. If a measure does not pass the four-part test above, leave it off.
Step two: write the definitions down
One page, signed by your buyer, your quality engineer and the supplier’s account contact. This is the step that prevents the two most common arguments: whose date counts, and what counts as a reject.
Step three: scale every metric to 0-100
Set a floor and a target for each measure, then interpolate between them. For OTIF: floor 80 percent scores 0, target 98 percent scores 100, and 90 percent lands at about 56. Anything at or below the floor scores 0, anything at or above target scores 100.
Step four: weight and multiply
Set weights by business risk, then compute the composite score. Use this example structure and adjust it to your category.
| Category | Weight | Example raw result | Scored 0-100 | Weighted points |
|---|---|---|---|---|
| Quality (defect PPM) | 35% | 1,000 PPM | 80 | 28.0 |
| Delivery (OTIF) | 30% | 96 percent | 90 | 27.0 |
| Cost to serve | 20% | 1.2 percent of unit price | 70 | 14.0 |
| Service (CAR closure) | 15% | 85 percent on time | 60 | 9.0 |
| Composite score | 100% | — | — | 78.0 |
0.35 × 80 = 28.0, 0.30 × 90 = 27.0, 0.20 × 70 = 14.0, and 0.15 × 60 = 9.0, giving a composite of 78. Set your bands: 90 and above is preferred supplier, 75 to 89 is acceptable, below 75 triggers a written improvement plan.
Step five: run it on a real cadence
Pull the data monthly from the ERP and receiving log so the numbers are ready before the meeting. Review the scorecard monthly with the buyer and quality engineer, run the full scorecard with the supplier quarterly, and reassess certifications, business continuity and financial health annually.
Pilot it on three to five strategic suppliers first. Once you have two quarters of history, you can show people a trend instead of arguing about a single number.
Supplier Scorecard Metrics That Matter by Supplier Type
The formula stays the same across suppliers. The weight changes, because the failure that hurts you most depends on what the supplier makes.
| Supplier type | Weight quality | Weight delivery | Weight cost | Watch most |
|---|---|---|---|---|
| Component manufacturer (machined, cast, molded) | 35-40% | 25-30% | 15-20% | PPM, process capability, Cpk trends |
| Packaging supplier | 20% | 40% | 20% | OTIF, print consistency, change control |
| Contract manufacturer | 30% | 30% | 15% | Schedule attainment, first-pass yield, flexibility on volume swings |
| Kitting and fulfillment provider | 20% | 35% | 15% | Pick accuracy, order cycle time, response to demand spikes |
| Logistics partner | 10% | 45% | 25% | On-time pickup, damage claims, track-and-trace data completeness |
Component suppliers earn the heaviest quality weighting because a defect found at incoming inspection becomes scrap, rework and a missed customer date. Logistics partners barely get a quality score at all, since their quality failure mode is a damaged carton rather than a nonconforming part.
How to Use Scorecard Results for Improvement
The score is a conversation starter, not a verdict. The teams that get real improvement out of it follow a few habits.
Show the order count next to every percentage. A supplier with 94 percent OTIF across 300 lines is a different story from one with 94 percent across four lines, and anyone who ranks on percentages alone will eventually be embarrassed by it.
Compare trends, not snapshots. Four quarters of movement tell you whether a supplier is stabilizing or sliding. When you do compare suppliers, use medians by category rather than company-wide averages, since one large plant’s misses can hide an entire supplier’s good record.
Agree escalation thresholds before you need them. Write down what happens at each band: at 75 and below a written improvement plan with dated actions, two quarters below 70 and no improvement means no new business, and single-source parts get a second-source qualification project opened at the same point.
Give improvement a public score. Suppliers notice when a supplier of the quarter gets named in the review, and that recognition costs nothing.
Keep the review human. A quarterly business review works when the buyer, the quality engineer and the supplier’s own account contact are in the same room, because most scorecard rows have an explanation behind the number that the spreadsheet cannot hold.
Frequently Asked Questions
What are the 5 key supplier evaluation criteria?
The five criteria most supplier scorecards are built on are quality, delivery reliability, cost and cost to serve, responsiveness and communication, and compliance and risk. Quality covers defect PPM and first-pass yield. Delivery covers OTIF and lead-time consistency. Cost adds expedites, rework and freight to the unit price. Responsiveness covers quotation and acknowledgment speed. Compliance covers certification status and audit findings. Weight them by what your category actually fails at.
What are 5 examples of metrics to measure performance?
Five examples used on most manufacturing supplier scorecards are on-time in-full delivery, defect rate in parts per million, lead-time variability, corrective action closure time, and cost to serve per unit. Each has a defined formula and a system of record behind it, so two people computing them from the same data get the same answer. A sixth worth adding is PO acknowledgment rate, which costs nothing to collect and catches slow suppliers early.
What are KPIs for suppliers?
Supplier KPIs are the measurable numbers used to rate a supplier’s quality, delivery, cost behavior, service and risk over a fixed period. A working KPI has four parts: a written definition, a formula, a named data source, and a decision attached to it. Most manufacturers start with five to eight and review them monthly in-house and quarterly with the supplier.
What are the 5 key performance indicators in logistics?
For a supplier acting as a logistics partner, the five that matter are on-time pickup, on-time in-full delivery to your dock, order cycle time, damage and shortage claims per thousand shipments, and cost per shipment including fuel and accessorial charges. Add track-and-trace completeness if you ship to customers directly, since incomplete data shows up as your own customer service cost rather than the carrier’s.
How many KPIs should a supplier scorecard have?
Five to eight. Under five, you miss a dimension that can hurt production, most often cost to serve or compliance. Over about ten, reviewers stop reading and the scorecard becomes a report nobody opens. If you want more, split it into a short scorecard for the quarterly review and a longer appendix of diagnostic metrics such as process capability and cycle-time breakdown.
Conclusion
Start with seven rows: OTIF, defect PPM, lead-time variability, corrective action closure, cost to serve, responsiveness and compliance status. Write the definition and the formula next to each one, set your own floors and targets rather than borrowing someone else’s, and pull the data monthly so the quarterly review is a conversation about trends.
Run it on three to five suppliers first, show the order count next to every percentage, and agree what happens at each score band before the first bad quarter. That last part is what turns a spreadsheet into a supplier that improves.