Micro-surveys are super popular right now. Each time we jump on the phone with a customer service agent, purchase something or visit a new website we can expect to be asked to take a survey. In an attempt to get more respondents marketers have shortened some of these surveys to a single question.
A single question survey however can be very dangerous. Consider this…
Last week my business credit card company thought they saw a fraudulent purchase and they suspended the credit card (without contacting me). At the same time my online accounting software tried to bill the monthly fee, found no active credit card and then, as is their policy, THEY suspended the service. Additionally there’s some quirk in the accounting software’s website that makes it impossible for me to click to the right page to resubscribe.
The account software company only has email support and the representatives do not speak english as their primary language. They keep sending me the list of steps to resubscribe, but those steps don’t work because of the bug in the system. I’m starting to feel a bit frantic because we are passing quarterly tax time.
With that background I am finding it rather difficult to answer the one question survey that just came from the credit card company.
As soon as I learned that my accounting software couldn’t bill the credit card I called up the credit card company to find out what was going on. The representative asked me a few questions, told me about the suspicious charge, which wasn’t actually suspicious and immediately turned the card back on. She was pleasant and efficient.
A few days after the call they call they sent an email asking:
Based on your recent call experience, if a family member, friend or colleague asked you to recommend a credit card, how likely would you be to recommend us?
I’m not sure if I would recommend them. Not because of the call with the customer service agent, which was fine, but because of the bigger problem they caused me by not telling me they suspended my credit card.
Let’s say the credit card company did a similar thing to 25% of the respondents to the survey and all of them said they would not recommend. What would the credit card company do? Would they fire the customer service rep? Would they spend more money in customer service training? Would they send us a valuable promotion? They think they know why I wouldn’t recommend them but they don’t.
Micro-surveys can give us inaccurate results. We expect respondents to interpret the question the way we interpret it. We suspect we provided the correct forced choice answers. We get so focused on our question we forget about the broader context in which members are making decisions. Like my example the background is complex and muddy. The accounting software is in no way related to the credit card company but that experience is intertwined in my mind. It is hard for our members to be objective, especially when they are make a snap decision to select what seems like the right answer at the time.
The credit card company would be far better off not asking their one question survey (unless they need a fictitious reason to fire a bunch of customer service representatives) and instead figure out how to train and encourage their customer service representatives to ask the right questions, understand the answers and elevate key issues to get fixed.
Let’s get really clear about our project goals and then our research goals so we can make sure the method we choose helps us get accurate answers.