You may be all too familiar with the drill by now: applications and services need to be pushed out the door yesterday, and everyone is racing against the clock to get everything done. Today's digital enterprises want new functions and interfaces released almost every day, which clashes with developers and service designers' desires to put together something attractive and compelling for users -- which, by the way, will actually get used -- with a minimal amount of confused calls and emails about how to navigate this or find that.
In a recent post at the UXPin Studio site, Stefan Rössler, co-founder of Simplease and Userbrain, describes a seven-step process his design firms employ that "involves testing ideas as early as possible, rapid prototyping, and usability testing" -- before "they wreak havoc."
Here is a summary of Rössler's rapid UX design process:
Step 1. Identify the problem, ask a lot of questions: Find out why people want specific features by asking lots of questions and listening carefully to the answers. Rössler recommends identifying design problems -- which often may not be apparent early in the process -- with reverse engineering.
2. Sketch out ideas to solve the problem: Literally, Rössler recommends that you take a pad of paper and draw some quick sketches. "The reason for this is that your first ideas usually aren't very good -- because you don't understand the problem well enough at this point. Pushing out a multitude of potential solutions accelerates ideation in the same way a pressure cooker accelerates cooking. Once you get into a groove, it's safe to tone it down and let the ideas percolate."
3. Share sketches with your team: "At this stage, you're looking for what doesn't work," says Rössler, who is adamant about showing, but not explaining, the sketches. If the solution is as obvious and intuitive as it should be, explanations shouldn't be necessary.
4. Revise your sketches and prototype your epicenter: "Once you [find] an idea that might work, build a quick prototype," Rössler says, noting that the approach that works best for him has been to "start designing the most complex state (or screen) first and then backwards-engineer to gradually design a complete task flow."
5. Test your prototype with potential users: "Do everything you can to make things break, then discard whatever doesn't work," Rössler says.
6. Interpret your test results: It boils down to a simple result, Rössler relates: "If our prototype solves the problem, it's effective. If users aren't able to solve their problem with our prototype, it's ineffective." He suggests questions to ask at this stage, including: "If you think about the last time you did something like this, did you need this kind of feature? Can you please describe to me what exactly you needed when you did this for the last time? Is there a critical feature missing that you would absolutely need in order to solve the problem?"
7. Build a new prototype and keep testing: "The most effective way is by building quick-and-dirty prototypes, then watch them break during usability tests to reveal bad ideas," says Rössler. "Keep this up until you arrive at an idea that's really good." Then it's time to run with it.