Emotional intelligence

x Sponsored content on Research Live and in Impact magazine is editorially independent.
Find out more about advertising and sponsorship.

In reviewing the many reports on large surveys from research agencies that fed into these companies’ strategies, he felt that the area was ripe for disruption.

“There are a number of challenges with surveys,” Johnston says. “They’re retrospective, timely and costly to undertake – and sometimes they’re just giving you the answers you want to hear, to questions that aren’t always appropriate. Like a bank asking people to rate them on how innovative they are. Do people actually wake up in the morning and think about that? They’re more interested in: ‘how quickly can I get money from the ATM? Is my mobile banking performing well?’”

In 2011 Johnston — alongside the School of Psychology at Queen’s University Belfast — launched Adoreboard, which measures online sentiment around brands, and which was named as Best Technology Start-up 2014 by the Silicon Valley Global Leadership Forum. The platform scours thousands of online news and blog sources, as well as social media, for information on what people are saying about brands or topics. It can specifically detect over 24 emotions, including trust, admiration, rage and terror, within those sources.

When a brand name is entered into Adoreboard, a quantified perception score – an Adorescore – of anywhere between -100 and 100 is generated. This score combines three independent variables: the authority of what is being said (based on the trustworthiness of the author of the data); the emotion contained in what is being said; and the currency of what is being said (how ‘fresh’ or current the expressed opinion is).

The platform can be used to measure the emotional content of text. In a recent study for low-cost airline easyJet, Adoreboard measured the emotion of emails the company sent to customers. It was able to understand the correlation between the emotions expressed in the communications and the subsequent click-through rate.

Adoreboard then used its understanding of the model to predict how the airline could increase its click-through rates by appealing to certain emotions that would generate more interest from its customers.

Focusing on the importance of emotion for decision-making, the platform is based around research, by Queen’s University, on how emotion is expressed online, including work on how avatars express emotion and could interact with people.

“We use an approach known as common-sense reasoning,” explains Johnston, “which uses a knowledge graph on how people converse in every day life and creates a semantic link between the concepts that are expressed.

“For example, if I was to say ‘birthday cake’, in our linked graph we know it’s related to an event and ingredients, and we can create some fuzzy logic around the emotions that are typically expressed towards that, based on annotated evidence of how people react to things in everyday life through emotions.”

The idea is that Adoreboard allows users to carry out a more exploratory analysis of data, looking for concepts that are prominent around a brand rather than asking direct questions. It also means, says Johnston, that real-time research can be conducted at any point in a campaign’s life-cycle.

“Our vision is that we create a product that follows the person through their workflow,” he adds. “What happens with the traditional forms of marketing is that it’s usually the best ideas around the table that inform the creative, then there’s some research that underpins it.

“What we’re suggesting is that you can do this in real-time and you can start to understand your customer a lot more quickly. You are also able to iterate and test in a more experimental way to get your messaging correct.”

The real innovation and disruption that Adoreboard brings to the table, says Johnston, is the ability to pre-analyse content before it goes out, then to re-write that content to ensure certain emotions are expressed.

Does Johnston see Adoreboard as a direct competitor to traditional research? Yes and no. He sees the platform as being for the chief marketing officer, the brand manager or the head of insight, “to really empower them, to demystify research and make it easier to get to the heart of how customers feel about the brand in a more responsive way”.

However, there’s still a lot to be said for the human interpretation, he insists. “It’s really interesting in that there’s this idea of exploratory analytics – the promise of big data, and of analytics, is that it can do everything and removes the human element. But I think research in its traditional form – there’s a lot to be said for that human interpretation; the linked intelligence that means the expert is able to give their opinion on what the data says. There’s increasingly a use case for our platform being used to undertake exploratory research: that could be just an API [application programme interface] connector to our system where they are using us as one strand of their analysis.”

This article originally appeared in the July 2015 issue of Impact magazine