Jillian Ney, doctor of social media: "There's no silver bullet when analysing social data"

Social data
Share this content

As a brand, the power of social media is two-pronged. On one side there’s customer engagement – being interactive, building advocacy, dealing with queries and reacting to events in real-time. And then there’s the other side – analysing large sets of public data, creating segments and spotting trends.

With the latter, the common assertion has often been that the more data you can analyse, the more you understand. The sheer volume of data available via social networks underpins this idea – statistics from Brandwatch state there are now 2.3 billion active social media users, with 1 million new active mobile social users being added every day. The upshot is that over 4m Facebook posts are ‘liked’ every 60 seconds. There are 347,222 Tweets per minute.

Yet, to paraphrase IT pioneer Mitchell Kapor, for many marketers, to attempt to glean valuable insight from social data is “like taking a drink from a fire hydrant”.

Trying to help brands make sense of these eye-watering numbers is Dr. Jillian Ney, a digital behavioural scientist who has become regarded as the UK’s first “Doctor of Social Media”. Ney’s stance is that, whilst crunching data is a crucial part of the overall analysis, brands are better served drilling down into the psychological and behavioural reasons behind social trends, rather than dealing in purely quantitative measures.

With this in mind, we caught up with her to discuss how to make social data count without suffering from paralysis by analysis.   

Jillian Ney

MYC: Why is social media data seen as such a tricky thing for marketers to incorporate?

JN: I think most marketers are undeniably interested in social data, but quite often we’re all hoping there’s a silver bullet. That there’s some technology with a button we can press that will give us some magical insight. Unfortunately there just isn’t.

That being said, the larger organisations I’ve worked with are often committed to getting on top of the issues around social media. My impression is that technology is often prohibitive: some of the social tools available are not designed for marketers specifically, and the metrics they’re used to measuring. Impressions, likes, shares; these types of metrics aren’t that valid – the need with social data is to question what else is happening. There should be an element of action attached to an insight. When people are reporting data points rather than actual insights, their results can often be underwhelming.

MYC: Is this when behaviour becomes part of the measurement process?

JN: There’s certainly a need to look deeper into behaviour and psychology. I’ve worked with guys in this field to try and capture behaviour through social media and it’s always more revealing.

You need to be able to marry both the data crunching side of social media with the behavioural side. Data analysis on its own is fine but it’s only possible to enrich data in this field by understanding certain behaviours. You analyse social media conversations under certain triggers. Motivational factors that drive people to engage with a brand. Looking at the certain semantics people use in the process of purchasing decisions. There’s also the perception part – trying to unearth what people think about a brand when put into contact with that brand’s content. Establishing customer journeys are usually a key component of establishing some behavioural trends.  

MYC: What does this process involve?

JN: It differs depending on the brand and what they’re trying to do. One company I’ve worked with recently specifically honed in on verbal cues and attention triggers on Twitter. They found there was a massive desire to purchase from a specific customer persona. We went through all of their data systematically and split different verbal triggers up into different personas – we were able to establish a number of different things about each, such as which was theoretically the ‘costliest’ in terms of spend, which was an advocate, which was most likely to be influenced by promotions, etc.

Data analysis is good for many things but as with anything digital, it can often lack the necessary empathy

You need to be able to gather the data initially, which is where the technology comes in. We are using commercial tools such as Brandwatch and other vendors together to segment data, but the difference is that this leads to more of a manual process. It’s about getting to the crux of different conversations. This is about honing in on certain actions in isolation, and being less quantitative. It can mean physically logging different elements of a high volume of interaction, and trying to establish what each type of conversation or interaction means.

MYC: How does sentiment form part of the results you glean?  

JN: I’m not a massive lover of sentiment analysis. It always seems too rudimentary. If you’re a brand and you categorise everything someone says into one of three sentiments, you’re almost returning to a quantitative approach similar to likes and shares and not creating anything actionable at the end.

What I’ve been working on a lot more recently is taking conversations on an emotional basis, rather than just saying ‘is it positive, negative or neutral’. So, what are the drivers leading people to purchase with a brand. What are the shortcuts people use to make decisions. What swayed them. What was their reaction to different types of messaging. There’s psychological elements in there that can only be analysed at a more granular level.  

MYC: What core psychological elements would you say play out most on social networks?

JN: Nonconscious decision-making is becoming a major focus at the moment. It’s always the nonconscious part of the brain that makes decisions for us when purchasing, and more and more brands are becoming aware that their messaging, its timing, positioning, relevance has a major role in the nonconscious side.

Empathetic understanding is also crucial. Data analysis is good for many things but as with anything digital, it can often lack the necessary empathy. Reacting is such a crucial part of social media marketing, and on a personal level, empathy is a core driver for being successful. Going back to the idea of standard metrics – what’s more important – getting 100 likes or retweets to a post your brand made on social networks, or monitoring a feed and discovering a customer who was upset about a product or service and then putting the necessary mechanisms in place to turn their experience from a negative to a positive? Both are social media skills but the second event is arguably more important in the long run.

We’ve pinpointed three types of empathy:

  • Cognitive empathy - the thinking
  • Emotional empathy - the feeling              
  • Compassionate empathy - the action

Our approach to social data analysis touches on these three types of empathy. Cognitive empathy is the analysis of thought (context), the emotional empathy is putting a face with customer stories and the compassionate empathy is the strategy to respond to customers (actionable intelligence).

MYC: What should a brand’s end goal be in analysing social data?  

JN: If we can understand people from their social profiles, then you can start to understand a bit more about their purchasing behaviours, interests, personality and experiences over a longer period of time.  All this talk of personalisation in marketing can all be tied in with social validation.  Birthdays, holidays…it’s information that is openly shared on social media and can be used by brands.  It is about understanding who the customer is at that moment in time.  We all consume different content, information, products or services at different times depending on the context of our environment and who we are with.  It is about understanding who the audiences are at that moment in time.

To succeed as marketers you must understand human behaviour. Social media has brought us the largest data set to analyse and understand human behaviour.  However, because marketers like big numbers, tools and analysis have concentrated on this type of analysis not specifically in measuring human behaviour – it’s what I call People Metrics.  If you measure behaviour instead of numbers you find the insight you need to create more powerful communications and experiences.  In every online interaction your audiences is telling you what they need, how to engage with them, their hopes, fears and dreams – you just need to know where to look for this data.

About Chris Ward

Chris Ward

Chris is Editor of MyCustomer. He is a practiced editor, having worked as a copywriter for creative agency, Stranger Collective from 2009 to 2011 and subsequently as a journalist covering technology, marketing and customer service from 2011-2014 as editor of Business Cloud News. He joined MyCustomer in 2014.


Please login or register to join the discussion.

There are currently no replies, be the first to post a reply.