What is your approach to creating personas? Something that I’ve been flip-flopping about recently is the distinction between what I think of as behaviour-based personas and segment-based personas.
To understand the difference between these two classes of persona consider some real life examples.
If you were researching personas for library users a segment based approach might include the following personas:
- an older patron who is researching their family genealogy;
- a secondary student who is researching a homework topic;
- a migrant who doesn’t have the best command of the English language; and perhaps
- an under-served segment – youth from a low socioeconomic group who would greatly benefit from using the libraries, but at the same time face multiple barriers to joining.
These are all valid people, but currently I would lean away from this approach for two reasons:
- while this breakdown of people represents very important users of the library, they capture a very specific and narrow group of people. With this approach you could easily end up with a group of 20 equally important personas and still have people questioning “what about so-and-so?”
- for many of these people there are a range of tasks that are very similar, but with a different context, for example someone researching genealogical history might need help searching for a specific relative, and a student might need help looking for information on homework topic. While these two people are searching for different content, and most likely in different sources of information, there is a common task of needing support with information retrieval skills.
The approach that I’d currently lean towards would be defining the different behaviours that people exhibit and building personas from those.
For example in 2004 the British group econsultancy released a benchmarking report around online retail that outlines some different consumer search behaviours that they described as tracker, hunter or explorer.
So my current preference for personas is to move towards more abstract personas based around differences in behaviours instead of segment based personas. But I’d love to hear your thoughts and opinions on this topic!
p.s. If you’re new to personas, I’d highly recommend The User Is Always Right: A Practical Guide to Creating and Using Personas for the Web
Tags: Personas
If your post is loosely ’segments vs behaviours’ …You are leaning in the right direction.
There’s really no contest here. Segments do provide base level research for building personas but there is no question that they are not swappable for Personas themselves.
Market segments cover demographics and attitudes, while Personas are based on motivations and goals. It’s these that need to be designed for.
My approach has been to use market segments as a basis on who to recruit as participants in the research, then to let the personas develop from behaviours and goasl which emerge during the research sessions.
All the best …Nick
Hi Nick, thanks for your comment :)
Looking back this post was poorly written, I was having difficulty expressing my thoughts and I think I need to revisit this idea again because when I first started becoming exposed to personas it was natural (at least for me) to think about people from a market-segment perspective, and nowdays when I meet with a client who have created their own personas most often they are also based around market segments.
I like the way you’ve explained how market segments are used to recruit for persona research, and how the behaviours emerge from this research. I think a common trap for people not familiar with working with personas is to start research with pre-established ideas of what the personas will be based on market segments.