Tell Me How You'll Measure Me, And I'll Tell You How I'll Behave

I've seen real performance problems. But I've also seen teams measured in a way that makes good performance look like failure.

KPIs are supposed to create clarity. Focus. Accountability.

And they do - when they're designed properly.

But when they're not, they don't just fail to measure performance. They actively distort it. Understanding why KPIs fail - and how to fix them - remains one of the most underleveraged management skills in most organisations.

The Problem Most Teams Don't See

A KPI that depends on someone else's behaviour is not measuring your performance. It is measuring your exposure.

You can do your job well. Make the right decisions. Execute exactly as expected.

And still end up red - because the outcome depends on someone else.

That's not accountability. That's a design flaw.

Where It Breaks

The issue usually isn't the KPI itself. It's the hidden mix of control and dependency.

Ownership becomes unclear. Responsibility becomes blurred. And behaviour starts to shift - quietly.

People escalate more. Defend more. Optimize for how they're measured, not for what the business actually needs.

Not because they're wrong. Because the system is.

A Very Familiar Example

I see this repeatedly in procurement and logistics.

Procurement is measured on cost - unit cost, delivery cost, total spend. On paper, completely reasonable.

Then reality kicks in.

Sales commits to a timeline that can't be met. Production falls behind. A decision is made - often above procurement - to expedite. Air freight. Premium rates. Last-minute changes.

At the end of the quarter, procurement reviews performance. The numbers are red.

Not because procurement failed. Because the system forced a decision they didn't control. And yet - the KPI says they own it.

When the Number Has No Connection to Reality

Sales targets set for the board. Double-digit growth. Big, bold numbers that signal ambition and momentum.

What they sometimes don't reflect is the market. The product maturity. The competitive landscape. The fact that the industry is contracting, supply chains are under pressure, and the realistic growth this year - done well - is 6%.

But the board wants 18%. So 18% becomes the target. And the pressure cascades down to sales managers, operations, finance, and production. Everyone is now building plans around a number that was never grounded in operational reality.

Ambition is not a KPI. Growth targets should push the organisation forward - but they need to be anchored in market data, operational capacity, and honest assessment. A target that is never achievable stops being a motivator and becomes background noise. People stop believing in it, stop chasing it, and quietly start managing around it instead.

The most effective KPIs are ambitious enough to drive improvement and realistic enough to be taken seriously.

When OTIF Becomes a Weapon

On-Time In-Full is one of the most important supply chain KPIs. It is also one of the most abused.

The problem is not the metric. The problem is the lead time it is measured against.

When sales or management sets lead time commitments without understanding the product, the production process, or the supply chain behind it - the operations team is being set up to fail before the order is even confirmed.

"Everything has to be fast" is not a lead time. It is a feeling. And feelings do not run production schedules.

OTIF only means something when the lead time it references is realistic. Otherwise it is not a performance metric - it is a blame generator.

The Same Problem, Across Every Function

This is not a procurement problem. This is not a logistics problem. This is a measurement design problem. And it shows up everywhere:

• Customer complaint compensation measured under Quality - but the decision to compensate is made by Sales as a "commercial decision." Quality owns the KPI. Sales owns the cheque.

• Product launch KPIs that push Technology to scale before the product is ready - Operations receives something immature and is expected to deliver as if it isn't.

• Production waste % as a KPI, with no budget allocated for the machine maintenance or replacement that would actually reduce it.

• Number of leads as a sales KPI - which tells you nothing. Conversion rate is the metric. Quantity without quality is noise.

• Supplier payment terms are not measured at all in Finance KPIs until a key supplier quietly deprioritizes you because you are always the last to pay.

• KPIs that are never reviewed. Set once, never revisited. The market changed, the business changed but the metric from three years ago is still running. A KPI that is never challenged becomes a habit, not a measurement.

• Vanity KPIs. Metrics that look good in a presentation but don't drive any decision. If a KPI doesn't change how someone acts tomorrow - what is it for?

KPI Best Practices: How to Build Metrics That Actually Work

The fix is not fewer KPIs or softer targets. It is better design.

Three questions to ask before any KPI is set:

What behaviour will this metric incentivise?

Not in theory - in practice, given the pressures this team operates under.

Who controls this outcome?

If the team being measured cannot directly influence the result, the KPI is measuring the wrong thing or the wrong people.

Does this KPI see the full picture?

A single metric almost always has a blind spot. Cost without quality. Speed without accuracy. Volume without conversion. Balance the metric or balance it with others.

The Bottom Line

KPIs always drive behaviour. The question is not whether they work. The question is whether they are driving the right behaviour - in the right place - under the right control.

How do I know if my KPIs are creating the wrong behaviour?

Look for workarounds. When teams start optimising for the number rather than the outcome - reporting what looks good rather than what is true, making decisions that protect the metric rather than the business - the KPI has become the problem. If your logistics team is avoiding urgent deliveries even when the business needs them, ask what they are being measured on.

Who should set KPIs - management or the people doing the work?

Both. Management sets the strategic direction and the outcomes that matter. The people doing the work understand what is operationally realistic and what the unintended consequences of a given metric might be. KPIs set without that input tend to be technically correct and practically destructive.

How many KPIs is too many?

When people stop being able to tell you which ones actually matter - you have too many. A good rule: if a team cannot recite their top three KPIs and explain how their daily work connects to each one, the measurement system has lost its purpose.