A Supplier Scorecard for Wear Parts: Metrics That Change Behaviour

Most suppliers are scored once a year, on a form nobody reads, and then managed on price for the other eleven months. A wear parts supplier scorecard earns its place only if it changes what the buyer asks for at bid stage and what the supplier does between orders.

This page sets out what a scorecard is for, how to weight quality metrics, how to treat delivery and continuity, which commercial metrics matter beyond price, how to score technical support, what cadence keeps it alive, and how to roll it out without creating a second bureaucracy.

What the Scorecard Is For

It exists to change behaviour, not to rank suppliers.

A scorecard that does not change a purchase decision or a supplier action is administration; a short one that does both is worth maintaining.

Start by naming the two decisions the scorecard must influence. The first is award: which supplier gets the next programme or the next trial order. The second is corrective: what the supplier must fix before the next shipment. A scorecard that serves neither purpose should be deleted, because it consumes time from the people who inspect parts and chase deliveries.

That framing avoids the most common failure, which is scoring everything. Wear parts are bought against a specification, so the metrics that matter are the ones tied to conformity, continuity and evidence. Fitment failures, dimensional deviations, missing traceability references and late deliveries are all scoreable from documents the buyer already holds, and each one points to a specific action.

It also keeps the scorecard honest about what it cannot measure. A supplier cannot be scored on a wear life claim unless the buyer records change-out data, and a scorecard that grades a technical judgement without data produces confident nonsense. Where a metric has no record behind it, the honest move is to build the record first. Public procurement practice guidance such as that published by the US General Services Administration is a reasonable reference for keeping evaluation criteria documented and defensible.

Quality Metrics and How to Weight Them

Quality metrics should be weighted by what they cost the buyer, not by how easy they are to collect. A missing certificate is an annoyance; a part that does not fit is a lost shift.

Four metrics carry most of the weight. Conformity on arrival covers dimensions against the drawing, hole spacing and bolt pattern, insert position and visible joint defects. Evidence completeness covers whether the grade record, hardness and grain size results and dimensional report arrive with the delivery. Traceability covers whether the lot can be identified after installation, which is what makes an investigation possible. Fitment outcome covers first-article success on the actual machine.

Two supporting metrics are worth adding for a programme rather than a trial order. Specification responsiveness measures how quickly a supplier answers a technical query with something usable, such as a grade recommendation or a tolerance statement. Change control measures whether a process, material or subcontractor change is notified before it ships, which is the difference between a managed change and a surprise.

Quality metrics, evidence needed and suggested weight

Metric	Evidence the buyer needs	Weight band
Conformity on arrival	Incoming inspection record against drawing	Highest
Evidence completeness	Grade, hardness, grain size and dimensional documents	High
Traceability	Lot reference on the packing list and the part	High
First-article fitment	Fitment check on the machine that will run the part	High
Specification responsiveness	Written technical answers and response time	Moderate
Change notification	Notice of material, process or subcontractor change	Moderate

Weighting in bands rather than percentages is deliberate. Precise weights imply a precision the underlying records rarely have, and they invite arguments about arithmetic instead of arguments about parts.

Delivery, Lead Time and Continuity

Continuity is the metric that decides whether a supplier can stay on a programme, and it is measured by delivery performance against the terms recorded in the order documents rather than against a promise made in a proposal.

Score three things. Delivery to the agreed date covers whether the shipment arrived when the order documents said it would. Completeness covers whether the whole order arrived, because a partial delivery of edges without fasteners is a non-delivery for the crew. Continuity covers whether the supplier held the specification across the period, with no silent substitution of grade, profile or packaging.

Two structural questions back those metrics up: does the supplier manufacture the part or resell it, and does the supplier hold stock for the programme or produce to order? Those answers explain most delivery outcomes, and they are worth recording in the supplier file rather than rediscovered each season.

Where a supplier falls short, the scorecard should distinguish a capacity problem from a planning problem. A capacity problem is fixed with lead time and volume information; a planning problem is fixed with a conversation about order patterns. Conflating them is how a good supplier gets replaced by a worse one that promised faster.

Commercial Metrics Beyond Price

Price belongs on a scorecard as one line among several, and the lines around it explain whether the price was actually good. Total cost per unit of work is the honest comparator, and it depends on wear life as much as on the invoice.

Four commercial metrics are worth scoring. Price stability covers whether the effective cost changed between awards without a documented reason. Commercial clarity covers whether the terms that apply – volume bands, sampling, packaging, delivery windows – are stated in the written quotation or order documents rather than agreed verbally. Cost of failure covers the emergency orders, express freight and rework caused by non-conformance. Administration cost covers how much buyer time the supplier consumes in chasing documents and answering queries.

The last two are the metrics most buyers never record and most suppliers never see, and they are often the ones that change behaviour. A supplier who learns that a missing dimensional report costs the buyer two hours per delivery will fix the report. A supplier who is told only that the unit price is high will discount.

Where a buyer supplies parts under a wholesale or distributor arrangement, the same logic applies at a different level: the scorecard should measure whether the supply route supports the programme with consistent documentation and fitment data, which is why the wholesale supply route is worth documenting rather than assuming.

Technical Support and Responsiveness

Technical support is scoreable, but only if the buyer records what was asked and what came back. Response time and answer quality are separate metrics, and conflating them hides the suppliers who reply quickly with nothing useful.

Record three items for each technical interaction: the question, the time to a written answer, and whether the answer was specific enough to act on. A supplier who answers a fitment question with a drawing or a tolerance range scores well; one who answers with a datasheet scores poorly even if the reply arrived in an hour. Over a season, that record produces a defensible ranking that no sales conversation can displace.

Scoring should also capture whether the supplier volunteers a change. A supplier who flags that a grade is at the limit of its suitability for a surface, and proposes an alternative, is reducing the buyer’s risk; a supplier who accepts every enquiry and ships whatever was ordered is not. That behaviour is worth more than a small price advantage, and it is invisible unless it is scored.

On the safety side, it is reasonable to include whether the supplier provides the handling and guarding information the fleet needs for its own risk assessments. General industry safety material such as the framing in 29 CFR 1910 is a useful reference for what the buyer will be asked to document internally.

Review Cadence and Consequences

A scorecard that runs once a year is a snapshot; one that runs quarterly with a short agenda is a management tool. The cadence should match how often the buyer can act on the result.

Quarterly is usually right for an active programme, with an annual review for the summary picture. Each cycle needs three things: the metrics, one specific corrective action per underperforming supplier, and a decision about award volume or trial scope. A review without a consequence trains every participant to treat the score as decoration.

The corrective action should be written as a change, not a criticism: provide the dimensional report with each delivery, notify process changes before shipment, hold a first-article approval for a new attachment, or supply a fitment drawing for an existing one. Each of those is verifiable at the next cycle, which is what makes the loop work.

Where a supplier performs well, the consequence should be structural rather than celebratory: earlier involvement in specification discussions, a larger share of a programme, or a trial on a new route class. Suppliers respond to scope, and scope is the strongest signal a buyer has.

Finally, keep the record. Whether a review body audits the file, the pattern that matters is the same one documented in reviews by the US Government Accountability Office: criteria that were recorded in advance survive scrutiny, and criteria invented afterwards do not.

Rolling It Out Without Adding Bureaucracy

The failure mode is a scorecard with thirty metrics that nobody maintains after the second quarter. The fix is to start with four metrics, all of which come from documents the buyer already creates.

Begin with conformity on arrival, evidence completeness, delivery to the agreed date and administration cost. Those four can be filled in from an inspection sheet, a packing list, a delivery note and a timesheet, which means the scorecard is a by-product of work that is already happening. Add traceability and fitment outcome once a change-out log exists, then add commercial and technical metrics in the second year if they are still needed.

Keep the format to one page and the language to the clause the supplier must change. Suppliers read scorecards the way buyers read specifications: they act on the line that names a requirement. A metric such as documentation of measurement practice, framed against a published reference like the material on measurement and traceability from NIST, is actionable in a way that a general quality score is not.

Two other details make rollout easier. Publish the metrics and weights to bidders before award, so the scorecard is part of the requirement rather than a surprise afterwards. And keep one owner: a scorecard maintained by three departments is maintained by none. Technical and quality records from bodies such as ASM International can serve as neutral references for grade discussions, which is useful when a supplier disputes a score on material grounds. Where a supplier is supplying a government entity, the supplier registration platform at SAM.gov is also the place to check that eligibility requirements remain current.