social network services segmentation
Posted on 06. Dec, 2009 by wolfgang in all, social media
anderson analytics had published a new segmentation model for social network users (and non-users) back in july of this year (but i stumbled upon it via IQ2.0 today first). they profiled “americans” and found 7 segments – 4 segments of social network service users, and 3 that do not use them:
segments of users:
- fun seeker
- social media maven
- leisure follower
- business
segments of non users:
- social media pessimist
- time starved
- concerned
each of the segments is positioned along the dimensions benefits (useful and not useful) and barriers (no concerns/fun and many concerns/no fun):
source: anderson analytics, http://www.andersonanalytics.com/SNStype/
the criteria to match one of the above segments are not published, neither the methodology of the research study. the just tell us readers, that
The segmentation model above is based on several variables from our recent research study. However, we have selected just a couple of the variables from the model above which do a fairly good job at predicting someone’s membership in one of the segments.
uh. on a following page they published a short online questionaire with 4 questions to answer in order to position oneself in one of the above segments. those contain (1) the motivation/reasons of the participant to use sns as well as the strength of doing so on a six-part-scale (stay in touch with friends, doing business research, doing business network and express creativity), (2) the barriers the participants are believing someone else can have when using sns (danger of information theft, finding sns boring or prefering the phone), (3) the self-assessment of the participants regarding their internet-capabilities as well as the amount of their real-life-contacts in comparison to “the average person” and (4) a list of topics the participants are informing about on the internet (frequently, sometimes, rarely/never).
i for myself of course went through the tool. after answering the questions, a new screen appeared telling me, that their analysis showed, that i am a sns website user (= someone, who used sns in the last 30 days) and if this is correct. hehe, smart – i mostly agreed strong on question one (asking for my usage of sns). ofcourse i pressed “yes”, but even more interesting was, what happened by pressing “no” – because this seems to be the question that separates me from the two groups of segments “users” vs. “non-users”. is it that simple? anyhow, when i took the test the first time (pressed yes), it produced the following output:
ok, i am a social media maven, i use sns (yes, even from work), i have a blog (obviously), i am at least interested in the twitter-phaenomenon (even though there is really no increase in tweeting the last 6 month). and now a company knows, that i am one of 24 million americans, am 27 of age (on average 
), are an early adopter (!) and have an approximate yearly income of $71K – so i am an very attractive consumer for you, eh? but how do you get in touch with me? how do you reach me with your message? simple meet me at facebook (but there are the other 4 types in my sns-user-segment, too)? not that easy after all.
btw, if i only would have decided to say “no” to the question wether i am a sns website-user or not, i would have become a time starved – on the way to become a sns-user but just at the moment out of reach for social media marketing …
beside the fact, that i would like to learn more about their (methodical) approach (and there is one for sure!), i ask myself, how this segmentation can be of use? taking the available information into account, i doubt, that the covered variables really lead to disjunct, homogeneous sets of customers. i also can badly imagine, that those sets are very responsive. so is this really a good market segmentation (measurable, accessible by communication and distribution channels, different in its response to a marketing mix, durable (not changing too quickly), substantial enough to be profitable)?
make up your mind!
