Description
Several designers independently create different designs or low-fidelity prototypes, which are evaluated to learn what works before the best elements are merged into one design for iteration.
Individual designers tend to settle on an early, satisfactory but suboptimal solution, and group brainstorming does not reliably do better. Parallel design counters this by having several designers work independently on the same brief, aiming to saturate the design space before anything is chosen; the more varied and independent the proposals, the better the merged result. The HHS guidelines recommend starting a site with parallel design followed by iterative design and usability testing. Dow and colleagues' parallel prototyping adds evidence that building several distinct low-fidelity prototypes reduces fixation and 'hill climbing' and keeps critique on designs rather than designers. Each alternative is tested or inspected to learn how people respond to its elements, not to pick a winner, and the strongest elements are combined. In Nielsen's diversified parallel design each designer optimises for a different aspect, such as novice users, expert users or a non-verbal interface, and existing competitor products can stand in as further parallel designs.
Attributes
- Discipline
- Design & Product Practice
- Loss type
- Vestigial
- Stage
- Design
- Data
- Qualitative
- Effort
- Days
Steps
- Write one brief and give it to three or more designers or pairs who work without seeing each other's work.
- Have each produce a distinct design or low-fidelity prototype within the same short time box.
- Push for variety: if proposals converge, ask for further alternatives until the design space feels saturated.
- Test or inspect each alternative with a few users or experts, noting which elements work and why.
- Merge the best elements into one design, checking the result is coherent rather than a patchwork.
- Continue with iterative design and usability testing on the merged design.
Sources
- Research-Based Web Design & Usability Guidelines (HHS, 2006)
- Universal Methods of Design (Martin & Hanington, 2019)
- Usability Engineering (Nielsen, 1993)
- Human-Computer Interaction (Dix et al., 2004)
Also known as
- Independent parallel design
- Parallel prototyping
- Multiple low-fidelity prototypes
- Diversified parallel design
Categories
- Design
- Design & Product Practice