So many things that seemed apparent to me, were no so for the students. User testing experiences I had were similar.

I believe it is good practice to test and use personas until you develop that way of thinking. And even when you do —testing does not hurt. That does not mean that you have to user test every little assumption you make —If you do not develop some level of instinct over time, you’re in the wrong profession.

Also, there is a lot of ‘usability’ data that accumulated over hundreds of years —typography, color theory, perception, etc — As they say —those who do not learn from history…

To piggy-back on Steve’s post above, we’ve found that details about a persona–car they drive, age, number of kids–plays a larger role than making a user real or identifying to what demographic a persona may map.

The details of a persona may illuminate the most subtle, yet important thing about a user—-user motivation. Motivation can be heavily influenced by environment. This is where personas go and use cases fear to tread.

In the same way that we remind ourselves that users are not designers and we are not the target users, we should remind ourselves that the user will tell us how much detail is enough for a persona.

Without a deep understanding of a user’s enviornment and how they work within it, a persona (and a usability expert) can miss key motivators to user behavior.
For example, a mom with 2 kids and a Dodge Caravan. How might she be different from another Mom persona in which we don’t know that info?

The Mom with kids persona leads a hectic lifestyle. Her free time is fleeting. Her kids are always into something. She must keep one eye on them even when they are napping.

She is usually in a hurry and will not have time for sites that require a large time investment. Her goals are practical (perhaps the motivation behind the minivan) and speed-based.
Her world is much more real and important to her than any “domain” we create.

This info can help describe a personal goal–”I don’t want to feel like I’m wasting my time.”

That could have big design implications and help you decide what functions to keep and what to discard.

For example, it may be nice to have that nifty ESP search function. What better way to improve search results? Surely the users will see value in that.
But it takes nearly 5 full minutes to read your mind. So, this Mom (a high-priority user) would grow impatient with it–she can’t imagine sitting still that long.
The ESP search tool should not be included (even if it is cool) because it is too slow.
But this also opens a new opportunity–how can we reduce the ESP search tool’s search time so it provides real value for our target users?

Understanding motivations and influencers is key to good personas. How much (or few) details to include depends on when the details stop adding value and no longer matter.

As it happens, the February Cooper newsletter contained an article by Elaine Brechin about this very topic. Here’s a quote:

“Personas and market segments provide different kinds of information. Market segmentation provides a quantitative breakdown of the market, while personas provide a qualitative analysis of user behavior. ”

The rest of the article can be viewed at http://www.cooper.com/newsletters/2002_02/reconciling_market_segments_and_personas.htm

Secondly, how often do you “update” personas, for example, when a new product is developed or a new customer comes along with a different type of user base, such as an older population? Is it ok to constantly update or change your personas or should they be relatively permanent?

“Personas are a set of fictional, representative user archetypes based on the behaviors, attitudes, and goals of the people we interview in our research phase.”

So even though standard marketing demographic data plays no major role in creating realistic personas, there _is_ a justification of the validity of the personas in the user research (mostly interviews) that’s being done beforehand. It’s not just based on designers’ assumptions. I think this is an important aspect that often gets left out when discussing the use of personas. For me anyway, it took away some scepsis about the method…