In Tales from a Trading Desk, this week Matt Davey has picked up on an interesting article about UX and ROI.
I like this sound bite:
Also, if you create great UX it will always lead to better user satisfaction, joy and happiness – and it’s a well know fact that happier users are – better and more motivated and productive workers.
My take on what makes a great UX…
For me a great UX:
- Turns users into advocates leading to an increase in most of these ‘Hard’ and ‘Soft’ metrics by default
- Should be analogous with great typography: transparent, ease information comprehension and consumption
- Doesn’t interrupt the ‘experience flow’
- Considers ‘emotional design’ and the users emotional responses.
UX: Time to get emotional
I started thinking about how I capture the ‘emotion’ of an experience.
It would be interesting to add some emotional user metrics to the list of elements, over and above just ‘Customer Satisfaction’:
|
Hard |
Soft |
| Conversion / Acquisition | Engagement |
| Lead generations | Customer satisfaction |
| Retention | Loyalty to brand / Building champions |
| (Targeted) traffic | Utilization and product / service adoption |
| Viral referrals (not only videos) | Awareness |
| Channel migration | Ethics |
| Employee productivity | |
| Cost savings |
I’ve been following the development of Emotional Design over at the Design & Emotion Society for a while and researching the emotional responses to images and text – following some ideas from ‘Persuasive Imagery edited by Linda M. Scott and Rajeev Batra’.
Capturing the desired emotional response and the actual user response is a fuzzy area, but the tools are getting better all the time.
I’m wondering if quantifying the emotional response is the real driver for discovering valuable ROI within technology? Don Norman discusses this in an interview over at Design & Emotion.
I apply these factors to my UX Design at Caplin. Although the domain of SDPs (Single Dealer Platforms) for financial trading is typically seen as an area where emotions are kept in check, we are finding that the reality is different and user emotions matter as much here as anywhere else.
The most difficult thing about capturing emotional response to a piece of technology is to do so dispassionately. To approach the objectivity of something like Fitts law we need a battery of measures beyond simple introspection.
It’s difficult to separate the emotion in most of the measures you suggest, as there are so many confounding factors, and of course you can’t deliberately release a rubbish product to work as a control.
I’ve had moderate success with Likert scales and physiological measures like galvanic skin response and heart rate, and I’ve been part of projects that videoed users as they used software at IBM (measuring how the user performed tasks, although there’s no reason why it couldn’t also be used to try to capture microexpressions) but this was with software that was deliberately trying to induce a physiological response.
It’d be fun to try some of this stuff with an MRI.
Sure – at one of my previous companies we were going to by a brain scanner to understand a users actions, as per: http://www.guardian.co.uk/science/2007/feb/09/neuroscience.ethicsofscience ~ I was a bit uncomfortable with this, it opens up a whole can of ethical worms…