When it comes to measuring customer experience, one of the key questions I’m often asked is which is better: CSAT (Customer Satisfaction) or NPS (Net Promoter Score). In fact, this question has been a hot topic for a number of years and there is mixed opinion, and supporters and critics of both.

I believe both CSAT and NPS have merits, but they really do different jobs. While CSAT measures a users’ satisfaction with a product or service, NPS measures customer loyalty. CSAT is typically based on a number of statements that customers rate their agreement or disagreement with e.g. Overall I was satisfied with <product/service>. While NPS is based on a single question based on a customers’ overall interaction with a company i.e. How likely are you to recommend our company/product/service to your friends and colleagues?

So when I’m asked by a company whether they should be using CSAT or NPS, my first question is what do you want to measure and why?

  • Want to understand what customers think of a specific product or service, or measure impact of changes to a specific product or service? Then CSAT would be my recommendation.
  • Want to measure customer loyalty based on a customers’ perception of their end to end customer journey? Then NPS is better placed.

Why I like and use CSAT

The key benefit I’ve found with CSAT is that it’s based on a here and now reaction to a users’ satisfaction with a product or service (we try and get a CSAT score within 30 minutes of a product or service being used), whether that be through a website, IVR, smartphone or some other channel. Because NPS focuses on a customers’ wider experience of the brand, it can be difficult to pin point where the specific channel pain-points may be. In addition, just because someone says they would recommend brand x or product y, it doesn’t mean they actually will. NPS is focused on intention and not actual behaviour.

Measuring CSAT also allows us to see how channels may differ in terms of user satisfaction and to measure customer experience for specific tasks within an IVR, website or mobile application. It also means that when I want to understand and measure specific attributes of interaction (e.g. perceived time on task, navigation, terminology), a CSAT survey allows me to use more than one question to focus on these areas.

When I’ve used NPS to gauge reaction to a particular product or task, respondents have commented that ‘just because the IVR is usable, it doesn’t mean that I’d recommend the company’. Propensity to recommend and associated loyalty is based on customers’ entire experience and interaction with a brand, so in user research this again highlights the importance of understanding what is you want to measure. In this instance, NPS isn’t as relevant or as useful to me as CSAT.

The really important bit!

Whether you measure CSAT or NPS, it’s what you do with those measures that really matters. A measure alone won’t tell you anything about why your customers are detractors or promoters (NPS) or why you may be seeing a lower than expected satisfaction score. You may find an increase or decrease in your NPS score, but this is meaningless without understanding what is contributing to the changes and actively addressing the issues. My key recommendation with any customer experience measure is to supplement this with qualitative research (e.g. interviews, focus groups, user testing) to really understand the reasons behind the numbers and to ensure you have actionable outcomes to address any issues uncovered. A measure alone isn’t useful unless you use it to drive and optimise customer experience. So collect, understand and make changes!


Topics: Blog , IVR Design