I was recently asked:
We feel that our Net Promoter Score might be too high and had read somewhere that the type and quantity of the questions can artificially inflate the score.”
Okay, so maybe you feel it’s a good problem to have, a Net Promoter Score that is too high, but it’s a fair question.
Personally, I have not seen any NPS® research that indicates the type and quantity of questions can impact on the Net Promoter Score. If you have seen some research on this, please let me know in the comments. After all the score is based on the response to the very specific “would recommend” question.
Of course, it is possible to change the response to a survey question based on the questions that lead up to it. This classic Yes Minister segment is a great example of this approach. However, if you try not to skew the results like this then you should be fine.
The Transactional NPS questionnaire that we have settled on is:
- “Would recommend” question
- Why (qualitative response)
- Select themes – a list of checkbox themes that the customer selects to indicate which part of the service/product experience was the cause of their “why”.
The last question is important. The hardest part of the transactional NPS process is making sense of the qualitative content. With even moderate volumes of qualitative feedback it gets hard to tag all the responses with relevant codes in a timely manner. Qualitative feedback is vital to the process because it helps you understand how to change but it can be a burden to manage.
But by adding this last question we have a 3 question survey (high response rate) with accurate tagging of all comments. This makes root cause analysis relatively easy.
We now recommend (and sell) CustomerGauge because it can deliver this style of survey in an automated manner. It also has some good NPS specific analysis charts that let you do root cause analysis from within the tool.
Back to the inflationary question: while the survey questions themselves may not inflate, or deflate, the NPS other factors in the survey can impact on the score:
- Format of data collection: it has been found that different forms of data collection, telephone, internet, paper, do result in slightly different Net Promoter Scores.
- Timing: it has also been found by a variety of organisations that timing can impact on the Net Promoter Score. Generally the closer the transaction that the survey is completed, the higher the score. So surveys taken the next day result in higher scores than those taken a week later.
- (Update – Thanks Tim) Position of Would Recommend question in survey: Research shows a significant position effect for the ‘Recommend?’ question in surveys. In short, you get higher NPS scores if you ask the question first, lower if you ask it last in the survey.
All of the above really leads me back to my standing response to the “what should our NPS target be” question. I hear this question often and in my opinion your NPS target should be last quarter’s NPS plus a bit. While there are NPS benchmarks out there, these factors make comparing scores problematic and so the most important benchmark is your own score.