I've been working over the past three or four months on a methodology for measuring emotion.
It's not that I'm overly concerned at how often interfaces make people cry, it's more that pretty much everyone in this industry now recognises that 'usable' doesn't mean people will use it. I mean, we all know that a usable e-commerce site will suffer fewer dropouts and losses at the checkout due to usability issues, but that doesn't mean it will be an instant success - the prices have to be right, the product range has to suit the customer, the delivery slots have to be reasonable, the fulfillment on time, etc. Put another way, the user still needs to be 'persuaded' through the design.
Marketing has been at this for years, measuring the emotive impact of an ad or campaign. They question the consumers before and after exposure, and find how well the campaign has strengthened the core brand values. So why don't we do the same with exposure to an interface?
The answer is we can, and some of us have been doing so for years - though probably not as thoroughly or scientifically as we should. And that's what I've been trying to correct.
Some usability experts are working on a scheme of persuasion; persuasive design is a popular buzzword right now, and designs are measured to see how well they persuade customers to complete. But for my money, that's too simplistic and is more of a broadcast approach (interface broadcasting at and persuading users), whereas it should be a dialog approach (interface engaging with the user and providing the required experience - which should encourage them to repeat/complete).
The trouble is, that experience changes every time. Sometimes, as in an e-commerce site, the experience should be relatively boring and simple, although it should also entice and engage to build a relationship and dialog. At other times, the experience may deliberately confuse and frustrate, to an acceptable degree; think of a puzzle game, for example, which requires a degree of confusion and problem solving in order to deliver a satisfaction pay-off.
So the key is in understanding what the potential emotive aspects of the experience are, and then in setting goals and touch points along them. Then you can identify a way of measuring how well the experience met those goals.
I'm currently working towards a four by four grid of experiential measurement components that I am hoping will do this. If anyone else is working on something similar I'll happily share notes...!
In my view