Perhaps this is why one of the world’s biggest comms group , WPP,  recently signed a long term agreement with twitter. I did suggest on my facebook page that gaining access to valuable ‘customer insights’ was the real driver in this deal (28 Aug). The heaving mass of tweets (just like facebook posts) can be aggregated and fed into an algorithm to identify personality traits.

twitter streamSo analysts can take a ‘good  educated guess about your personality just from looking at 200 of your Twitter messages’. And now this can be ‘done in a disciplined and automated fashion using the science of data analytics’.

Old Segmentation v Psycholinguistics

Old segmentation based on demographics (age, sex, marital status, home, income) and behavioural analysis (what you read and/or what & how you buy) can now be enhanced (if not replaced) by psycholinguistics (analysis of anyone’s choice of words when using twitter).

Market researchers know that a buyer’s behaviour and decision-making process is heavily influenced by their intrinsic  traits e.g. people who are happy or depressed use different words.  

Based on psycholinguistics, data analytics can now derive people’s traits from their ‘linguistic  footprints’. So extracting correlations between a person’s word choice (& activity patterns) can reveal these intrinsic  traits. Something we couldn’t dream of just a few years ago.

Taking just 20 seconds to analyse 200 tweets (or  2,500 to 3,000 words) , IBM researchers claim to  create a personality evaluation that is accurate to within 10%.

In the meantime, I’m just a bit concerned that IBM might categorise me as a ‘face-enabled, automated bill board ad junky’ (if you are wondering what tweeted Freudian insights dig deep down into my dark secrets  then click the tweets on the blackboard to read into my personality or at least, into some of my interests.

Part 2/2 asks ‘why bother?’ and looks at the benefits of this kind of psycholinguistics analysis.