myspace views counter
Close

CX & Customer Success resources

in just about every form you could want!

Search Waypoint Resources

Search phrase

Filter by resource type

-- Article --

Net Promoter Neutral Isn’t

Posted on July 9, 2010 , by Steve Bernstein
CATEGORIES: Loyalty Research
TAGS: 
There’s always debate on how anchoring scales can impact results.  Recently there was a lively discussion on LinkedIn asking if anchoring the midpoint (score of 5 on a 0 to 10-point scale) by labeling the “5” as “Neutral” on the questionnaire can change Net Promoter scores.  Net Promoter methodology finds that a response of 0 to 6 on this scale is a Detractor (i.e. the research found that a 5 isn’t Neutral at all – these people are in fact likely to spread negative word-of-mouth), so it’s important to get this right. 
Labeling the midpoint "5-Neutral" certainly impacts results

Labeling the midpoint "5-Neutral" certainly impacts results (Click for larger view)


In our research we compared 2 statistically significant and similar samples and graphed the survey results for the recommend question as well as the diagnostic sub-attributes.  The top graph shows the results with the “Neutral” label on the midpoint while the bottom graph shows the results without the label.  Without crunching any number, you can see just by looking at the graphs that there certainly are differences in the responses – the “Neutral” label draws respondents to score the 5.

 
What you decide to do with this really depends on your objectives (and many thanks to Matt for a great response in the LinkedIn thread on this topic): 

  1. If your objective is to acquire a score that helps you more accurately benchmark your company against others in your industry then you need a consistent methodology.  Since the original research used the “Neutral” anchoring then this will be important to maintain.  That said, 3rd-party benchmarks are suspect:  they are always changing over time and you can never be sure if you are really comparing apples-to-apples (variations in sample, timing due to seasonality and market conditions etc).  I question how 3rd-party benchmarks can really help you, especially when doing your own benchmark is both less expensive and generally much more accurate and actionable.
  2. On the other hand, if your focus is on improvement and growth then you want to carefully consider the “Neutral” decision.  The research consistently finds that people scoring you a “5”aren’t neutral at all – these people are detractors! – so correctly identifying these respondents for action and improvement is critical.

What do you want to accomplish?  If you want different NPS scores without having to actually change anything in your organization you can certainly use this simple trick.