Personas by MIT Media Lab

I read Matt Rhodes, “MIT’s Personas project and owning your brand equity online” and was quite impressed by “Personas“.

Rhodes indicates the drawback of the product, which is the fact that it is name based and since most names are shared, your Personas results won’t necessarily be all about you.  He notes a better way to “show you how the Internet sees you” would be through the use of shared credentials (such as OpenID).

I agree with Rhodes on this point, but I also see tremendous value in tracking your results over time as a measurement tool.

The two primary measurements I see of immediate importance are one, how the profile shifts to categories related to you, as compared to other individuals sharing your name over time, and second the extent to which areas related to you grow or shrink individually based on where you are placing your emphasis.

My name only renders two searches, a client’s name I used compared over 30.  I knew this client had a more ‘common’ name, because we’d worked on building his online presence and it was much more difficult than my own.

I ran the Personas search on my name multiple times and each time got a slightly different grid.  So I took five sets of results and estimated a percent for each category, then averaged them to achieve a composite of my composites.

Three categories ranked over 10% on average:  education, online, and politics. Management came in at eight followed by sports at seven.  I would agree that these are my top online categories based on my current and past work experience.

In two of the five sets “aggression” scored a high percent (10% & 16%) and I haven’t quite figured that one out.  Others I wasn’t too surprised to see such as religion, family, social, genealogy, etc.

Personas is very interesting, and I’ll be curious to see what uses can be applied to it, particularly related to reputation and identity management.


Comments are closed.