How can you better hear and engage the voice of your customers? As lean practitioners, we all do so by personally going to the source—the Gemba. Today this approach includes the use of virtual Gemba, such as the examination of large volumes of internet Big Data.
But are you utilizing the Small Data your front line workers experience every day?
What do I mean by Small Data? Let me give you an example. Recently I was upgraded to first class (always a nice surprise) and looking forward to a meal. On the menu card I had two choices: pasta and fish. When I ordered the fish, the attendant apologized that they had run out. It occurred to me that since I was sitting only halfway back in first class, then most of the people ahead of me must have chosen the fish.
I asked the attendant if this is what happened. “Yes.” Could he have predicted it? “Yes,” he said, “it happens on every flight.” Interesting. He then told me that only 3 out of 16 people wanted pasta, resulting in 5 unhappy people in first class. That’s a 31.25% failure rate with their most preferred customers. “What’s more,” he added with a disappointed glance, “I predicted this would happen the moment I saw the menu.”
We continued talking. Apparently the meal distribution used to be 60% protein and 40% vegetarian, but a while back it was changed to 50/50, and he didn’t know why—was it a cost decision? He was frustrated that many of his customers don’t get what they want, and he regularly apologizes, while feeling powerless to do anything about the problem.
Is anyone listening to the voice of the customer in the menu planning process? This got me thinking… How would you approach this from a lean perspective? How could you reduce this failure rate and improve customer satisfaction for your most preferred customers? What sort of Small Data could your frontline staff gather (simply, quickly, and at low cost) in order to understand problems and improve customer service and satisfaction?
Small Data can be found everywhere, with every customer transaction and interaction. It is derived from a small population that resembles the larger population and is valid for use in small experiments (kaizen) to drive improvement and spark innovation.
Lean startups are familiar with using small data and rapid experiments. A new software product with a small user base can be tested, one experiment at a time, observing user behavior as they respond to each change. Can a large, established enterprise think and act this way too?
Of course they can. They can start with the point of frequent customer interaction, the Gemba. This is why a customer call center can be so valuable, yet many companies (especially big ones) treat it as a cost center, staffing with lowest cost individuals who are incented by call volume statistics rather than meaningful customer interactions.
In the case of the airline, they could start by selecting a small group of experienced flight attendants, and ask each one open-ended questions: what challenges do they hear from customers? What do they wish they could do better? Then they could listen for the common themes or patterns that emerge. Next, they could work to validate these discoveries by sampling a larger pool, this time in a more structured format that enables them to quantify and visualize the data.
This process should help identify the key question that every lean practitioner should always ask: What is the problem and why is it important?
The key lesson with Small Data is to help everyone pay attention to the little things and make evidence-based observations whenever they can. This means helping frontline workers learn to see waste and spot opportunities for improvement and innovation. It also means helping managers and supervisors to be more attentive to frontline workers who experience those interactions every day, and make a point to become better listeners. It’s a win-win-win effort. Small Data. Small Investment. Big Value.
For more on listening to the virtual voice of the customer, see the video of Steve Bell’s presentation at the 2012 Lean Transformation Summit.