David Kluskiewicz

Avatar

A Complete Customer Loyalty Survey in (almost) One Question

Bain fellow Fred Reichheld has a theory on how to measure customer loyalty. By simply asking, “Would you recommend us?” you can, with reasonable accuracy determine whether customers are buying in or shopping around. It’s called Net Promoter Score.
Although this survey method may appear too simple for a lot of organizations, it may make more sense than “boiling the ocean for a cup of tea.” Well developed surveys can produce helpful data, but in the age of consensus, they can take an eternity to approve and execute, rendering them inefficient.

What is especially useful about Reichheld’s method is its repeatability. We can often learn more about something like customer loyalty by examining how it changes over time than we can by trying to analyze every little facet of it at a single point in time. If organizations could get even a few people to answer this question (and the follow on, “Why?” if their answer is 8 or below) regularly, they could discover customer satisfaction problems well in advance.

Tufte Thinks We Present Too Simply

Edward Tufte, masterful critic of design and skilled creator of graphs and charts presented to me and 350 or so other people at the Omni New Haven. Unlike reading Tufte’s work, hearing him speak reveals another side - someone who is infuriated with bureaucracy and someone who believes that we are much smarter than we think.

As he does throughout his 4 books on analytical design, he extolled the virtues of artfully rendered charts, graphs, diagrams and lists. He described each of the 9 principles of analytical design, which could be used by anyone to create “workaday” presentations. The principles really are practical, and if applied, more intelligent conversations might be generated. The basics principles are:

  • Make comparisons.
  • Show causality.
  • Show multi-variate data.
  • Document everything, and tell your audience about it.
  • Completely integrate words, numbers and images. They are all important.
  • Serious presentations largely stand and fall by relevance and integrity of content.
  • Show information adjacent in space, not stacked in time.
  • Use small multiples.
  • Put everything on the universal grid.

Some of these seem obvious, but when you look at most data (i.e. charts and graphs), you’ll see how many of these rules are ignored. Bottom line: most data are presented to support a point, not solve a problem.

Then, came the interesting parts.

Tufte advocates the use of sparklines, a combination of words, a number and a graph that fit in the same space as an average length English word. He suggests that current methods of organizing data by their “accident of production” (i.e. powerpoint, word, excel, email) is wasteful. We’ve become lazy by separating paragraphs from diagrams from charts. We need to make some effort to bring that data together to present a cohesive argument.

Tufte described weaknesses of the web. First, computer monitors operate at 1/10 the resolution of print rendering complex images imprecise. Second, because of their size and resolution, they are unable to present data all in a single visual field, a format with which humans work very well.

Working with Tufte’s principles, and checking presentations and reports against them, could help reduce a lot of information clutter and finally spark the conversations that those presentations and reports were meant to invoke.

,

What's on my mind?

Some of this, some of that.