All projects: Gel, Jobs, Gootodo, Games, Uncle Mark, Goovite, Blog, Bit Literacy
Asking customers for what you don't already know
Apr 2, 2008
A friend of mine runs a company whose site sells a wide range of products, all sharing a particular attribute. For the sake of example, let's say his site sells organic fruit only - bananas, pears, oranges, apples, and so on - all healthy and extra-tasty.
My friend knew his company's site could use some improvement. Maybe the navigation could be better organized, or the checkout process could be streamlined, or perhaps the product pages could have a better layout. So he decided to run some customer research.
Most companies at this stage sign on for usability tests, which typically involve a facilitator reading from a prepared script, and asking the user to complete several tasks. "Find a group of eight bananas for under a dollar." "Add to the shopping cart some Anjou pears, and then some Fuji apples."
These tasks can be helpful to zoom in on a known, tactical problem. But they're not very good at revealing problems that the company isn't already aware of. Users are forced to do the facilitator's bidding, and that's that.
To my friend's credit, he decided against task-oriented usability tests and instead ran "listening labs," the non-directed research that my company Creative Good has used, and evangelized, for a number of years. In a listening lab, the facilitator's job is to invite the customer to demonstrate their own actual customer experience with the site. Rather than concocting an arbitrary set of tasks, the listening lab allows customers to use the site "for real," as they would at home or work.
How do you get a customer to use a site naturally? Don't give them arbitrary, pre-written tasks; instead, talk to them first, then build the tasks around their context.
For example: Before using an organic fruit site, the facilitator might ask the customer how they usually buy fruit, online or offline; what their favorite or common purchases are; whether and how the Web plays a part in their purchases; and then ask the customer to use the site to place a common purchase.
As I said, my friend went into the listening labs hoping to get some feedback on his navigation, or checkout, or other tactical elements of the site. And yes, he got some of that.
But what he really learned surprised him. It was something much more basic, and much bigger, than any tactical improvement.
Here's what he told me when I saw him after the labs:
"The customers didn't even realize that we only sell organic! That's the whole point of our site, and they missed it! It totally transformed how we're thinking about our redesign."
This was something he never would have thought to ask in a pre-written question, since he wasn't aware of the issue customers were having. What do you call it when you're made aware of something you didn't know before? I call it LEARNING - and that's what customers bring about, when they're allowed to show the way.
Customer research can be surprising, if the labs are run in the right way. I'd even say that it should be surprising. After all, being customer-centric means being open to learning something from customers that you don't already know.
- - -
See also:
• Four Words to Improve User Research
• Tips on Moderating Listening Labs
I think we should combine both approaches. Using questionnaires have the advantage of getting customer feedbacks on specific/particular topics which really concern you. The feedback would be more specific. On the other hand, just asking the customers and allowing them to air their opinions freely involve some “noise” because the customers might just tell you something not relevant to your questions. In terms of product discoveries, I think the second approach could make a huge difference in that the “noise” might turn out to be a piece of great music
This reminds me of the Customer Service surveys I often get which typically include such questions as:
1. Did the Rep say their name
2. Were they courteous
3. Did you feel they understood your issue
4. Do you have confidence that your issue was resolved satisfactorily
and so forth.
In my experience #1 and #2 pretty much happen. I don't think you need a survey to determine that. I would imagine listening in on calls, a strategy often employed by Customer Service Depts. will let you know pretty quick whether that is happening or not. Generally speaking, at least in those companies that administer such surveys, the Reps are generally pleasant enough.
However, are they *effective*? That is, did they resolve your issue? Questions #3 and #4 get at it, but given they way most companies seem to be set up, such questions are all but useless. While the Rep may handle the call they may not be the person who actually executes the activity needed to resolve a customer's issue. They merely initiate the process whereby it is handed off to the appropriate individual who may be in anothet Dept. altogether, to deal with. In this respect, they are as much at the mercy of that other Dept as the customer is. This breakdown is what I've typically experienced. I've often had to call back several times to get something resolved because someone *else* dropped the ball, and it did not become apparent that this was the case until some time (perhaps several weeks) had passed. Yet ludicrously, it is only the customer service dept that seems to be getting measured.
And given the simplistic questions that are asked, it seems that the company is asking questions about issues they "know" about, rather than uncovering issues with internal processes elsewhere in the company, that need to be fixed that migh improve the overall customer experience.