Design Technique 12: Parallel Design
How many heads are better than one? The technique of Parallel Design is probably what it sounds like when you heard the term. You just get two or more people to design independently and without any interference so that you can compare their interface designs and then you choose the best aspects of each design to include in the final solution. Its a rather simple concept for a very powerful methodology which has some interesting science behind it. So how do you go about that again?
Stage 1 - Independent Creativity
OK at the first stage you just send a number of people in different directions with identical information about what the design needs to be. The emphasis on this stage is independence so none have an influence on the others. When they have all done their initial design each is brought back together and their work is compared.
Stage 2 - Revisiting Designs with New Ideas
The second stage involves the designers going away again and improving on what they did in the first stage by using ideas from the other designs provided at the end of stage one. That sounds simple enough. There is no set in stone rule about the number of times this can be done but apparently there is well documented evidence (and I remember doing a tutorial at university on this at one time) every single design will show improvement. That’s a bold claim but when you think about it there’s logical reasons why - your designs only include enhancements on what you already did yourself, and likewise your fellow designers pull the best of your ideas into their work.
Stage 3 - Finalise the Design
The final stage is to finalise the actual design utilising as many of the best ideas presented by the designers as possible. Perhaps it is a hybrid of many designs or one design which evolved ahead of the rest.
The Tradeoff - Better Usability at a Cost
Beyond the expense of having more designers working on the project at one time there are the obvious benefits of casting a wider net for ideas and the speed at which evolution of these ideas into a final product can be realised. Jakob Nielssen states on his site a case study
In a case study of a screen-based telephone interface, measured usability when going from version 1 to version 2 was improved by 18% when using traditional iterative design and by 70% when using parallel design
Those are impressive improvements in usability and may be worth the cost of Parallel Design. Also, on a smaller front for the little developer in a small shop, this could conceivably be scaled down to the point where both designers in a firm go to their office and make initial designs of an interface and go from there. This wouldn’t be that expensive in-house I’m guessing. And its not something you would do on every project either. But there’s something really cool about brainstorming independently toward a ‘best solution’.
