NPS, CSAT or Customer Effort Score – which metric is right for your business?

NPS, CSAT or Customer Effort Score?

As the saying goes – you can’t manage what you don’t measure.  To overcome this many companies look to a single metric that they can use to measure the health of their company and that business leaders can use to predict future performance of the organisation.

No one question can give all the answers needed across a business – but compromises have to be made between the simplicity of a single question and a drawn out survey that results in a poor response rate.  A single question can be asked frequently and across multiple industries and countries giving business leaders a sense of momentum of their own organisation and comparison with others.

Which question to choose?

With many options out there, which single question is the best measure of today and predictor of future performance?

NPS – Net Promoter Score

The Net Promoter Score is perhaps the most famous of these metrics – sometimes known as The Ultimate Question!  The question is founded on the assumption that happy customers are more likely to recommend a company to others:

On a scale of 0-10 how likely would you be to recommend [company name] to a friend or colleague?

Responders are split into three categories:

  • 9-10 – Promoters
  • 7-8 – Passive
  • 0-6 – Detractors

The Net Promoter Score is a simple calculation of the percentage of detractors subtracted from the percentage or promoters.  Any score over zero is a good start and the top businesses in the world would be happy with a score over 50.

Potential Concerns

Whilst NPS provides a simple question concerns have been raised that agreeing you are ‘likely to recommend’ does not infer that you have or will recommend and therefore might not be a good predictor of future performance.

CSAT – Customer Satisfaction Scores

CSAT is a broad term that covers many question types that all attempt to uncover how satisfied current customers are with the product, or a particular interaction they have just completed.

A typical example would be customers being asked to rate a customer support interaction or their renewal process with a question such as:

How would you rate your experience with your recent support requirement?

  • Very dissatisfied
  • Somewhat dissatisfied
  • Neither satisfied nor disatisfied
  • Somewhat satisfied
  • Very satisfied

The CSAT score is derived by looking at the percentage of respondents that select somewhat or very satisfied and a score of above 70% would make many companies feel they were doing enough.

Potential concerns

The CSAT score is probably the weakest predictor of future behaviour of the three.  The questions are often framed around a specific interaction – a support event or a product delivery – and the customer is focused on that rather than their wider relationship with the company.

It is also important to choose the wording of the options correctly.  The selections we’ve listed make it easy for customers to select 3 or 4 without having to commit to a strong point of view.  It should also be remembered that mildly satisfied or dissatisfied customers are less likely to complete your survey thus skewing the results.

CES – Customer Effort Score

The newest metric to gain popularity is the Customer Effort Score.  Extensive research over 5 years by the team that created it showed that once a customer had received a satisfactory experience there was very little increase in loyalty by wowing them even further.

The team found that the real uplift in loyalty came from having a very poor customer experience to having one that meets their expectations.    The best way of bumping an experience up to satisfactory, according to the research, is to remove as many of the barriers as possible from the customer’s path.

Typical obstacles in a customer’s path might be:

  • an overly complex IVR with many dead end choices
  • multiple transfers between departments
  • having to call in multiple times to resolve a problem
  • not listening to preferences or selections made
  • having to switch channel from social, to email, to phone to resolve a problem

The CES is derived from asking a simple question:

How much effort did you personally have to put forth to handle your request?

Scores range from 1 (very low effort) to 5 (very high effort)


Learn How to Use Post Call Surveys to Collect CSAT & NPS Info


To repeat what we said at the start – no one question can determine the future success of your business – but some questions can give better actionable insight than others.  CES gives you a focus on removing obstacles from your customers path and encourages you to contact any customers that are scoring you anything other than 1.

It is also important to remember that any metric requires regular, continual measurement rather than a single audit.  You need your team to understand that the metric selected is important and that continual improvement in it is a core business goal.

What metric do you follow in your business?  Why did you select that metric above others?  We’d love to hear your thoughts here on the blog.


Categories: Customer Service

  • Catherine Rasman

    We recently engaged in our first NPS effort in Q3, hoping to learn more about our company’s brand and our customers’ level of loyalty. We’ve been measuring CSAT for years, haven’t found the feedback particularly helpful, although the reporting the score paints a very happy customer picture. I’m recommending a replacement on CSAT surveys on our Support team to CES – hoping to get a better understanding of the customer’s perspective, without their fear of hurting the feelings of the Support rep who took their call.

  • michael olmsted

    Why not ask all three and try to correlate the answers?
    We ask the loyalty question
    We ask a Customer Effort question
    We ask a series of 4 Csat questions.

    Would it not be logical to try and look to all the answers to see how they relate?

  • Siddhartha Datta

    How different are these concepts based on industry, is it the same or tweaks or absolutely different

  • Eric D.

    Well said. well researched

  • I would like to know how to draw CSAT indicatore line (from green to Red)
    as we have now applying CSAT Scores between 1-5, while 1 – 2.9 is Red indicatore (we have to call customers and understand their cases and resolve it) 3.0 – 3.9 is good and 4.0 – 5.0 is excellent service.

    can someone help us with the correct and standard Scores that is applied as global standard?

  • Hasan

    In IVR, some people might think it is well organized while other can feel it too granular. Similarly in CES how one knows that his system is not solving his customer problem, if he knew he would have fixed it. As company we should focus on detractors who are likely to share bad experience with other or will never return. So we need a way to find such customer who had bad experiences and then a system which can track these user behaviours through various source like his social presence or his experience at competitors site. Based on above I think, CSAT will still be important to know what went wrong with customer and they are will likely to share them ( if not email then via call), another important metrics is first call resolution that also gives an indication how well his experience could be.

  • Juan Tavarez

    HI Hasan, as you explain there is a great importance in work with NPS, but NPS is much more holistic,
    CES is Focus in clean your customer journey, in made much more easier for the customer to get what he need, if you could do that better an faster and easier that your competitors you will be ahead and them will bring the word of mouth, the increase in purchase etc. etc. I prefer to use all of them NPS and CES and CSAT and FCR and cover all the possibilities.

  • Praveen Harkawat

    Hi, I think CSF is more important then any other metrics. NPS does not give clear picture of business as it’s focussed only on one project / deliverable. Customer Effort metrics will have it’s own challenges from data collection point of view. Please share some examples if available.