Think Customers: The 1to1 Blog

I think the NPS concept is great for its simplicity and intuitive nature... however, I'm wary of anything that claims to be the "ultimate question."

There is no magic bullet metric - linking customer experience to attitudes to actual behaviors is tricky and time-consuming.

I've elaborated my thoughts on this subject at our blog

Zach Conen

We have used the NPS rating with some hesitancy and have concentrated more on analyzing and promoting customer advocacy. In our geographical area of smaller cities and towns in western Canada, we have traditional loyalists, those who deal with our offices because their parents and grandparents did before them. They are loyal to us because they always have been, not necessarily because they have thought of the value we in particular bring to the transaction. They rate us highly, and likely will swing the NPS rating to a level where we forget our blue ocean and rely on incomplete information. It is this group that will also be loyal to others through habit rather than emotional connections, and when our blue ocean turns to red they will question that loyalty. Loyalty must be for the right reasons.

We also find that behavioral patterns effect the NPS rating. If our group of respondents are introverted technical types, few will likely become advocates telling their friends how much they like our services. They may personally be very happy with their touch points with our company, but they behaviorally will not be preaching the good word. If however they are weighted to extroverted creative types, we have a bunch of sales people ready to tell whoever will listen.

We would rather save ourselves general effort and identify the advocate personality and type through their point of contact in our local offices, and reward the advocates with points and give-aways for referring new business. And all customers for staying with us through several other services.

Do we do customer focus groups and customer surveys? We do, and we find very valuable information coming back. But we also balance customer response with market samples as the type of customer that volunteers to respond is already - you guessed it - an advocate, or is a "body for hire", someone with some time to spare that will fill out a survey either on line or in person for cash. Neither are really representative of the average customer. In 2006 we engaged Ipsos-Reid up here north of the border to do a general survey of typical consumers of our products and services, and we learned a lot more by combining these two results. But the most important thing is to do something with those results. We discuss focus groups and surveys with other businesses, and many have done these, particularly the focus group side. What surprised us is that so few had acted on the results and little had changed in spite of the process and the findings. Perhaps it had to do with change itself, but at the end of the day the customer provides both the direction and the authority, and neither should be ignored.

Fred Reichheld's work on Net Promoter Score is a powerful companion to Don and Martha's Return on Customer; both are readily understood across the corporation by senior and lesser executives, both are dynamic and respond quickly to improvement initiatives, both get the bean counter's attention, both can be translated to department programs and both can be explained in an elevator pitch!

From my experience in over 30 blue-chip customer projects, such concepts are like riding a bike; if you wait till you know how to do it perfectly before you hop on - you never ever learn. Early adopters of NPS and indeed ROC derive initial benefits, wobbly at times, but gain a significant competitive advantage as they adapt and customise the processes to achieve their own objectives. Years later, processes are entrenched in the corporate psyche while competitors bodily announce they are adopting a "proven" methodology.

I'm exploring some interesting combinations of NPS, ROC, Hoffmyr's Conversion Model and Touch point Mapping to leverage the power of four different models that achieve a holistic set of metrics rather than single facets. Have others worked with this combination?

Everyone makes very valid points here.

I'd like to make another point. I like to make the results to the "ultimate question" actionable by taking those who said they are very likely to recommend and invite them to be a company evangelist. How you that customer segment (and vet them) will vary based on industry, customer type and more but the potential to engage your biggest advocates even further is very powerful.

I've done this successfully at a pure play dot com (community) and other companies.

This takes the ideas from the book, "Creating Customer Evangelists" by Huba and McConnell (an awesome book by the way) and combines it with the ultimate question.

Happy to talk with others about this.
Kim Proctor
Customer Relationship Consultant
Customers That Click

Here are a few comments on NPS.

1) It’s a fantastic tool for getting companies to take customer service seriously. Because its easy and simple to understand NPS has a great advantage over almost all customer service tools which senior management don’t understand and therefore don’t trust.

2) Like most catalysts of change NPS starts the conversation it does not fix everything. Once the door has been opened tools to fill out the picture are much easier to introduce.

3) I use an NPS style approach along with other questions in customer surveys but I also introduce a related question into the employee questionnaire along the lines of “How likely is it that customer’s would recommend us to a friend/colleague?” Looking at the question from both sides often throws up some interesting differences that need to be further investigated. I’ve found that asking employees to state why they think customers would recommend the company to a friend usually generates some surprising responses too.

John

Ron's suggestion that companies focus on expectations is deceptively appealing.

It is appealing to ask customers what they need, want and expect. But it is deceptive to expect them to be able to answer easily and for you to make sense of what they say.

It is for this reason, and despite more than a decade of involvement in service quality/customer satisfaction research, that I have stopped asking about them. Instead, I ask about the discrete outcomes that customers are looking to achieve. This is of course only one of a number of things that you should survey customers about.

Take a look at the Strategyn in the US and OMD in the UK for more on outcome-driven innovation.

Graham Hill

The "promise of actionability" that Michael Lowenstein refers to is an empty promise of NPS, if a firm ONLY asks the "will you refer us" question.

I believe a better line of questioning (sorry for the negative connotation) for customers is around their expectations -- what were they expecting from the firm/product/relationship, and how well is the firm meeting those expectations.

Granted, you have to do a little bit of the hard work to prompt customers for what those "expectations" are (i.e., you can't make it an open ended question).

But the "simplicity" of NPS that so many people allude to is a liability -- not an asset. Why are accepting something that isn't actionable, useful, or right just because it's easy?

Net Promoter's chief attractions are its simplicity and promise of actionability; however, when the technique is considered from the perspective of good, defensible customer research science, it has multiple serious methodological flaws and interpretive challenges. I've written an article on the subject, which is available to anyone on request directly from me; or it can also be found on www.crmguru.com, where I am a member of the Advisory Council.

Michael Lowenstein, CMC
Vice President and Senior Consultant
Harris Interactive Loyalty

John

Your post, and Mark and Linda's comments raise three important points.

Firstly, most companies do not do thier homework when it comes down to understanding customer behaviour, what influences it and what they can do about it. And many companies are very reluctant to hear the truth in case someone becomes responsible for doing something about it.

Secondly, as the British economist John Kay points out, the vast majority of so-called customer 'data'is soft data. It comes with a long list of caveats attached, all of which reduces its predictive accuracy and thus its usefulness. A measure like Net Promoter is about as subjective as they get.

Finally, innovation takes place on the boundary between sinplicity (where nothing really changes) and chaos (where things continuously change). (That's why free-markets are the best sources of innovation in general.) As the history of innovation shows, unless a company is in this 'complexity zone', they are probably not going to be all that uncomfortable and thus, all that innovative.

Graham Hill
Independent CRM Consultant
Interim CRM Manager

I think John Gaffney's comments are spot-on-perfect. Way too many companies look to have their ears tickled by customer feedback. Ask your customers if they will recommend you for x, y, or z, and if not why. All aspects of the business must be allowed to be evaluated by the customer "to recommend" including email content, timing, and relevance. Also needed to be examined are products by products, total collections, packaging, presentation, pricing, branding, communication, shopping channels, shipping, delivery, customer services, and surveys. Even if customers "recommend overall" there are likely aspects of the business that need to improve; hence the need to examine all products and customer touchpoints. Emails are one such critical aspect that needs examined. Do they speak to the customers' needs (and wants)and schedules. Do customers recommend that others receive them? What is driving the resounding yes? What is driving ANY hesitation?

Are there examples of your organization using survey results to ask tough questions of your customer base, and what did you learn from those challenging situations?

Ah yes, that "element of discomfort." That's a real show-stopper in many companies. Why? Because the culture doesn't support the idea of being uncomfortable as a necessary stage of learning. When the only metrics that are taken seriously are end results (revenue, leads generated, etc.), there isn't really a space for learning. Without learning, there is no innovation. But, guess what, to learn you generally have to be willing to be UNCOMFORTABLE.

Check out "Love Your Mistakes" and "Get Caught with Your Assumptions Down" in my free ebook at http://www.fordbusinessconsulting.com/ecommerce/ebook1.html.