The design process: Part 1

Unlike the scientific method, the design process is not a well-defined thing. Most design professionals have a preferred process, and many are quite stubborn about theirs being the right one. Usually, there is some good logic to a well-developed process, but I've always had trouble taking people seriously when they claim that theirs is the only way to do things. Based on my admittedly limited experience, the most accurate representation of the true process is something like this: (Creative Commons, Damien Newman)

In his book Designing for Interaction, Dan Saffer describes four distinct methods for design: User-Centered Design, Activity-Centered Design, Systems Design, and Genius Design. Without going into the details of each of these methods, suffice it to say that they are quite different. As far as I can tell from what I am being taught at the Segal Design Institute, User-Centered design is the hot one right now. UCD places the user at the center of the process, and allows a design to form and be refined based on the needs (latent or expressed) of this user.

Of course, as Daniel Nettle explains, humans are notoriously bad at knowing what will make them happy. This presents a problem for the practitioner of user-centered (or human-centered) design. Because of this, rather than simply asking people what they want, a designer needs to make rather unscientific inferences based on ethnography and a slew of other observational data-gathering tools.

From this qualitative data, design concepts somehow emerge. But how? Designers are typically encouraged to shelf their own personal biases and prejudices in forming conclusions from what they see in the field. But the real world seems a lot more complicated. From a given set of observations, there are infinite conclusions to be drawn. How the data is clustered, what meaning is ascribed to it, and what initial solutions are proposed will all be different with a different set of designers working on the job. This is ostensibly a good thing (there's more than one way to skin a cat, after all) but this is the time that those latent prejudices can slip back in.

Designers are people too, and all people have their own agenda. For an academic project, the goal is to get a good grade; for a client, it is to impress the client; for a business, it is to make something marketable that fits within the constraints of the company's structure and culture. How can a designer avoid pushing their interpretation of the data--and thus the initial design concepts--toward meeting this unrelated agenda?

Why not embrace the fact that designers have biases? It seems logical that the more you try to keep yourself out of something, the less you understand how you're skewing the results. There must be a better way of doing things.

More on this to come.