I frankly am beginning to feel that I’m shouting into a void here. Companies are spending more time and money equipping the CEO and team with information, while starving the thousands of ground-level employees who, frankly, can have more impact on the company’s success simply through their day-to-day actions.
One ray of hope: an article in the December Harvard Business Review (co-authored by Fred Reichheld, the creator of the Net Promoter Score – a simple metric that somehow captures the complexity of customer perception) entitled, “Closing The Customer Feedback Loop.”
As opposed to the conventional wisdom of gathering masses of data and trying to detect high-level patterns in them, Reichheld and his coauthors talk about getting more granular – gathering information at the customer transaction level, creating small rollups of the data, and sharing them where they can do the most good – with the front-line employees and first-level management who directly impact the customer experience.
I agree with their prescriptions, but it still leaves the problem of what to tell upper management. Is there anything wrong with high-level management dashboards? Well, yes. Something of the danger in this is described in today’s WSJ article on Simpson’s Paradox (”When Combined Data Reveal the Flaw of Averages“). The first example cited: while today’s overall unemployment rate is lower than the 1982 level, unemployment at each educational level is higher. (The overall rate is lower because there are more people at higher educational levels, which have lower unemployment, than there were in 1982.) The article states: “Compared with a similarly educated worker in 1983, ‘the worker today has higher unemployment at every educational level.’”
There’s always something lost in summarization. In the case of Simpson’s Paradox, the result of the loss is a flawed conclusion, or at minimum missing a greater point of the story. Overall unemployment today is lower than 1982, but people today have been hit harder than their 1982 counterparts.
Dashboards distort reality as well. Executives rely on machines crunching millions or billions of numbers to present them an easily readable story of what is happening in their businesses. Yet the farther the statistics are distanced from the on-the-ground reality, the more likely they are to lie.
What can be done? Let’s get back to “Closing the Customer Feedback Loop.” On-the-ground data gathering and interpretation by those close to it makes all the sense in the world. But in communicating with upper management, there needs to be less sharing of numbers, and more sharing of individual stories. You can’t get any more granular than that. You can read a vibrant story in a minute or two. And stories fall into patterns–something more subtle and nuanced than statistics–that help senior management understand what’s going on. And human experiences are more understandable than the simplest dashboard.
There are tools to do help you gather and use stories. Rakontu, an open-source story-sharing platform, is one. Enterprise 2.0 tools such as blogs would also work for this purpose. So what’s stopping us? Or am I still shouting into the void?
(Photo by awegedebe via stock.xchng)
Related posts:
GE uses “net promoter score” – one of my earliest posts!
On Rakontu
Time to listen to front-line employees
How B2B customers talk
“Enterprise 2.0″ review
Technology is great, and so is avoiding the acorns
