The ‘net promoter score’ could drive up quality, but we have to ask the right questions.

Illustration about patient experience

The magic number for service satisfaction?

The prime minister’s proposal for a “friends and family” test asking patients whether they would recommend a hospital sounds simple. But how exactly are trusts to get their answers? More importantly, what are they going to do with the answers once they have them?

There is much debate currently over the “net promoter score”, an 11-point patient rating from which organisations can create an overall score (see box).

How NPS works

The Net Promoter survey asks customers if they would recommend a service or product to others using an 11-point scale (0-10). Usually a second free text question asks for the main reason for the score. It breaks the answers down into three types of customer and uses these to generate a score, the NPS:

  • Promoters (score 9-10) are loyal and will continue to recommend you to others and repeat purchase or be receptive to upselling techniques
  • Passives (score 7-8) are largely satisfied but are not immune to competitors’ offers
  • Detractors (0-6) are unhappy and are likely to either bad mouth your offering or seek to damage you in one way or another.

The overall score is the percentage of promoters minus detractors. A sensitive indicator, scores can range from -100 (everyone is a detractor) to +100 (everyone is a promoter). In industry, a positive score is regarded as “good” and a score of +50 excellent.

On one side, NHS West Midlands and NHS Midlands and East and others are promoting NPS, suggesting this commercial tool holds the solution. On the other, a recent report for the Care Quality Commission by the Picker Institute Europe says patients like the scale but adds: “We do not recommend the net promoter score for NHS use.” So who is right?

I’m going to attempt to answer that question from my perspective of working both in the NHS and commercial sectors, as a certified NPS practioner and a member of the Market Research Society.

In my view, the NPS can definitely work in the NHS if it is adapted to reflect the differences between a publicly funded healthcare system that is trying to manage demand and a commercial sector organisation that is trying to grow demand. Users must understand that NPS is more than a score on the door; to deliver its full potential, users need to understand how to link the score to behaviour to drive improvements.

NPS in the commercial sector

It’s worth taking a moment to look back at how and where NPS developed. It was the brainchild of American business guru Fred Reichheld and global management consultancy Bain and Company and was presented to the world in the Harvard Business Review of December 2003 as the “one number you need to grow”.

They proposed asking customers one simple question: would you recommend a service or a product to a friend or family member? The answers could be translated into a score that would measure customer loyalty (see box, right) rather than simply satisfaction. Driving this number up would lead to increased profit through lower cost per acquisition of new customers and easier repeat selling opportunities.

This takes us to the second element of NPS: a free text question, asking customers for the main reason they gave their score. When a customer gives a poor rating, the company can pick this up immediately and try to resolve any issues. For example: “I see you gave us a poor score because your package did not arrive on time. Can I offer you a voucher for free postage next time?”

The aim of service recovery is to turn detractors into promoters who have a long-term relationship with the company and to drive growth and profit. In NPS parlance, the common goal is profit.

Clearly, the NHS thinks in a different way. At the risk of stating the obvious, profit and growth are not the raison d’etre of the health service. Indeed, the NHS is striving to drive down or at least manage demand by improving the long-term health of the population.

So the common goal of profit used by NPS in business does not work in the NHS. In health settings a legitimate common goal or shared value might be “cure” or “outcome”. Equally, it may be “being treated with dignity” or “being treated in a clean environment”. This changes the philosophy that underpins commercial NPS thinking.

Then there is the nature of choice and substitution. In commercial settings there is almost always a competitor ready to take your customers - hence the emphasis of NPS on loyalty rather than satisfaction alone. In the NHS, choice is more complex.

Next, there is the process of service or experience recovery. The backbone of NPS is the ability to follow up poor experiences or when a passive or detractor is identified. It is doubtful that the NHS has the people and the communication channels to do this with the same rigour as the commercial sector.

Nor is there any currency to compensate patients for a poor experience. How could the NHS recover a poor experience? A promise that next time you break your leg you will be seen first? A token for the coffee machine in A&E? These are ridiculous suggestions but the point is that the NHS organisations need to consider different mechanisms for influencing their NPS.

There are some interesting paradoxes around transparency and data sharing. Commercial organisations do not tend to publish their NPS, as no two companies operate the same way. Yet the NHS proposes to publish scores, largely as a mechanism for supporting choice.

However, the score itself is not the point of NPS.  The score is only the headline indicator; it is knowing what influences this that is the key. A score is a rallying point but it does not address why people responded the way they did.

Commercial organisations know who is answering surveys and can respond to individual experiences. This is extremely unlikely to be the case in the NHS where information governance and patient confidentiality require anonymous responses. Most IT systems would simply not allow well established NPS tools such as cloud-based Outlook plug-ins to exist inside the NHS infrastructure.  This means that to achieve the cultural and operational shift that NPS demands, we need to find another methodology to “close the loop” when engaging with a dissatisfied patient.

Should the NHS use NPS?

So can NPS be useful in the NHS? A qualified yes, although it does need to be adapted. Launching a poorly designed friends and family NPS programme will result in inaccurate and incomplete data providing top-line scores that do not necessarily indicate the quality of the experience patients receive.

Correctly configured and implemented, NPS can deliver a simple and effective means to embed patient experience at the very heart of the way NHS organisations behave. The NHS needs to understand that NPS is not a research tool or methodology but a way of driving quality focused behaviour and operations. As such, it can be extremely powerful.

First, the NHS needs to ask the right question based around the right common value; hence the need for board involvement in NPS.

For example, the question: “How likely would you be to recommend our services to your friends and family?” risks eliciting the response: “I would not recommend my friends and family become unwell.” Rephrasing to “How likely would you be to recommend/suggest our service to a friend, family member or someone with a similar need?” is less ambiguous and may be closer to what you want to know.

The NHS needs to find ways to feed back information to the staff who influence patient experience. This needs to be as near real time as possible to be relevant and meaningful and to take place at a level where staff can take remedial action - for example, ward level or outpatient clinic.

As staff get closer to the data, so culturally it will become easier to accept the importance of an open relationship between staff, patients and their experience. Ensuring patients are aware of this process and encouraging them to feed back regularly fuels the process.

The Picker/CQC report that claims NPS should not be used in the NHS is the right answer to the wrong question. Has NPS got a place in national surveys or traditional “research” methodologies? Probably not - but it was not designed for this purpose. Combined with good causal analysis, proper top down involvement (not “sponsorship”) and intelligent survey strategies, NPS is not only a valid metric but a powerful behaviour driver and, as good operators know, staff behaviour drives customer experience.

So as you consider NPS, ask yourself: are you really seeking to drive improvements - or simply seeking to publish a headline number and fulfil a centrally directed requirement? The choice is yours.

Toby Knightley-Day is managing director of Fr3dom Health.

Find out more

Overarching questions for patient surveys: development report for the Care Quality Commission