1. What is it, in one sentence
Goodhart’s Law says that once you turn a measure into a target, people start gaming it and it stops telling you the truth.
2. What it means to businesses
Every business runs on metrics: sales per head, call times, NPS, conversion rate. They start as honest signals of how things are going. The moment you attach bonuses, rankings or praise to one, behaviour bends around the number rather than around what the number was meant to represent.
You then get a dashboard full of green ticks and a business quietly going wrong. The metric looks healthy because people have learned how to hit it, not because customers are happier.
3. Customer opportunity
Understanding the law helps you measure what customers actually care about, not what’s easy to count. Pair a number with a quality check, such as repeat purchase alongside satisfaction, or speed alongside resolution, and you reward the real outcome.
Customers notice when a business is chasing its own scorecard (the rushed call, the pushy upsell). They also notice when it isn’t. Get this right and you build trust that competitors tied to their targets can’t match.
4. Business threat
The threat is a slow, hidden one. Gaming rarely looks like cheating at first. It looks like a keen team working hard to hit its goal. Over time you can end up with distorted data, damaged customer relationships and, in the worst cases, scandal.
The risk grows the more weight you put on a single number. One metric tied to pay or promotion is an invitation to optimise it at any cost.
5. Three real business examples
- Wells Fargo: Aggressive cross-selling targets pushed staff to open accounts customers never asked for. The sales numbers looked excellent until the scandal broke in 2016 and trust took the hit.
- NHS England (A&E waiting times): A target to treat patients within four hours led to reported cases of patients being moved around or held in ambulances to stop the clock, without any real improvement in care.
- Tesco: In 2014 the retailer admitted it had overstated profit by recording supplier income too early, in a culture under pressure to hit its numbers. A reminder that even the financial measures get gamed.
6. How can we use data to maximise this effect
- Balanced pairs: Pair every target metric with a counter-metric that would worsen if it were gamed, such as sales volume with returns or complaints.
- Outcome data: Add customer-sourced measures like repeat purchase, churn and verbatim feedback that are harder to manipulate.
- Anomaly detection: Watch for sudden jumps, unusual patterns or results clustering just above a threshold. Those are the fingerprints of gaming.
- Rotating metrics: Review and refresh targets regularly, so no single number becomes something people can learn to beat.
7. When is it best to use this technique for retailers
- Setting store and team targets: Reward a blend of sales, customer satisfaction and returns rather than sales alone.
- Running promotions: Judge success on incremental profit and repeat custom, not just footfall or units shifted.
- Reviewing KPIs and bonuses: Run a “how could we game this?” session before a metric goes live.
- Reading online metrics: Treat conversion rate, reviews and ratings with care, because they too can be pushed in unhelpful ways.
The lesson isn’t to stop measuring, because you’d be flying blind. It’s to hold your numbers lightly, like a good friend’s advice: useful, but not to be followed off a cliff.
Leave a Reply