A gold mine of personality data?

Written by Web News Editor on . Posted in Features

Twitter
There seems to be no end to the potential for Statistics to harvest the internet – blogs, tweets, e-mails – to find patterns in our innermost thoughts and emotions and to understand what makes us tick as humans. Think back to the Joy of Stats and the vast ‘madness movement’ map which analysed our digital traces and unwrapped what ‘mankind’ is feeling at any given time.
 
What might be an everyday way of using this knowledge? In their bid to match customers to products and services, marketeers are seeking new sources of intelligence and it seems that social media - a mine of big and valuable data - and until now mainly a tool for informing consumers about new products and services, is a good place to start.
 
 
Researchers at IBM’s Almaden Research Centre in California certainly think so. For Dr Eben Haber and team, in the future, marketing will be less about “demographics” (age, sex, marital status, location, income etc) and current buying habits and more about tailoring promotion of goods and services to customers’ deep-rooted personalities, values and needs. E.g. are they ’extrovert’ and so more likely to respond to a mobile phone ad that offers excitement rather than convenience or security?. Or are they ‘agreeable’?  ‘conscientious’? ‘neurotic’ or ‘open to experience’? all traits which underpin different purchasing behaviours and offer valuable information to marketing departments.
 
Inspired by the findings of earlier research undertaken at the University of Colorado, which looked at the association between frequency of certain categories of words used by bloggers and their personality traits, the IBM team has developed software that takes streams of Twitter data and searches for words that indicate a tweeter’s personality, values and needs.
 
Yes, extroverts are more likely to use terms like “bar”, “crowd’” etc and neurotics to refer to things being “awful” and “depressing”,  but they began to see new associations too e.g. between aspects of ‘agreeableness’ such as trust and co-operativeness with “summer” and “unusual”. They also began work on matching people’s values e.g. things like loyalty, accuracy and self-enhancement their needs e.g. excitement, control or acceptance. In brief, getting more inside who we are as customers.
 
Proving the power of statistics - in this case, statistical computing – to bring out the information in data…the IBM team analysed 90m users’ Twitter data across 3 months and they found that analysis of 50 of our tweets offered a fairly good prediction of our personality. Analysing 200 tweets did the job.
 
Social media data should aid the quest for more direct, less’ hit and miss’ marketing (recent US research cited in The Economist’s ‘No Hiding Place’ found that just 3% of those who received unsolicited post bought anything as a result and when it came to online ads, just 0.01% did). It may not offer insight into the widest customer base but young people, in particular, are heavy users and mining the social media data of say 18-24 year olds should offer a clearer picture of the types of purchases this age group may be persuaded to make.
 
 
 
 

Join the RSS

Join the RSS

Become part of an organisation which works to advance statistics and support statisticians

Copyright 2019 Royal Statistical Society. All Rights Reserved.
12 Errol Street, London, EC1Y 8LX. UK registered charity in England and Wales. No.306096

Twitter Facebook YouTube RSS feed RSS feed RSS newsletter

We use cookies to understand how you use our site and to improve your experience. By continuing to use our site, you accept our use of cookies and Terms of Use.