Design

The Danger of Asking Users

It is well known that a successful product or a service must resonate with its users – that is what user-centered design (UCD) is about. However, often the idea of "user-centredness" is interpreted so that the users know best what a new product or a service should be. This can go very wrong when the users have little idea what the future technology is. Here are few examples: It is year 1887. Imagine a photography company thinking about new ideas for its business. The gelatin based glass plate business is too competitive to bring in much profits – something new must be invented. How about a consumer camera so easy that anyone could take and enjoy photographs. The photography company sends its UX researcher to ask potential users if they would like to take photographs. Photography in 1887 was hard. The cameras were expensive and the glass plates cumbersome to use. The new dry plates were more usable and faster than the old wet plates but still the process of getting paper prints from those plates took time, money, patience, and experience. The idea that every man and woman would take photographs had to compete against this image of a tedious and cumbersome task dominated by men with long beards (the geeks of late 19th century). What would the UX researcher ask his sample of potential users? "Would you like to take photos?" or perhaps "Would you like to take photos if you had ?" or perhaps "Would you take photos if it was easy?". Whatever the question, the way it is formulated will critically shape the answers. Jump to year 1980. An electronics company has an idea that people could listen to music while they are moving by foot. They send out their UX researcher to ask whether people would like to listen music while they walk, jog, or sit on a park bench. What would have the potential users answered? "No, I think a record player on my lap would be very inconvenient!" or perhaps "No, I listen to music in my living room with a glass of red wine and together with my husband". We are lucky that George Eastman and his company that later became Kodak did not listen to the consumers in 1887, and that Sony in the 1980s built the Walkman although the thought of "mobile music" was probably quite alien. And yes, supposedly Henry Ford said that if he would have listened to the users, he would be building faster horse carts. User-Centric Design (UCD) principles have been fighting an upstream battle last two decades. However, once these principles gathered mass and speed they became hard to stop. UCD has such a stronghold that it is hardly questioned. In other words, criticizing UCD today is not politically correct because the user is the king, isn't he/she. It is time to reign in those galloping horses of UCD and acknowledge the great work done in changing software processes to acknowledge the end users. It is important to understand that the users and the context of use are critical for good design. However, at the same time we should re-think the proper role and weigh of UCD in projects. The role of UCD is different in coming up with wild new ideas and in planning the implementation of those ideas. If the new product is radically different, how on earth can the users relate to it before they can get their hands on it? How to ask users about something radical? Second, the weight of UCD in development should vary depending on the kind of technology being built. If the goal is to make software to sell ideas or sell other products, the software must support the idea more than end users' preferences. Software is so inexpensive to make that it is often built only to demonstrate an idea, much like a powerpoint presentation. However, my point is not that we should throw all UCD principles out of the window. We should understand those principles in the context of what is being built and for whom, and why some UCD principles and methods apply better to project A than to project B. So pick up all those UCD school books but this time read them with a pinch of salt. For further reading I recommend: Norman. Technology first, needs last: the research-product gulf. interactions (2010) vol. 17 (2). Anderson and Tushman. Technological Discontinuities and Dominant Designs: A Cyclical Model of Technological Change. Administrative Science Quarterly (1990) vol. 35 (4) pp. 604-633. (Thanks to Mikael Johnson for the discussions preceding this post)