Building a Supplier Performance Review That Works

Supplier performance reviews fail for predictable reasons. The scorecard measures everything, so no supplier can tell what matters. The data arrives in the meeting for the first time, so the meeting is spent arguing about the numbers. And nothing is written down afterwards, so the same issues reappear next quarter with the same surprise.

A review that works is narrow, predictable and action-led. The scorecard is the agenda, not the outcome.

Measure few things, and the right few

Three to five measures create a conversation; twenty create defensiveness. For most supply relationships the core set is:

• Delivery reliability - on-time-in-full against the originally confirmed date, not the most recently revised one. Re-baselining the date is the single most common way OTIF is inflated.
• Quality - acceptance rate or defects per million, with the cost of rejection made visible.
• Lead time performance - the average matters, but variability matters more, because variability is what you buffer against.
• Responsiveness - time to acknowledge, time to resolve, and how the supplier handles exceptions.
• Cost and total cost - price against agreement plus expediting, premium freight and rework, so cheap suppliers that generate cost elsewhere are visible.

Add category-specific measures only when a decision depends on them. Sustainability, capacity commitments and technical support belong on some scorecards and not on most.

Differentiate by supplier segment

Not every supplier deserves the same process. Segment by spend and by risk or supply criticality:

• Strategic - quarterly business review, joint improvement plan, executive sponsor on both sides, forward-looking topics such as capacity, roadmap and risk.
• Leverage / high spend, low risk - quarterly scorecard focused on cost, service and compliance.
• Bottleneck / low spend, high risk - the group most often neglected. Focus on continuity: capacity, second sourcing, buffer strategy, contingency.
• Routine - automated exception reporting only. A meeting is not justified.

Trying to run a full review with every supplier guarantees that the reviews that matter get the least preparation.

Run it as a cadence, not an event

1. Publish the scorecard at least a week ahead. Data disputes should be resolved before the meeting, in writing.
2. Agree the definitions once, in the contract or in an appendix. What counts as on time, what window applies, whose date is the reference, how partial deliveries are treated.
3. Spend the meeting on causes and actions - not on reading numbers aloud. Two or three actions per quarter, each with an owner on both sides and a date.
4. Open the next meeting with last quarter's actions. This single habit does more for supplier performance than any change to the metric set.
5. Escalate on a defined path. Sustained underperformance should trigger a documented improvement plan with a review point, not an unpredictable emotional escalation.

Track the actions, not just the score. A scorecard measures the past; the action list is the only part that changes the future.

Common mistakes

• Scoring on revised dates. If the supplier can move the promise, OTIF measures paperwork rather than reliability.
• Ignoring your own contribution. Late forecasts, rushed orders, specification changes and unpaid invoices degrade supplier performance. Mature reviews include measures of the buyer's own behaviour.
• Weighted composite scores nobody can decompose. A single number of 82% tells neither side what to do. Show the components.
• Only reviewing poor performers. Strong suppliers are where growth, capacity priority and innovation come from - silence is a wasted relationship.
• No link to sourcing decisions. If performance data never influences allocation or renewal, the review has no consequence and suppliers will treat it accordingly.

Frequently asked questions

How often should reviews happen?

Quarterly for strategic and high-spend suppliers, semi-annual or annual for the rest, with monthly exception reporting in between. Frequency without preparation is worse than a slower, well-prepared cadence.

Should the scorecard be shared with the supplier?

Always. A scorecard the supplier cannot see is an internal report, not a performance management tool.

Where should the data come from?

Your own transactional records, on definitions both parties have agreed. Supplier self-reported performance is useful as a cross-check, never as the primary source.