Daily Archives: October 22, 2016

Lean Startup Leadership and Net Promoter Score (NPS)

Leading an integrated effort to win market share through NPS increase and Lean Startup leadership to rapid prototyping and validated learning


Fortune 100 companies have been using NPS increasingly to assess customer loyalty and following through with task forces to implement possible NPS impacting projects. These efforts rarely lead to the expected results. The long lead time between the measurement and the completion of the projects lessens the overall effectiveness of the impact. In addition, the NPS impacting projects themselves are rarely ‘scientific’ – many of the forecasted positive impacts are not realized. The task force in itself, led by a busy senior executive is constrained, thus as time passes, reducing the attention to the effort.

With our clients we have been achieving great success applying Lean Agile / Lean Startup leadership to NPS projects, alleviating the challenges and leading to breakthrough positive NPS impacting solutions.

What we’ve been doing

Applying Lean Agile thinking to impact how customers think about our business can seem unrelated; We’ve formulated an integrated model to help our clients deliver value faster by coupling lessons learnt from Lean Startup leadership and feedback from their NPS (net promoter score) results to achieve faster scaling of the combined method. On top of which, we’ve been using concepts introduced by John Kotter – XLR8.

Let’s start with a few definitions:

What Is Net Promoter? Net Promoter or Net Promoter Score (NPS®) is a management tool that can be used to gauge the loyalty of a firm’s customer relationships. It serves as an alternative to traditional customer satisfaction research and claims to be correlated with revenue growth.

Lean startup is a methodology for developing businesses and products. The methodology aims to shorten product development cycles by adopting a combination of business-hypothesis-driven experimentation, iterative product releases, and validated learning. The central hypothesis of the lean startup methodology is that if startup companies invest their time into iteratively building products or services to meet the needs of early customers, they can reduce the market risks and sidestep the need for large amounts of initial project funding and expensive product launches and failures.

XLR8 – In his latest book, Dr. Kotter explains how traditional organizational hierarchies evolved to meet the daily demands of running an enterprise. For most companies, the hierarchy is the singular operating system at the heart of the firm. But the reality is, this system simply is not built for an environment where change has become the norm. Kotter advocates a new system—a second, more agile, network-like structure that operates in concert with the hierarchy to create what he calls a “dual operating system”—one that allows companies to capitalize on rapid-fire strategic challenges and still make their numbers.

How do the three concepts tie together?           

Traditionally companies start the NPS journey by surveying their consumers. They send out emails, printed surveys, questionnaires and the like, to receive many data points on the NPS spectrum. Some use statistical methods to gauge the NPS of the consumer population; other will survey the entire population. Often the number of respondents to an email blast is around 5-7 percent. That’s expected, however can lead to statistical skewing of results. Many times respondents will take the time to provide written feedback as well.

Traditionally, companies will spend time analyzing the results and the written feedback. Then, they will create a task force to identify projects / methods to impact the NPS score. They will roll out these projects, which tend to be cross functional, and re-assess the results after project delivery.

The process is lengthy!

For a Fortune 100 company the data set can be exhaustive, sifting through the written feedback might require an automated tool and further investment. The turnaround time for merely understanding the feedback can be longer than 3 months, which in today’s competitive climate doesn’t cut it.

The next step is acting upon the results. Often a senior VP will own the task force with the charter of identifying the opportunities to improve the score.

Once more, the process is time consuming and lengthy.

The senior VP is busy, their time limited and while initially there’s much fanfare, the task force meets twice a month at best after the initial kickoff and focused effort has subsided.

The NPS improvement projects that the task force identifies tend to be big, requiring investment of resources and time, often outside the direct command of the senior VP, as well as cross functional collaboration from IT resources. These will be stretched thin anyways and will hardly be able to support the NPS mandated projects. The turnaround time for actual delivery of the project to impact the NPS is six months at a minimum!

This makes the NPS measurement, while worthy as a data point, not an actionable metric for most organizations. Thus, the challenge with NPS as a metric is not so much the actual measurement and the feedback received, rather it is the tendency of organizations to over analyze, try to optimize a solution, over dedicate resources, and move too slow towards a solution.

What can be done differently?

That’s exactly where Lean Startup leadership and Kotter XLR8 approach could tilt the scale.

One of my clients received 9,000 responses to an email survey sent to 150,000 consumers. The NPS was negative. Surprisingly, 80% of the 9000 responses included written feedback – 7200 in total. The next obvious step for my client was to learn from all the written feedback.



How many surveys written feedback do we have to read in order to have statistical confidence in a certain hypothesis we have about low NPS outcome?

The number is quite small, probably smaller then we would be willing to admit.

Approximately 5-10 responses have to be read in order to provide a good enough confidence level.

Wait a minute – am I saying that we can trash most of the written responses? Doesn’t that sound irresponsible? Shouldn’t we read all the written responses? We could, however we would not learn much from reading the N+1 survey that we didn’t already know from reading the N survey. That N tends to be small – maybe 5 or 10.

Think about it this way – when you haven’t read any survey your knowledge is zero. After reading the first response your knowledge increased exponentially (actually it increased infinitely as your starting point was zero). The next survey you read will increase your knowledge, but not as much as the first one, the same is true for the third and the fourth and the fifth and so on. How many of those surveys do you need to read to be mostly confident about what irks your consumers?

Probably 5 or 10, to be on the safe side 20 or 30; but not 100 or 5000 and definitely not 7,200. Also, there’s no need to invest in a costly data collection application.

This sounds counterintuitive – since we have always been trained to analyze all the data points; suggesting doing anything less does sound irresponsible. The main message is that the level of uncertainty is reduced fast and the knowledge increases exponentially. If you wish to read more – Douglas Hubbard, How to Measure Anything: Finding the Value of Intangibles in Business is a great place to start.

Okay, we’ve read 20 surveys and are now confident about some of our customers’ concerns impacting our negative NPS. What next?

According to Lean Startup leadership, our next step is to identify a hypothesis that is a possible cause for the score, and a remediation for the specific issues analyzed from the survey.

Traditionally the senior VP led task force would spend a considerable amount of time identifying opportunities. This is especially true if the analysis itself was time consuming; actually the amount of time spent on opportunity identification tends to be directly correlated with the amount of time spent on analysis. Possibly this is a psychological effect, we aren’t comfortable spending a short amount of time identifying opportunities when the data collection and analysis was lengthy.

The opportunities would then translate into projects that supposedly will improve the NPS. In line with time spent – just as the analysis and the identification phase, the projects that would be identified would be lengthy. But that’s just the easy part of the problem.

The BIG Problem is that most of the projects to impact the NPS will have no impact on improving it! Some will actually impact it negatively!

Generally speaking – we don’t really know ahead of time which projects would actually impact the NPS positively, much like we don’t know which features that we develop would delight our consumers. Read more about it from the Chaos report. Instead, using Lean Startup leadership we identify the most important and uncertain assumptions we’ve made concerning NPS impacting projects. These would become a basis for validation planning. Each validation plan would prove or disprove a hypothesis about impacting our NPS.

Lean Startup leadership of product delivery is based on the Build-Measure-Learn hypothesize incremental solution to impact our NPS score. Essentially, we substitute the long term strategic projects to improve the NPS score, which might or might not impact it, with shorter incremental, short validations that contribute to a growing base of validated learning.

The game changer is the time frame required; analyzing a smaller set of responses, ideating solutions to impact them, validating a hypothesis vis-à-vis a rudimentary incremental prototype solution is much faster than the alternative traditional approach.

So far so good, however, we haven’t solved the problem of the entire process led by a busy time constrained senior VP, yet.

The next step is scaling the process of improving the NPS on an organizational level.

This is exactly where the thinking offered by John Kotter XLR8 helps us.


How do you scale the approach?

We use free agents in the distributed informal organization. In other words, we make the project for impacting the NPS an organization wide initiative. The senior VP led task force will own the vision and the process and will run the first iteration of validated learning through hypothesis validation and prototyping. However it would also drive other teams to journey through discovery-hypothesis-prototype-validated learning cycle.

We empower the distributed teams to follow through following steps:

  • Discovery through their own analysis of a subset of the written feedback in the NPS reports;
  • Hypothesize to identify a possible cause and remediation for the specific issues analyzed from the survey;
  • Prototype a product, solution or result that validates the hypothesis or disproves it;
  • Validated learning based on the results of the previous step leads to endure or pivot.

Essentially we are creating distributed, loosely managed, independent teams using Lean Startup leadership to ideate many possible prototypes supporting hypothesis, leading to validated learning that positively impact the NPS score.

Instead of having a single task force spending a year to deliver a limited set of projects, where only a few would have a positive impact on the NPS; we are jump starting numerous teams, running independent experiments, following through on those validated experiments that have merit and quickly pivoting from ones that don’t. The senior VP led task force provides a vision and resources to the teams where needed and aggregates the individual contributions – but is not managing per se, rather it is leading the effort and provides guidance where needed.

What are the benefits?

  • Duration sliced by a factor – from year long to several months;
  • Time to analyze slashed by following sound statistical methods;
  • Time to a working prototype slashed by a factor;
  • Waste on none impacting solution – removed;
  • Impact multiplied, by shifting from a centralized manage oriented task force, to a distributed, independent, loosely managed structure;

Using the model described, integrating three seemingly unrelated concepts, exhibits innovative out of the box thinking. We received synergetic benefits that augments both the internal organizational leadership and at the same time enhances customer satisfaction through NPS measurement.

If you are facing similar challenges in your organization, embracing the model would lead to great benefits. We are excited to learn from your experiences and how it worked for you, please share…