Exploring Emotion Analytics

“I like good strong words that mean something…”
- Louisa May Alcott, Little Women 

"It’s not what you said, it’s the way you said it.”
- My mum

That is a direct quote from my mum that followed me around for my pre-teen and adolescent years.  I’ve since rediscovered it as my children have grown older and I find myself recycling those phrases that I said I would never use.  Whether I actually intended to convey the tone that my mum implied in her response is a moot point but reliving the phrase got me thinking, how good are we, as humans, at actually reading the emotions behind the words and phrases we use?

There’s a saying that ‘communication is the response you get’.  When we are face-to-face with colleagues, family or friends we can see the response our communication gets. We see the slight lowering of the eyebrows when they are confused and we respond, re-phrasing what we say to lessen this confusion.  Or we see their eyebrows raise, a slight upturn in the corners of the mouth and a wrinkling of the forehead when the penny drops or the light bulb appears over their head.  Reading people’s emotions generally works pretty well in these situations.

But we now live in an age of text communication the like of which has never been experienced before.  For example, a colleague of mine recently purchased themselves a new iPhone X.  After showing me face recognition and animojis he then explained to me that he’d been quite shocked one Saturday morning when one of his friends rang him up!  His thought process went like this: “What are you doing!?”; followed by “Just message me like normal people do!”.  My colleague’s expectation that communication should, in the first instance, be conducted via the typed word made me think about communication generally and how it has changed and then, of course, my working head woke up and I began to think of it in a job perspective.

You see, I work for Hitch and head up the Behavioural Insight team - I’ve conducted research for 20-odd years, with qualitative research really being my bag.  For most of these years we’ve relied on transcripts, audio and video recordings of interviews and focus groups to help us understand, not only what people said but, to quote my mum, the way they said it – rewinding and playing C90s or VHS cassettes backward and forwards to listen for the subtleties of communication.  However, we don’t always have video to review to look for non-verbal cues or shifts in tone of voice.  Furthermore, much research is now done digitally via ‘controlled’ platforms such as online focus groups or forums and, more and more regularly through social media and already existing online groups and forums.  As Professor Bing Lui said,

“For the first time in human history, we have a huge volume of opinionated data recorded in digital form for analysis.”

But how can we trust that we can analyse and measure, not just what people said, but the way that they said it, when all we have are words on a page, or a Tweet or an email or an open-ended survey question?  And, why does it matter?

Well, approximately 11 million bits of information hit our brains at any moment.  And our poor brains can only consciously process around 7 (+ or -2) pieces of that information in any one moment.  That means that around 99.99% of our brain processes occur unconsciously… AND … we make around 10,000 decisions per day.  Our brains usually run on autopilot, despite making us believe we know what we are doing.

What’s making our decisions?  Well, it has been estimated that between 80-95% of consumer decisions are driven by emotions*.  If that’s the case, then 8-9,500 of our 10,000 daily decisions are being driven by emotion.

So, we can see why understanding the emotional drivers and barriers is important.  But it is notoriously difficult and even more difficult to do quickly with large amounts of qualitative data.  By using state of the art Emotion Analytics artificial intelligence software, qualitative data is turned into evidenced, actionable insights, quantifying the individual emotions expressed along with qualitative topic analysis.  We, at Hitch, have developed a package that marries Emotion Analytics with tried and tested research approaches, to deliver insight to help develop policy and evaluate strategy, campaigns, brands or just capture how people feel.

But don’t think this is just for the Big Boys.  Sure we can think about large scale tracking studies or specifically designed surveys but we can also think about the amount of data we all get from our Twitter feeds, or emails to our customer service teams, posts to our Facebook pages or in-situ interviews.  All this can be used to understand how people are feeling and what it is that is actually making them feel that way.

So mum, maybe we can say, “It’s not what you said, it’s the way it’s written.”

https://hbswk.hbs.edu/item/the-subconscious-mind-of-the-consumer-and-how-to-reach-it

 

 

 

Copyright © 2018 Hitch