Background to NPS:
Fundamental to any Customer Experience Management (CEM) program, is the measurement of that experience and the appropriate use of the correct metrics
Let us examine a few components of this including:
- Metrics being used
- Interpretation of those metrics
- Setting of appropriate benchmarks
There are a few golden-rules in CEM measurement, among these are:
- Consistency: the organisation needs a consistent survey framework and system of metrics. Too often we seem programs using different scales in different surveys. Avoid this!
- Enterprise wide: in order to evaluate and measure performance across the group, it is crucial to use the same core metrics across all areas
- Medium Matching: As survey platforms becoming increasingly “Omni-channel”, making sure you focus on the right survey technology for the right customer and touch-point becomes more important. Make sure you are using the channel which gets you the closest to your CEM goals, whatever they may be.
Over and above these three rules, a few other consideration are worth mentioning.
- Understand the limitations that are applicable to all metrics and how to minimise these
- Understand and use both absolute (rate your experience on a scale of 0 to 10) and relative metrics (above or below expectations)
- Set appropriate targets
- Interpret the scores and metrics correctly
Net Promoter Score:
This image describes how NPS works. Remember it is not a %!
Net Promoter score (NPS)
A summary of the method of calculation is as follows:
NPS = (% of Promoters – % of Detractors)
NPS is not without its limitations and criticism. However it has become a widely utilised metric across all organisations as a measurement of customer loyalty. NPS scores are often compared across companies and industries, but some caution should be exercised since those comparisons are often not a like for like comparison.
What is a good NPS score?
There is no simple answer to this question but to keep it simple here are a few pointers:
- Generally a negative NPS score is a bad score
- Any positive score is good starting point
What is of more value than the score (particularly a positive score) is:
- How that score compares across areas in the organisation, and
- How that score changes in time
Sometimes organisation apply different types of analysis to the result of NPS questions. This undermines the correlation study performed showing the behavioural clustering illustrated above. Applying arithmetic to the above model which is not in line with the model developed by Reicheld is meaningless and misleading.
Whenever any proprietary model is used in Market Research, ensure that the model is fully understood and that it is applied in its full form or if an adaptation is used, that the adaptation is logically consistent with the model and that is still capitalises on the usefulness of the model as developed.
Benchmarks and CEM Targets
Unfortunately there are again no simple answers, instead we offer some guiding principles:
- Some metrics or scores are clearly bad or poor e.g. overall rating of 5 out of 10 or a negative NPS score à these are clearly poor and should be improved
- Equally some scores are likely very good e.g. overall average rating of 8.5 or 60% of clients saying expectations are being exceeded
- The more challenging scores are the “average scores” e.g. NPS of 50 or overall rating of 7 à these require more judgmentThe best way to set benchmark or target for CEM scores is:
Look at longitudinal data à are the scores getting better?
Look at the movement across both absolute and relative measures
- Consider the distribution of the scores à an overall average score of 7 out of 10 can be arrived at by a score of 10 and 4 or a score of 7 and 7 à the latter is clearly a more desirable outcome (consistency of service is crucial)
- Evaluate scores across different business units or functional areas to inform appropriate targets
DO NOT simply increase the score by a fixed percentage each year!
This simple guide requires more in-depth discussion; Mail us to set up an obligation free chat if you want to understand more.