Why You Shouldn’t Set an NPS Target

Improving customer satisfaction is one of the most powerful levers available to any organisation – commercial or social housing – that wants to prosper financially, satisfy a regulator, and build a leading reputation. So the logic seems obvious: set high NPS targets, tie them to bonuses, watch performance climb.

Unfortunately, that logic is wrong. And the consequences of getting this particular decision wrong are more serious than most senior leaders realise.

Setting NPS targets sounds sensible, but in practice it often leads to gaming, distorted reporting, and worse customer outcomes.

In this post, we’ll explain why satisfaction scores – including NPS – make dangerously bad targets, what Goodhart’s Law has to do with it, and some much more effective targets to set instead. (NPS is great, by the way, but it was never intended to be used this way).

Targets are one of the most powerful tools you have. Use them carefully.

Good targets give your people clarity, alignment, and a shared sense of purpose. They make progress visible and, when well-chosen, they’re motivating. Set the right targets and your organisation moves purposefully in the direction you want it to.

Set the wrong targets, and you get something far worse than missed objectives. You get teams optimising hard for the wrong things – often without realising it, despite your (and their) best intentions.

Consider a target to reduce complaints. It sounds entirely reasonable: fewer complaints suggests happier customers or residents. But anyone who has managed a complaints function knows what happens next. The first order of business is defining exactly what constitutes a complaint.

That’s only fair, isn’t it?

Things that aren’t strictly complaints should be excluded so they don’t count against you?

The number drops straight away. Then the definition gets tightened further. Customer issues that would once have been processed efficiently by a complaints team get reclassified as “queries”, “feedback”, or “service requests”, or even ignored completely. The ‘complaint’ numbers fall. The target is met. Bonuses are paid. All before anything whatsoever has improved for the people you serve.

This poorly designed targeting isn’t hypothetical. It happens regularly, in organisations of all sizes and sectors.

Should You Set Targets for NPS?

In most cases, no.

Using NPS as a target creates the exact conditions that cause metrics to become unreliable. Once bonuses, performance reviews, or departmental success depend on achieving a particular score, the organisation’s incentives shift away from improving customer experience and toward improving the number itself.

That does not necessarily happen maliciously. In many cases, good people simply respond rationally to the incentives placed in front of them.

And that is exactly why NPS targets are problematic.

Why NPS – and any satisfaction score – makes a poor target

Satisfaction scores, whether NPS, CSAT or any other variant, feel like they should make natural targets. They’re simple, they’re comparable over time, and everyone understands the direction of travel. But there are three structural problems that make them unsuitable as the primary target.

They’re subjective

A satisfaction score is a customer or resident’s summary of how they feel about an organisation. Feelings are real and important – but they’re not objective. Two organisations could deliver identical services and receive very different scores based purely on communication style, survey design, or the mood of respondents on the day.

They’re proxies, not primary measures.

When a customer or resident scores you a 6 out of 10, they’re not telling you they want better satisfaction. They’re telling you something specific was wrong: a repair took too long, a call went unreturned, a letter arrived in language nobody could understand, an online system was impossible to navigate. The score is a signal that something underlying needs fixing. Targeting the signal rather than the underlying cause is like covering your car’s warning light with a piece of tape. The dashboard looks fine. The engine is still failing.

They can be gamed.

This is the most serious problem, and it’s more common than most boards realise. Satisfaction scores can be nudged upward through survey design choices that have nothing to do with service quality: reversing the answer scale, adding colour-coding that steers respondents toward higher numbers, using on-screen text that implicitly discourages low scores, or – in the most troubling cases – staff telling customers or residents which score to give because “anything below a 9 affects my performance review.” If satisfaction scores are linked to bonuses, the pressure to game them becomes acute.

Goodhart’s Law: the principle every C-suite leader should know

The economist Charles Goodhart captured this dynamic precisely in 1975. His principle, now known as Goodhart’s Law, states simply: when a measure becomes a target, it ceases to be a good measure.

Applied to customer experience: if you set NPS targets of 70, you will almost certainly get an NPS of 70. But the mechanisms your organisation will deploy to achieve it will not all actually improve service for customers. The score becomes detached from the reality it was designed to reflect. Your board gets reassuring numbers. Your customers get no material improvement.

Fred Reichheld, the creator of NPS, has spoken candidly about this problem. He did not design the Net Promoter Score to be a management target or a basis for individual incentives. He designed it as a diagnostic – a simple way of understanding whether your customers would voluntarily endorse you to others, which correlates reliably with long-term business health.

Used as a diagnostic measure and a starting point for identifying the most important business improvements, NPS is powerful. Making it a target corrupts it.

What to set targets for instead

The good news is that the alternative is both simpler and more effective. Rather than targeting the score, target the specific, objectively measurable factors that drive the score upward.

Good targets in customer experience share three characteristics. They are objectively measurable – two people measuring the same thing should reach the same result. They cannot easily be gamed – because they reflect operational reality, not perception. And they correspond directly to the things that customers or residents actually want you to be better at.

In practice, this means targets like these:

Speed of resolution.

From a customer raising a complaint to receiving a substantive response? How long does it take from a resident reporting a repair to the repair being completed? These are measurable, comparable over time, and entirely within your operational control.

First contact resolution rate.

What proportion of issues are fully resolved on the first interaction, without the customer or resident needing to chase? This is one of the most powerful drivers of satisfaction, and one of the clearest indicators of operational effectiveness.

Accessibility.

How easy is it to reach the right person to help with your particular problem, or obtain a satisfactory answer from a website?

Quality of outcome.

In social housing, for example: after a repair, what proportion of residents confirm the work was completed to their satisfaction and their home was left clean and tidy? In commercial contexts: what proportion of orders are delivered on time, in full, to the agreed specification?

Responsiveness to vulnerable customers

For regulated organisations – particularly in social housing and financial services – are there measurable targets around identifying customers in vulnerable circumstances and ensuring they receive an appropriately supported service within a defined timeframe?

Each of these targets connects directly to something specific that customers want you to do better. Improving them will cause satisfaction scores to rise – organically, durably, and honestly. That is the right sequence: fix the underlying factors, and watch the satisfaction score follow.

The hidden cost of getting this wrong

There is a final reason that a C-suite audience should understand.

Low satisfaction scores, managed correctly, are commercially valuable. They identify customers or residents who are at risk in time to save them before they leave, complain publicly, or escalate to the regulator. An organisation with a well-designed feedback process that surfaces low scores quickly and responds to them effectively will retain customers and reputation it would otherwise lose, and will resolve issues before they become formal complaints.

If your teams are under pressure to keep scores artificially high, those early warning signals get suppressed.

Low scores may be hard to receive emotionally, especially when we’ve built a team that genuinely cares, but they’re a valuable means to an end. Don’t create a culture which punishes them.

In summary

Set targets for the specific, objectively measurable things that your customers or residents want you to be brilliant at. Measure satisfaction scores as a diagnostic, not a destination. And do not link NPS or any satisfaction metric to individual bonuses – you will get the number, and lose the insight.

The organisations that consistently improve customer satisfaction are the ones where leaders resist the temptation of a tidy headline figure, and do the harder, more rewarding work of fixing the actual causes of dissatisfaction – one at a time, systematically, and with accountability.

That’s how satisfaction improves. And that’s how organisations succeed.