Welcome to the Listening Post!

Jeffrey, 
 
Thanks for this. Clear, clean, and concise. I am glad you talked about the verbosity in the second example, I don't have to. 
 
I am still sitting on the fence on using propensity to recommend (or NPS questions as clients call them) at all, but I do so when coerced (we will only hire you if you do it). 
 
And in those cases, I only use it for correlation - not to report directly (yes, they can do whatever they want after I leave - right?) 
 
I am still looking for the convincing element to put me over the fence to using it without caveats, but have not found it. 
 
Will welcome to hear any of your other readers perspectives and what they re doing that works without problems. 
 
Thanks 
Esteban

I think the open-ended comments you provided highlight one of the key weaknesses of NPS, and the alternate scale offers one possible way to get at better data. Another problem with NPS is that the collapse of the scale to three categories introduces measurement error and reduces precision and predictive power. There's lots of "noise" in it.  
 
What we need to be thinking about is what can drive changes in customers' behavior and get them to recommend? A satisfying customer experience is the key. We need to quantify the relationships between the drivers of satisfaction, satisfaction, and desired behavior like recommending a product or company. Then we can prioritize efforts to move the needle and measure our success. 
 
 
 

NPS asks a simple conditional question. It asks would you... The respondent can interpret this any way they want. e.g. I would if my company let me. I would if I were asked to recommend. etc. 
 
One company I know dealt with the dilemma of getting value from NPS this way. They first ask the NPS question. Then they ask follow-up questions based on the response(s). Detractors are asked what makes them hesitant to recommend (the company reaches out to each of these customers personally to try to rectify any issues). Passives are asked no follow-up questions. And promoters are asked "Will you recommend us/our product?" This simple question permits the respondent to reply with "yes, no or maybe/need more information".  
 
In the case of the company I mention (B to B), they have successfully increased reference accounts by over 20% in less than 3 months by using the NPS responses to generate "leads" for references. 
 
Hope this suggests another way to use the score.

Andrew, 
 
There are a couple of problems with your statement. 
 
You say that a company you know ask the NPS question, then asks follow up questions. This is not the NPS method - they are likely asking similar questions to what NPS asks (would you do biz with us again?) but not the same. See, the methodology for NPS calls for the "golden question" the only question that needs to be asked to determine everything. Then it follows by cross-referencing with other metrics (for example, promoters spend 30% more in the first three months of each purpose that non-promoters. There is nothing wrong with asking the "golden question" with other questions - it is juts not the same concept. However, you highlight a very interesting point that NPS detractors have been making forever - how can you have a single question determine the entire outcome and evaluation of your organization? It preposterous. And that is what happens with NPS - management is so excited that they have a single metric that shows the health of the business, they forget everything else. And quality suffers, and relationships falter, etc. Either use NPS (which I don't recommend) for specific groups where promoters matter, or don't use similar questions focusing instead on your needs and demands from the system. 
 
Interesting take on it, though.

Have you ever considered that the follow-up questions might be the issue rather than the NPS question?

Sorry for the delay in responding, Guy. Hectic week. I used the example of these follow-up questions because it was a survey that I could share the results of. In other studies, I have seen comments to a generic open-ended question ("Why did you give that rating?") that don't match the NPS interpretation of the scale.

Does telling our customers how the NPS scoring is done prior to giving the question, invalidate the score because the students now know which scores to pick and which to avoid depending on if they want to support or not? i.e. "Explaining how 0-6 are detractors, 7-8 are not used, and that 9-10 are attractors.)

The difficulty with using NPS for benchmarking is that because it is so easy to use, many organizations adopt it and modify it, yet report their score as if it were consistent. I would say that your instructions will definitely result in you getting very different answers from anyone else using NPS. I would not include those instructions.

I have to agree, and would add the additional caution that by including an explanation of how the NPS scoring works, you are essentially telling them how they 'should' respond.

Gary - I would not advise dismissing 7-8 scores as 'not used'. It's true that many firms look only at the 'Net'/ part i.e. the promoters - detractors percentage, but our consultancy advises using each of the three scores individually as well. For example, scores of 10 / 80 /10 give you an NPS of zero, but so does 40/20/40. Yet the customer experience of the two is telling you a very different story. The latter shows inconsistency, whilst the former suggests a lack of distinctiveness and/or notability. 
If you want to get the best out of NPS, esp. what is driving the NPS headline, you need to run qual research at the 3-score micro-level. 

It seems the biggest misunderstanding about NPS is the idea that it is the only question or metric that matters. If one believes NPS is intended to be the only measure, he is surely going to fail. NPS is intended to be your anchor question when measuring the health of a company's relationship with its customers. Underneath the question, an organization must understand what causes customers to recommend or promote. As the business gains a clearer understanding of what generates promoters or creates detractors, it is able to adjust its operations or offerings to provide the experience its customers seek. A good NPS program will be able to prove these correlations through direct links to revenue. In some instance, a correlation will NOT exists. Where this occurs, the company must find a better measure such as CES or perhaps CSAT. NPS isn't the end all, be all for all companies.