Survey Design - Don’t foget your audience
Christen Uber September 25, 2007
Recently, I had the pleasurable experience of moving (I am totally being facetious). As we all have experienced, before moving residences, it is necessary to transfer, disconnect and/or reconnect services from one residency to another. All of which are time consuming hassles. Well, after completing my move and acquiring the necessary services at the new residence, I begin to be bombarded with the utility and Telephone Companies calling surveying me about my experience and my customer service satisfaction. One company’s survey process particularly aggravated me both personally as well as professionally. Personally there were many times I wanted to discontinue the call however, because of my professional vocation I was intrigued on the company’s survey design and to what extent they would go to acquire the perspective of their new client - me.
The survey began with acquiring some of the basic demographic information regarding my “purchase”. In my particular situation, I was just transferring over my DSL service from one residence to another utilizing my existing modem, which my provider provided me 3 years ago - so not really a purchase. The customer service representative asked me several questions requiring me to rate my satisfaction from 1 to 10: 1 being very unsatisfied to 10 equaling extremely satisfied. We reached a section where the representative specifically wanted to obtain my opinion regarding the DSL set-up process. I informed the representative the “set-up” process was not applicable to my situation. I had set-up my own modem with the existing software and hardware they provided 3 years ago. The representative proceeded to tell me that “N/A” was not an option and continued to read through the multiple questions which had no relevance in my situation. I continued to reply N/A and the representative continued to tell me that was not an option. Thus, I chose to rate all of the necessary steps a 2. By this point I became very aggravated and voiced my distain and patience to continue. The representative assured me that we were almost done. However, the following questions were again multiple parts with very little relevance to my particular situation. Detailing my specific situation to the representative, the only response I received in return was a response I would expect from an IVR system with no autonomy to accommodate or rectify the situation in a customer service friendly manner. I am sorry but I have to ask the questions just as they are.
My point in explaining and detailing my experience is to remind companies acquiring information from their customers - surveying is not all about just obtaining information and measurable data. Surveying and polling your customers is also an opportunity to acquire the voice of your client. Therefore provide the audience, from whom you want to hear from, the opportunity to voice it. Train the customer service staff, who is obtaining the information, to rate a clients response based on the information provided not just the information you are looking to obtain. Design the survey so the client feels it is applicable to the services and/or products that they actually acquired and utilized from your company. If I had that type of experience, I would have been left with a favorable impression of the company and its follow-up and would not have found it necessary to write this blog.
Posted in Human Resources Analytics
September 26th, 2007 at 1:49 pm
You present a very interesting blog, Christen. How do you design surveys with some form of A.I. to adapt to the answers. Do sections of the survey begin with yes/no questions to trigger automatic delivery of follow up questions which are relevant and applicable to the subject. e.g, “Did you buy or lease a modem from us?” Yes / No. The A.I. waits for an answer… “Yes”… A.I. tells the survey to open up the qualitative inquiry set of questions pertaining to the experience of buying or leasing a modem. Follow up question - “How would your rate the delivery time of your puchased / leased modem?”…”How satisifed are you with the setup of your purchased / leased modem?” The A.I. would also enable or prevent future yes/no and qualitative questions based on previous yes/no answers. With simple logic and a little artificial intelligence surveys can shape and reshape based on the customer… You still gather data on the experience, but the drill down is richer… now you have the ability to segment your “previously owned modem” users from the “purchased/leased modem” users and their experiences with the services. Imagine telling the VP of Operations or Sales that 45% of our customers are “previously owned modem” customers. Wow - business strategies shift… “Can we now send these customers coupons for a discount on new and expensive modems so they can replace their slower/older modems with lightning fast, fire wall enabled super modems,” they might ask… It’s a survey to gauge what you want and need and feel… but it’s how they use it that makes them a superior business or just a run of the mill internet provider.
September 26th, 2007 at 1:56 pm
UPDATE-
The A.I. is a “Basic Skip Logic” embedded into the survey for the exact purpose of enabling or preventing question sequences which have no relevancy.
See Survey Monkey’s Write Up