As Chris Berry noted in his recent post on personas and market segments, the intent is to impart a sense of empathy. While I would agree with this statement I think there is even more to it. Personas should evolve. That is, as new information comes to light that information should be incorporated into the personas. One might also find the information does not fit one of the persona types and so a new persona must be created.
Personas arise out of a lot of research – some of it quantitative, some of it qualitative. They act as archetypes to aid in decision making and act as stand-ins for real users. Some companies feel personas are flaky, like having imaginary friends and only stop at the archetype level of persona development. However, there is real value in creating a story around a persona.
Here is a great example of a persona from Step Two Design (they even tell you how to create them!):
Bob is 52 years old and works as a mechanic with an organisation offering road service to customers when their car breaks down. He has worked in the job for the past 12 years and knows it well. Many of the younger mechanics ask Bob for advice when they meet up in the depot as he always knows the answer to tricky mechanical problems. Bob likes sharing his knowledge with the younger guys, as it makes him feel a valued part of the team.
Now that is much easier to relate to than: Males, ages 44-46 years old. Median income of $46,000. Limited computer experience.
And as Chris noted, personas don’t always work. But is that the fault of the persona or the fault of the person using it? If the persona is based on real data then I have a hard time thinking of how it couldn’t help. But if that persona is based off of rapidly changing information, or erroneous information then it could do a tremendous amount of harm.
What would happen if you noted that a new group arose? There was a segment that either newly developed or wasn’t fully realized at the time of persona creation? You would need to either incorporate that data into Bob, or create a new persona.
The issue behind it all is that the web isn’t static, it is constently changing. Ever heard of AOL? Of course you have. Know anyone still using it? Probably not. What about MySpace? Also there is the inevitable the march of time. Your persona for Bob has his age at 52. In five years time, will Bob still be useful? Probably the difference between 52 and 57 isn’t that large. But what if your target demographic is 22? There is a much larger difference between the interestes of a 22 year old and a 27 year old.
9 months ago the median age of someone using twitter was 18-24 years old, that is slowly slipping into a older demographic 25-34 years old
In conclusion, use personas but don’t let them get stagnent, your personas represent people and people change.