Joe Leech: How to design with science
Joe Leech was talking about how scientific evidence can make your deisgn stronger at DotYork, these are my notes from his talk.
- Product goes live and it's lost the initial vision
- The client continually questions the design
- Being micromanaged by the client
Eye tracking is a great piece of science you can use in design - it shows where people are actually looking, it can prove that people are (or aren't) looking at items in the design.
"Design is not so much a design issue as a a power struggle." ~ Alan Cooper
"The biggest myth ever perpetuated in the design field is taht good design sells itself." ~ Mike Montiero
When people are cllecting train tickets they often have bags with them - people don't have their hands free. People don't print out booking referecnes, they look at their email: putting the reference in the subject line means people can find the reference quicker on their phone.
Don't mark the mandatory fields in forms - mark the optional fields. Even a small number of people having problems filling in forms can show a better way to approach a problem.
Everything we do is business (of a sort) we need to understand how they work so we can understand what the goals are. However, even Google an Microsoft have trouble proving the return on investment of advertising.
Five business models:
- Addiliate fees
Data is really important when doing thigns online. Don't be scared of the way optimisers talk about data. You can measure an inordinate number of things - you can get analysis paralysis. You can't always judge the details of a design on data, you need to be able to talk about data and numbers to defend your work.
Measure two or three things: timescale -> benchmark -> thresholds -> actions.
Don't associate designs with yourself, your feelings, your thoughts. It invites people to have opposing oppinions based on themselves, their feelings and their thoughts. Don't confuse design with art. (Design is a job)
Subliminal messaging caused a 58% increase in popcorn sales. However, this study was completely made up - it did launch a career. It hasn't died but people are still trying it. Use good theories not bad theories.
Try and discover a relevant theory to your problem.
Use all these techniques to build a strong argument to justify your position.