Bag of Chips (part 2 of 3)


I Want It All, For Free (Questions)

Over the years there have a been a steady stream of 'proven systems' for understanding customer needs. In most cases they attempt to format every question so that it produces numerical, weighted answers.  These answers are then added, multiplied, and weighted some more to produce composite weighted nonsense. There is a better way.

There are essentially three kinds of questions and they relate very closely to the three kinds of surveys discussed in the previous section.  Each question is asked differently, the data is collected differently, and the results are collated differently.

Question #1:  Open-ended questions:  In these you are asking very open questions and then listening intently to the customer. You want to find out what it is they need from you and your product.  Your goal is to determine the product, the modifications, and the Voice of the Customer input.

  • What kind of terrain do you need this product to cross?
  • What tests might you suggest we could do during our development to make sure this will meet your needs?
  • What problems do you currently have with similar machines?
  • What does the user need this machine to do? To feel like?

Question #2:  Specification questions:  These questions will allow you to determine what the specification value should be. You should have taken the answers to your open ended questions and determined the feature, now you need the spec for that feature (e.g. the machine is self propelled, how fast should it go?).  You need to have done some engineering and understand the trade-offs involved.  You are going to propose specifications and get the response.  You will also propose the alternative that you compromised to get this specification. At this point, there are a limited number of answers (e.g. the machine can’t go over 60 mph and it will cost more if it goes over 30 mph), so you will use a standard set of answers.  We have found these multiple choice answers most useful for specification questions:

  1. I will not buy without this
  2. this is a compelling sales advantage
  3. Interesting
  4. I am indifferent
  5. I would prefer the alternative

Here are some examples in which the person surveyed would be asked to choose one of the five answers listed above:

What are your thoughts on the following specifications for the XRY Industrial Roadster:

  • A drive speed of 25 mph (alternative is 50 mph, but the unit would cost you an additional $5,000)
  • Square headlamps (alternative is round)
  • Black body color with orange roof (alternative is custom color but it costs $3,000 and adds 2 weeks to your lead time)

At this point you are constraining your customer.  They are helping you design their product and they, like you, need to be constrained by the laws of physics and product cost.  Consider an answer to the second question:  I would only buy this if it had square headlamps.  Is it obvious the difference between that and “square headlamps are a compelling sales advantage”?  Consider how an engineering team might use this information in developing the product and making compromises.  Consider the difference between these results on square headlamps:

5 customers will only buy with square headlamps

20 find them a compelling sales advantage

0 just find them interesting

5 were indifferent

20 would prefer the alternative (round?)

Would it make a difference who the customers were and how they answered?  Of course.  Compare this to a typical survey where you ask people to rank the importance of square headlamps from 1-5.  The average is about 2.2, saying our customers are, on average, indifferent. The survey method above shows the true situation where 45 of 50 are far from indifferent.

Question #3:  Configuration Question:  This is the final survey in a major design program and is usually done just prior to the first prototype phase.  At this point you will have completed a significant portion of the design and will have 2 or 3 major directions you can go (if you only have one option, then press on and skip this step, there are no questions to ask).  In these questions you will be constraining your customer even more.  They only have a couple of options, the intent is not to offer the sky and sun, just the earth (you asked about the sky and the sun three months ago in the first survey).

For configuration questions, you will need to show your customer a summary of the options.  These should be simplified specifications sheets with the major specs only.  Don’t give the fuel tank capacity if the capacity is the same for all units (unless, of course, fuel tank capacity is a major spec for these products).  You then ask them to rank each configuration with one of the following answers:

  1. I will only buy this configuration
  2. First choice
  3. Second choice
  4. I will not buy this


15 Minutes for a Jacket (The Prototype Survey)

One of the most crucial steps in creating a survey is conducting a prototype.  We realized this all too clearly after sending 6 teams out to answer a set of questions on our booms (hmmm, important side note, don’t use jargon your audience doesn’t understand, a boom is a large self-propelled aerial for lifting people in construction, some would call it a “cherry picker”).  Each team, consisting of a product manager and an engineer, was sent to different parts of the country, and were given the same carefully thought out set of questions. When everyone returned, we realized that each team had discovered there was additional information, clarification, or follow-up questions required to get the help our subjects answer our questions. Each had addressed this in different ways.  The end result was that we conducted six completely separate and non-comparable surveys. We had to do it again.  Four months wasted, several thousand dollars down the drain, and quite a few people thinking we didn’t know what we were doing (we didn’t).

There is such an easy way to solve this.  Arrange with one of your best and most trusted customers to conduct the first survey with them over the phone.  Get your entire survey team around a conference phone and let one person conduct the survey while everyone else listens.  When complete, thank your customer, send them a real nice jacket with your company logo, and then refine your survey.  Before you leave the room, discuss the short comings, unclear areas, and general fog.  Fix the questions, add pictures where it might be useful, think about the various answers you might get.  Refine the questions and discuss how you will ask each.

This prototype phone conference survey is also the opportunity to train your team on conducting the survey.  The person conducting the prototype should be one of your best, most experienced, and, most of all, most objective customer surveyors.  Illustrate how to ask the questions objectively, avoid selling, and, most of all, listen.  Have a teaching moment at the end (or maybe a teaching hour).  When you start your survey you will have a common set of questions, a trained crew playing from the same sheet of music, and a good customer wearing a new jacket with your logo.


Three is About Right (Arranging the Visits)

In 2005 I visited a customer in order to better understand the market for our light towers.  This particular rental yard had many light towers and was part of a larger company that used our product almost exclusively. I was counting on some good info. When we went in to meet with the branch manager it became immediately evident that he was not planning on seeing us. He was new to the job, came from outside the industry, and, quite frankly, didn’t care.  Blank stares were the primary form of communication, I left with our engineer and sales person completely frustrated.  I hate customer visits like that.

Customer visits are intimidating, unnerving, and take a lot of time.  If they were easy, everyone would do them all of the time.  They are also our greatest opportunity.  They can serve a variety of purposes.

  • Show your customer that you care about what they do
  • Expose people to the customer’s point of view
  • Give a first-hand view of the market
  • Create a passion and urgency within a design or manufacturing team to create what the customer wants.
  • Develop a bond with your sales force
  • Produce relationships to be used in the future
  • ….and get information for your survey

To accomplish these goals, the visits must be conducted by the right team of people.  Objectivity, consistency, and solid information will depend on a team with varied skills and approaches.

The best survey team consists of three people.  Three is probably also the maximum, somehow it just gets real awkward to have four people crammed into a tiny office with one customer.  Here is our optimum team:

  1. The survey professional:  This is the product manager, engineer, or marketing person responsible for conducting the survey.  They have arranged the trip, participated in the prototype survey, and are primarily responsible for asking questions and recording the answers.
  2. Design Engineer or Manufacturing Manager:  This person will be on the team that will bring this project to fruition.  There are two reasons they are along.  First, it is to see first hand the needs they have to meet and to understand their customer. Second, it is to provide a second set of eyes to try and avoid bias.
  3. Sales Person:  A key element in the team, the sales person sets up the visit, knows the customer, and makes the introductions.  They have the relationship that allows true candor in the interview.  A great sales person personally knows each customer you visit, sets up a range of customers with different views and preference, and has absolutely no ability to be objective whatsoever.  You have to love them!

The team travels together for 3-4 days, visits 8-12 customers and gets 5-8 good surveys. They aren’t all good and some will be disqualified.  Others will show exceptional insight and might get special treatment.