CSAT, NPS, CES, VES - What are your thoughts?

We've heard at past TSIA conference keynote sessions that Net Promoter Score (NPS) is "dead".

For Customer Satisfaction (CSAT), TSIA has indicated it is still a viable/reasonable metric to track.

What about Customer Effort Score (CES), Value Enhancement Score (VES)? Are you tracking these as a main or significant metric for your Support organization?


  • Alexander Ziegler
    Alexander Ziegler Founding Member | Expert ✭✭✭
    edited September 2023

    Interesting. I am not sure why somebody thinks NPS is dead - research numbers show that more than 50% of companies are using NPS and this number is still growing (not speaking about Training).

    Let me focus on the Training industry with the next lines: I think the biggest fact is that there is lots of research in certain industries on its success, and once there is lots of data the industries are more or less completely converting to NPS as they have proof of its value. I have overseen various research initiatives around NPS in Training, and we're still at the beginning of getting reliable data in our Training area. But: There is research with thousands of participants in long-term studies in the Fortune 100 space in Training that suggests that CSAT is good and working - but NPS can be superior dependent on the situation and be either valuable as an "add-on" or even as replacement. It is important to mention that most research I'm aware of that indicates that NPS is working usually suggests using the number of NPS as an indicator but also adds a question that outlines the "why". Data then indicates that the pure number of NPS is an excellent indicator for problems - and then the information from the open question helps to identify problems.

    From general research outside of Training most research comes to the conclusion that CSAT is good, and some research is putting CSAT seriously above other measures so that trends are really not suggesting to use any of the other measures mentioned above except thinking of NPS as an alternative.

    I have recently started a research project together with a team of psychologists to evaluate if NPS at all works in the training industry to build a foundation that we're all missing in this discussion, meaning we compare the NPS results against other questions and also take psychological aspects into consideration. The survey was sent to more than 10k students and just closed this week, so I expect to have later this year a general indicator that either supports the usage of NPS in Training or not. There is already research suggesting NPS can be stronger than CSAT (see above), but this research did not ask the general question if NPS is suitable for the Training Industry at all, so my team and I were worried that we missed out on the foundation.

    In the case this new research would indicate that NPS generates valid results I would be interested in working on a broader approach with this group to do work around what the right NPS numbers are for the Training industry. As we cover in TSIA hundreds of thousands of students we should be able to provide some really valuable data in this case.

    If anybody is interested in more details please ping me. I am focussing at the moment on the Training area as I think there is still lots to do.

    I am not sure if the initial question was meant to be NPS in general, but I would be surprised why anybody should state NPS is dead as it is according to research growing. If you need any help in a special area (support, services....) we should look into this in detail.

  • Carlos Alves
    Carlos Alves TSIA Administrator, Moderator, Founding Member | admin

    Hi @Neal Hatton ,

    NPS is not dead, but the issue is how many companies have been using it. Companies often do not apply it correctly, rely only on NPS metrics, and cannot go deeper into root causes to create effective action plans.

    You may refer to this blog entry and TSIA report "State of Customer Success 2023," where NPS usage is discussed.

    Great comments, @Alexander Ziegler ! You got me curious to know more about the study you mentioned.

    @DarleneKelly , any thoughts?

  • Alexander Ziegler
    Alexander Ziegler Founding Member | Expert ✭✭✭

    @Neal Hatton , yes - NPS is used by more any more companies, but I see lots of wrong usage: NPS works especially well as indicator in industries where an industry average already exists - but then companies still need to implement always an open question to be able to dig into the root causes for a high or low NPS compared to the industry average. Details how to use NPS to improve quality beyond CSAT can be found in this research.

  • HI Alexander, wondering if you can share already some high-level insights from your study results?

  • Alexander Ziegler
    Alexander Ziegler Founding Member | Expert ✭✭✭

    Hi, yes, of course: IBM belongs to the 60% of Fortune 1000 companies using NPS in lots of areas. We introduced in Training NPS 10 years ago, when our PSAT and CSAT were on a highest level, but we felt we need to improve, especially as we moved to a partner led model at that time. After 18 months we could see 1 partner with 15 points lower NPS than the others, they looked to the comments and saw that their pdf format they used was not appropriate - and after resolving their NPS moved to the normal level. And this is just one example where the combination of the pure number PLUS the comments is really strong to adjust challenges. And it is simple as the Clients tell you what bothers them the most once you add the question "Tell me why you rates us like this". So, we are using in IBM also beyond Training the NPS plus an additional question, and we everywhere create actions based on the comments - and this works. Do not think that it is around the number. The number helps and is an indicator - but the key is to read the comments.

    Besides this I did with thousands of evaluations an A/B test by not using the 0-10 scale but using just three options "Would you recommend / I am neutral / I would not recommend" as some people said the scale is not fair. This test showed that the result was (with a small deviation) the same, meaning we could see independent on the scale where the problems were and people still commented in the same way.

    So, I can only highly speak about NPS. If you have any detailed questions please feel free to also ping me directly.