Tuesday, November 18, 2014

Are You Putting Marbles in a Bowl?

Image courtesy of Pixabay
Are you listening to act - or are you just putting marbles in a bowl?

Probably the most important component of listening to the voice of the customer is acting on what you hear. In order to do that, we must first optimize how we are listening.

What do I mean by that?

When we ask customers for feedback, it's imperative that we make the most of that conversation. I'm specifically referring to surveys, but I suppose this could apply to other listening posts. We must ask questions in a way that gets us the information we need in the clearest, most-detailed way possible. We can't improve the experience if we don't know what's wrong. We can't coach our employees if we don't know what to coach them on, nor can we praise and recognize without knowing what or why.

I recently attended a conference where, after some of the presentations, attendees were asked to rate the speakers. In order to do so, we were told to use marbles; as we left the room, we could pick a colored marble that matched how we felt about the session: green for spot on, yellow for OK but missed the mark, and red for not so much. After one particular session, I saw that the bowl contained quite a few more yellow marbles than green ones.

This got me thinking, as these things often do.

How on earth does this tell the organizer how or why this particular speaker or presentation missed the mark?

Listening is great, but listening without understanding is pointless. Marbles might tell us sentiment, but they don't tell us why. Using marbles might be a creative way to measure performance, but that's all it is. It's not insightful at all.

That brings up a few important points to remember when you're designing a survey:
  • Worry less about how it looks or how fun it is and more about what it will tell you
  • Ask the right questions; ask for understanding
  • Probe for details; don't just focus on that "one number"
  • Always offer a text box that allows respondents to provide feedback in an unstructured way
  • Don't focus on the metric; focus on the customer and how to better the experience
  • Assign an owner to each question and hold that owner accountable for actions on that feedback
  • Ask questions in a way that ensures the feedback will be actionable
Any initiative to improve the customer experience will be unsuccessful without understanding the customer and his needs. To do that, we must have the right data at our fingertips.

Want more tips on survey design? Take a look at this post: 22 Tips for Proper Survey Design.

Statistics were magic like this: they could tell you with near-certainty that a thing would occur, without a hint of when or where. -Hugh Howey, Shift


  1. Hi Annette, it was great meeting you at that conference and I agree with your comments and the importance of being to understand what you are measuring. Otherwise you are just gather data that you can't apply.

    My one concern is the statement you make, "Worry less about how it looks or how fun it is and more about what it will tell you." That got me thinking as well! My UX designer and usability experience says that your survey has to be usable across all channels and easy to complete. I have observed a lot of survey designs that immediately makes me close the browser and say, "Yeah, I am not doing that."

    For me, your survey is a customer interaction that you should consider in "look and feel" because that experience can have a duration of 10 minutes to complete the survey. Your survey design may determine if that customer will ever fill out another survey for you again.

    1. It was great meeting you there, as well.

      When I wrote that line, I knew I'd get in trouble for saying that! I'm definitely not discounting the look or the usability... after all, surveys are another touchpoint in the customer experience. To make my point, though, I do think we need to focus first on what we're asking, why, and what we're going to do with it... and then build it to look and work to ensure a solid customer experience.

      Annette :-)

  2. Hi Annette,
    I wonder what would happen to the results/ratings if we looked at the style of the presentation given and then separated out the participants by learning style (e.g. VAK)?

    Do you think we pay enough attention, if any at all, to learning style in survey design?


    1. That's a fair question, Adrian. I agree with your point about different learning styles... which makes it a challenge for both speaker and audience.

  3. Fabulous post Annette. That is precisely why I think NPS scores and vendors offer little more than snake oil.

    1. Thanks, James. There's definitely an approach to doing it right so that you can learn something or get something meaningful out of it. Unfortunately, so many just ask to check a box rather than putting some science to it.