When Metrics Mislead: Goodhart’s Law in change programmes

In a nutshell

Goodhart’s Law and the illusion of progress in change programmes

When a major bank celebrated 100% completion of mandatory compliance training, leadership was thrilled, until audit results revealed widespread misunderstanding of critical procedures.

What happened?

Goodhart’s Law: “When a measure becomes a target, it ceases to be a good measure.”

Employees had rushed through training to tick the box, not to learn. This pattern repeats constantly in change management: the project team hits its metrics yet misses its goals.

This blog explains how to avoid this trap.

Key takeaways

• Metrics drive behaviour, but sometimes it’s the wrong behaviour
• Successful change requires measuring what matters, not just what’s convenient
• Practical approaches are needed to keep metrics meaningful while driving real change

Metrics and KPIs are valuable tools. They bring clarity and structure to complex work. But they can backfire when overemphasised or mindlessly pursued. A KPI should inform decisions, not replace them. The metric is there to guide you, not become the destination.

The Measure vs. the Goal

We’ve all seen it: teams so focused on a target that they lose sight of its purpose. Take the example of assessing a chef by the number of meals served. It’s measurable, but it tells you nothing about flavour, quality or whether anyone will come back.

That’s Goodhart’s Law in action. Coined by economist Charles Goodhart, it reflects a simple truth: once a measure becomes the goal, it stops being a reliable proxy for success. People adjust their behaviour to hit the number, not to achieve the outcome it was meant to represent.

In change management, metrics are comforting. They provide evidence of progress: training completion rates, user adoption stats, and engagement scores. They’re the dashboard dials, but stare at them too long, and you risk veering off the road.

Set a target like “90% of users must log into Workday in Week 1” and teams will find a way to meet it, even if users just poke around and log out. The dashboard will light up green, but it doesn’t reflect real adoption.

At best, it’s amber.

How Goodhart’s Law derails change initiatives

Change initiatives often hinge on KPIs: adoption rates, cost savings, morale boosts. Targets bring focus and accountability. But people are clever. When they know what metric matters most, they’ll optimise for it, sometimes at the expense of the change itself.

This isn’t malicious. It’s natural. If you tie bonuses, reputations, or job security to a number, that number will go up, whether or not anything’s truly improved.

Other cognitive quirks also come into play. Leaders may fall into confirmation bias, seeing what they want to see in the data. The act of measurement can itself influence behaviour. People act differently when they know they’re being measured. Goodhart’s Law sits at this crossroads: turn a measure into a target, and you often change the behaviour it was meant to track.

Let’s bring this to life with a few examples from our work at Change Associates.

Partner experience over performance targets
The John Lewis Partnership

One of the more subtle, yet powerful, drivers of change is employee morale. At the John Lewis Partnership, they recognised that what really mattered wasn’t just hitting performance metrics, but how their Partners (employees) experienced their roles day to day.

We helped map the Partner journey in detail, identifying opportunities to improve onboarding, engagement and motivation. Rather than treating engagement surveys as vanity metrics, the business used them as indicators of where and how to act. They didn’t chase a “90% engagement score”; they asked what could be improved and how to do it.

The result was better insight, smarter decisions and a clear link between employee experience and customer outcomes.

A case of using metrics as a compass, not a scoreboard.

Read the full John Lewis Partnership case study here. 

The count vs. the change: Lessons from INEOS

Another example is our work with INEOS FPS, the energy business.

Our task was to map and improve 130 critical processes. It would’ve been easy to declare victory once the 130 documents were delivered. But that would have missed the point. The value wasn’t in ticking off process maps but in the improvements we made along the way.

Had we focused only on the number, the team might’ve rushed through, producing documents that met the brief but lacked substance. Instead, the goal remained clear: capture knowledge and improve operations. We engaged operators, challenged assumptions, and refined processes.

The count was a milestone, not the mission.

It’s a lesson we’ve seen repeatedly: measure completion, but prize improvement. The number of documents, logins, or survey responses only matters if it leads to something more meaningful.

Read the full INEOS case study here.

Warning Signs: When you hit the number but miss the goal

It’s easy to imagine what happens when the number gets ticked, but nothing really changes. Consider an organisation which defines success for a major system implementation as 100% completion of a two-hour e-learning module. By launch day, the dashboard looked great, every name ticked off.

But the reality tells another story. Post-launch feedback reveals that many employees still don’t know how to use the new system. Most had clicked quickly through the training just to reach the certificate at the end, without taking in the content.

The team had optimised for the target, completion, not for the real goal: capability.

Five practical ways to keep metrics meaningful

So, how do you stop a useful metric from going rogue?

Here are a few principles we share with clients:

1. Use a mix of measures. No single number can tell the whole story. Combine hard stats (logins, documents) with softer indicators (sentiment, confidence).
2. Check in qualitatively. Talk to people. Use interviews, manager feedback and focus groups to understand what’s really happening.
3. Watch for anomalies. Sudden jumps in a metric, especially near deadlines, can be signs of gaming. Ever noticed how many one-to-one reviews are completed as you approach bonus time?
4. Adjust targets. As change progresses, some metrics become less relevant. Feel free to evolve or discard them if they no longer have meaning.
5. Watch the culture. Ultimately, no measure stays “clean” in a culture that rewards hitting numbers at any cost. Celebrate how results were achieved, not just that they were.
   

Create a balanced measurement framework

Consider using a 3 x 3 matrix approach in your change management metrics:

This structure ensures you’re tracking the full journey of change, not just convenient numbers. For example, in a CRM implementation, you might track early engagement in design workshops (leading), current system usage patterns (performance), and eventual customer satisfaction scores (lagging).

Beyond Measurement: Reconnecting with your change purpose

In an earlier blog, we discussed the McNamara Fallacy, valuing only what’s easy to measure. Goodhart’s Law takes that conversation further: even good measures lose value when turned into rigid targets.

This isn’t an argument against metrics. You need them. But use them wisely. Look beyond the numbers. When a KPI goes up, ask what might be going down. When a target is consistently hit, ask if it’s still the right target. Use data to start conversations, not to shut them down.

Goodhart’s Law reminds us to lead with both head and heart. The head knows that what gets measured gets managed, while the heart knows that not everything that matters can be measured.

So, by all means, set your targets. But keep Goodhart’s caution in mind. If hitting the number becomes the goal in itself, pause, and reconnect with the change you’re trying to make.

Your next step: A metric audit

This week, select one key change initiative in your organisation and ask:

• Are we measuring what matters or what’s convenient?
• Have our metrics become the goal rather than the guide?
• What behaviours are our measurements encouraging?
• What would we measure differently if we started today?

Remember, the goal isn’t perfect measurement, it’s meaningful change. Keep your metrics in service to that mission, not the other way around.

After all, the aim isn’t just to win the metrics game, it’s to improve the business, its performance, and the lives of the people in it.

Get in touch to discuss how we can help you to keep your change initiative focused on the right metrics.  

Image by MidJourney