Too much analysis can confuse people about how they really feel
Here's a short follow-up on my recent post "Don't listen too much on users - observe them instead". The presentation "Is qualitative research holding us back?" by Jason Oke that is available on slideshare.net pretty much supports the point I was trying to make. Here are a couple of quotes used in the presentation that pretty much frames what the presentation and my previous post is about:
“The consumer does not behave as they say, they do not say what they think and they do not think what they feel” - David Ogilvy"Too much analysis can confuse people about how they really feel. there are severe limits to what we can discover through self-reflection.” - Timothy Wilson, university of virginia in New York Times, dec 29 2005
So, when studying and trying to understand human behavior, it is usually a better strategy to observe people (in their natural environments) than to interview them.
Most users that visit a web site or use an application act on instinct and choose the option that seems to be the most intuitive one of the ones presented to them. If they fail to achieve their goals with that option, they try something else. If no option or path of options on the site allows them to achieve their goals, they leave the site. What is important is that each of these choices are made within a fraction of a second. Most users do not stop for several seconds to analyze the site or their own behavior. That only happens when they are tested and asked to tell someone what they are doing. Or something is fundamentally wrong with the site or app.
So when you want to do quick usability tests, such as showing some people two design alternatives to get their reactions on them, your primary aim should be to observe the users and draw conclusions from their spontaneous reactions. If they say something spontaneously within a second or two, note it down. But once they think too long before they react or they start asking you questions, the test is over.