Follow

Employee Net Promoter (eNPS) scores


An Employee Net Promoter Score (eNPS) is an employee focused version of the Net Promoter Score (NPS) popularized in customer and market research. An eNPS is commonly derived from a single question that asks whether people would recommend their organization as a place to work. This type of question is a common component in many Engagement measures including our own Employee Engagement Index. This means that while we don't recommend using eNPS as your only outcome measure, or any other single question measure, we can provide an eNPS measure for customers who have a need to provide stakeholders with this (you can ask your coach for more information on turning this on). 

Here's some more detail on our thinking, research and approach to eNPS.

1. How we calculate our eNPS score

Most customer oriented NPS research uses an 11-point (0-10) scale (more on this below). However, we use 5-point Likert response formats in our surveys for better usability (and industry benchmarking). Our own research (also others, e.g. here and here), has demonstrated the validity of a 5-point based eNPS calculated as follows:

  • Promoters = percentage of strongly agrees (5's)
  • Detractors = percentage of strongly disagrees, disagrees and neutrals (1-3's)
  • eNPS = Promoters - Detractors

2. Why the 5-point eNPS and not the 11-point eNPS?

Fred Reicheld (originator of the NPS idea), acknowledged that different scales could be used and even provided a case study using a 5-point scale. Additionally, our research suggests that whether we use a 5-point or 11-point scale we are measuring exactly the same underlying feeling or intention. 

By using the 5-point version of the question we can provide our world class benchmarks for eNPS and we can help you understand how you're going compared to thousands of other companies.

Smoothed density scatterplot of eNPS scores using 5-point and 0-10 (11-point) scales. The regression line shows a near perfect correlation (r=.91). N.B. Darker zones represent larger numbers of respondents in the same spot on the chart.

Screen_Shot_2017-08-10_at_6.49.02_pm.png

3. Why the 0-10 NPS scale isn't necessarily the best scale

The customer oriented response scale ranges from 0 to 10 (it's actually an 11 point scale). These types of score 'out of 10' methods were once popular for telephone interviews. Researchers also sometimes used the 0-10 option because some people would mistakenly use 1 as a good score. However, these types of scales are much less user friendly for online surveys and especially on mobile devices. Furthermore, the majority of evidence suggests that fully labelled response scales are the most reliable response method. 

4. The 0-10 eNPS scale or any other single question is never as reliable an outcome measure as a multi-question index

It is a commonly known statistical recommendation that when it comes to self-report scale metrics for an important outcome we are far better off using a multi-question approach (we combine 4-5 questions into a single index for Engagement for example). Even within the customer and market research studies have shown that using multiple questions is more reliable and accurate than using a singular NPS type metric or question. Our own data observations tell us that single item outcome measures result in weaker statistical outcomes and reliability. 

Reliability levels (Cronbach's Alpha) shown as the number of questions in an index is reduced. Reliability below .70 are often considered questionable. NB It is not possible to even calculate reliability for single item measures. 

reliability_chart.png

In conclusion

We recommend that people using eNPS, for whatever reasons, still use Engagement or other multi-item outcome measures for the primary analytics in their surveys. These measures will be more reliable over time (i.e. not move around as much due to random noise), provide stronger and more reliable statistical signals in analyses such as driver/impact analyses and have stronger relationships with turnover metrics etc. 

 

 



Was this article helpful?
3 out of 4 found this helpful
Have more questions? Submit a request

Comments