Customer Experience Professionals Beware: NPS Obsession May Kill Us All

In this hyper-complicated world of 24 x 7 digital marketing and connected customer experiences, the desire to find a single “silver bullet” – some magical means by which to measure and solve for all problems, has perhaps never been greater.  After all, wouldn’t it be fantastic if such a secret weapon were to actually exist? 

We need only look back to 455 BCE to Aeschylus's Agamemnon and the often cited phrase, “Live by the sword, die by the sword” to understand the perils of looking at the world through a mono-dimensional lens.  Such a grizzly outcome could become our destiny if those of us who recognize both the need for and the mighty potential of customer-centric strategy and actions continue down a path of singular NPS Obsession.

Don’t get me wrong, there is nothing wrong with NPS per se.  It cuts to the chase with an extremely logical and outstanding premise: “How likely is it that you would recommend [brand] to a friend or colleague?”  This is not one of those difficult to track, touchy-feely “customer satisfaction” questions.  It’s not yet another milk-toast approach to understanding customer sentiment.  It’s a great question to ask and in the absence of any other data, no customer experience leader nor marketer should do without.

The perils with focusing on NPS rest in that this single criteria of measurement is dominating most efforts to measure and subsequently improve customer experience quality and value. The very customer experiences that we should be designing, measuring, modifying and deploying with clock-like precision simply cannot and should not rely on the NPS metric alone.

Hard Dollars Vs. Soft Sentiments

We must ask ourselves – when is the last time we put some NPS in the bank?  How often do we cite NPS as an asset in your balance sheets?  Or getting down to brass tacks, can we truly quantify a direct correlation between NPS and the value of our customer base / business and are we able to understand the levers that effect the two with a clearly defined ROI Model?

Nevertheless, as we look at the growing number of survey based CX applications and service offerings bombarding the market, the overwhelming majority are focused on the usage of survey based data and indeed often solely on NPS.

If we stop to reflect on this increasingly survey dominated world (see my recent blog: Beware the Survey Tsunami…) we should recognize that there will always be a net void between what people say and what people do.  Call it human nature, call it hypocrisy, call it left vs. right brain thinking, the fact of the matter is that relying on whether or not a customer declares that he/she will or will not recommend a brand to a friend is a path fraught with (excessive) risk.  

It gets worse…when it comes to effectively scaling the insights that can be derived from customer feedback, the challenges become bigger as survey responses inevitably decline with growing consumer rejection of being surveyed every moment of every day and with that decline extensibility in a true 1:1 manner is impossible. 

Nevertheless, a dizzying array of Fortune 500 companies have recently bought into this survey driven, NPS focus (obsession?) and are using this choice to support critical KPI’s, business objectives, variable compensation and as a result, management behaviors.

Follow The Money

I would suggest that there is a better way.  Let’s be practical and simply start with the money. What one puts into the bank and what one takes out of the bank, whether your job is to drive revenues or control costs, clearly measurable and predictable financial performance remains central to our understanding of performance.

So how about measuring Customer Lifetime Value (CLTV)?  And for the increasingly stressed marketers out there let’s not forget about Marketing Return On Investment (MROI) – I am sure the CFO does not forget when budget time comes around...

Every action, experience, and message that brands create while interacting with customers has the opportunity to either increase or decrease the value of the brand’s relationship with that customer.  I readily agree that NPS is a very helpful metric, an important contributor to measuring value yet it cannot be a silver bullet and the aforementioned gaps and scalability issues are serious ones that need tackling, the sooner the better.

Three Implications For A Multi-Metric CX Approach

To start with, better measurement inevitably will lead to better outcomes.  Increased customer value (CLTV), higher MROI as well as a much more efficient design and deployment cycles for customer experience creation.   This is the “easy” part.

Despite the potential rewards, the challenges are nevertheless significant.  In order to accurately understand and act on the proposed broader selection of metrics there are substantive data, application and organizational considerations that cannot be ignored.  Here below is a simple, high level overview of some of the major issues that we need to address:

Data: If survey / attitudinal data is fraught with flaws which diminish predictability and scalability then much larger behavioral data sets need to become a central part of the supporting customer experience analytics and insight generating platform(s).  However, few, if any, companies have cracked the code on cost-effectively integrating and analyzing these rapidly-growing behavioral data sets that are derived from the omni-channel world their customers are navigating in at all hours of the day.  This remains job one and it requires a very precise understanding of how data should support customer experience goals and underlying strategy – before you embark on one of those budget killing, deadline missing, “data lake” projects.  The questions of which data, when (real time or not?) and what to analyze are critical yet your strategy needs to lead - not follow - this analysis.

Applications:  If you have solved the data set problem - congratulations – but you are not out of the woods.  Speed is now the rule and “real-time” is increasingly cited as the essential standard, the ability to assemble, integrate and analyze the right data sets is only as valuable as an organization’s ability to also leverage those insights in a timely and value creating manner.  This not only requires high degrees of automation; it can also require significant integration.

Most large companies have now adopted one of the many cloud-based campaign planning and management platforms and many of those that have yet to do so are in the active consideration phase.  Therefore, the ability to integrate one’s “Insight Creation” capabilities / platform with one’s “Campaign Planning and Management” capabilities platform becomes sacrosanct.  With very few exceptions (none?) the enterprise application stack is an ever-evolving combination of legacy application, systems and data platforms, each of which will need to work together in order to successfully support critical, customer experience enhancing uses cases including: content personalization and optimization, cross-channel messaging and trigger-based up-sell and cross-sell initiatives.

Organization:  You have solved the data set challenge, you have integrated your systems and applications so that the linkages between insight creation and and insight actionability are highly automated and effective – that’s fantastic.  Well done!  Now comes the hard part.  Organizing your “people assets” to effectively serve your “customer assets.”

Companies that have embraced NPS as the/a key KPI and who are compensating management based upon that score have already taken important strides to get their organizations to think about truly serving customers.  However, not surprisingly, when one seeks to add cost and revenue considerations into the mix and support just two more metrics such as CLTV and MROI, the organizational issues and complexity multiply dramatically.  No longer is “great service at any cost” an acceptable approach.  Quickly the need to differentiate customer experiences and levels of service based upon the current and potential value of the customer in question becomes a critical necessity and drives the need for segment / persona-based planning.  With that a need arises for the ability to produce and update segments dynamically and with that need arises yet another data question.

Conclusion:  Nobody said it would be easy.

Perhaps one of the drivers for the explosion of the NPS mindset is its relative ease of adoption.  Yet in a world where performance requires accurate, objective measurement of those variables which truly impact value combined with razor-sharp precision, reliance on NPS alone could deliver you the sharp side of a dangerous sword.