Set clear expectations

I was reading an article today and one of those ‘take this quiz’ games actually caught my eye. When I was finished the article I had to make a phone call – I knew I would likely be on hold for parts of the call. I thought it would be fun to take the quiz to kill time while I waited.

I started the quiz at Driving.ca. It was fun at first. You had to link logo to car company names. I liked how many questions were about European car brands that I’m not too familiar with. But here’s the kicker, my call was a taking a long time, so was the quiz. I’m 18 questions in… and here’s what starts running through my head: where’s the end? How long is this quiz? If I wasn’t stuck on hold I’d be a bit frustrated at how long it was taking… What’s the point of this quiz? Who made it? Why in the hell did they think anyone would want to play for so long? How much longer will this take? F%*& I’m kind of roped in here now and I do want to know my score – I’m a bit driven to finish.

My call ends. I play a few more questions. I start to look around the screen to see if there are any other indicators of how long it will take. There are none.

I finally abandon the game. I’m actually pretty cheesed off at this point. I feel like my time has been wasted and if I’d had been given better indication of how long this ‘stupid‘ pop quiz would take I could have made a better decision at the beginning to start the game or not.

I take a deep breath and remind myself one of the golden rules of designing a smooth multi-step process: Set clear expectations for the audience up front. Early in my career as an interaction designer I worked at a market research company. We built online surveys. We collected a lot data about participant thresholds – you know how long people will remain engaged in a survey. It’s short, about 12 questions is recommended for surveys where the participant isn’t incented.

We used to always recommend to our clients to make sure to explain the approximate duration in time and exact duration in questions at the begining. We spent a lot time designing ‘progress indicators’.

Despite all our advice our clients often felt the need to cram as many questions as they could into a survey. Often to the point of severely impacting the survey’s abandonment rate. I always used to think that part of the need to ‘cram questions’ was a ‘buyer’s syndrome’. They are buying a service and feel that the more questions they get, the better value for their money they are getting – if you guage value on a questions per dollar scale. The funniest thing is… when we’d be working on these projects and the number of questions had gone sky high, everyone knew there was a problem-including our client. The inevitable solution – suggested by the client – take the progress scale off the survey and remove the text in the instructions that explains the estimated duration and number of questions to be answered.

Our response was always the same… don’t. You must set clear expectations up front, and you have to show the participant their progress as they move through. What’s at risk is your reputation and credibility. You are trying to measure your customer’s attitudes in order to improve your service delivery to them. But by creating a situation where YOU WILL frustrate them. And you ultimately risk losing a customer.