Suppliers are good at a great many things. What they are best at is delivering whatever you measure.
This article is an argument for handling scorecards and KPIs with some care.
You get what you measure
Every metric on a scorecard is a signal. It tells a supplier what will be discussed at the next review, what affects the odds of renewal, and what an account manager will be held to.
Suppliers respond to that rationally. They improve what is measured. And because their capacity, like yours, is finite, that improvement often comes at the expense of something that is not measured.
Steer on price, and quality shifts
The best-known example. A scorecard that leans heavily on cost reduction gets cost reduction. The question is where it comes from.
Sometimes from genuine efficiency. More often from places that are not on the scorecard: a cheaper material that just meets specification, less experienced engineers on the maintenance contract, a longer response time on questions that are not urgent. Each defensible on its own. Together, a supplier who performs better on paper and worse in practice.
Steer on delivery reliability, and lead times grow
Delivery reliability — on time and in full — is one of the most widely used metrics, and rightly so. But a supplier who is judged on it will try to protect that score by agreeing more generous lead times.
A part that always arrived in two weeks is now promised in three. Delivery reliability goes up, so does your stock, and flexibility goes down. On the scorecard it looks like progress.
Steer on quality, and flexibility disappears
A supplier judged strictly on rejects and complaints becomes cautious. They take less risk, deviate less readily from the standard process, and say no more often to an urgent request or a modification.
That is fine as long as everything goes to plan. But the moment you need a supplier who thinks along with you, tries something or makes an exception, it turns out your scorecard has taught them precisely the opposite.
Steer on responsiveness, and you get yes to everything
And then the metric that looks most harmless of all: how quickly does the supplier respond? Steer on that and you get fast answers. Not necessarily good ones.
An account manager who knows they are judged on responsiveness says yes quickly. Yes, we can do that. Yes, we will sort it out. Yes, that fits the schedule. The commitments arrive quickly. The delivery arrives later, or not at all.
No single number tells the whole story
The underlying point: every metric shows only part of a relationship. Delivery reliability says nothing about quality. Cost reduction says nothing about innovation. A quality score says nothing about how a supplier behaves in a crisis.
A scorecard is therefore always a simplification. That is no objection, as long as you treat it as one.
Why the scorecard is indispensable anyway
This is not an argument for abolishing scorecards. Without measurable agreements, supplier management becomes a matter of instinct, and then the most recent incident decides the verdict on a relationship of years.
A good scorecard makes the conversation factual. It shows trends you would otherwise miss. And it gives both you and the supplier something to hold on to. The question is not whether you measure, but how you use the result.
Signals, not verdicts
Use the results as direction, not as sentence. A falling score is a reason to start a conversation, not a conclusion. A rising score is a reason to look at where it came from.
With every improvement, look also at what is happening beside the measured metric. Has delivery reliability risen? Then look at lead times as well. Have costs come down? Then look at complaints from users. Anyone who examines only one side of the balance sees only what the supplier wants to show.
One metric at a time
A common mistake is to introduce an extensive scorecard all at once. Ten metrics, all at the same time, each with its own weighting.
The effect is that you set ten behavioural changes in motion at once and can no longer see which came from where. Better to start with a few metrics that genuinely matter for this category, watch how the supplier responds, and only then add the next. The right combination also differs from category to category: what suits a strategic supplier of critical components does not suit a supplier of consumables.
Measure your own side too
A scorecard almost always looks in one direction. But a supplier’s performance depends on you as well.
Are orders placed on time, or often at short notice? Does the specification stay stable, or change halfway through? Are invoices paid on time? Suppliers, like any other business, decide where to put their best people and their scarce capacity. A customer who orders late, changes often and pays late is not the customer who gets helped first when things get tight.
Bring that into the conversation and you get a fairer picture — and often a better supplier.
The conversation counts, not the number
In the end a scorecard is not a goal but a reason for a good conversation: what are we seeing, where does it come from, and what do we expect from each other next quarter?
A supplier who can be honest in that conversation about what is not working is perhaps worth more than a supplier who scores a pass on everything. The first tells you what you need to know. The second tells you what you want to hear.
Conclusion
Your scorecard does not only measure how a supplier performs. It also determines what that supplier gets better at, and what they quietly give up. Steer on one metric and you get that one metric, and pay the price somewhere else.
So use the numbers as the beginning of the conversation, not the end. Look at what happens beside the score and you see the supplier as they really are.
