The Business of Data podcast: Episode 5 – What is Social Listening and how can it be used ethically by companies?

The Business of Data podcast series with James McCarthy, CMO, Exonar

Having worked in the tech industry for over 25 years, James is fascinated by the impact that technology, and particularly data, has on organisations and the people they serve.

In this episode, we’re tackling the fascinating subject of social insight and social listening. Everyone knows how big a part social media plays in the lives of consumers. After a series of data breaches and the likes of the Facebook and Cambridge Analytica scandal in 2018, any use of consumer data by companies can be perceived as a bad thing. We talk about whether social insight, gathered from billions of people’s public rants on social media is a force for good, and how it can help organisations stay relevant and in touch with the views of society

If you’d rather read the transcript than listen to the podcast (we’d always encourage you to listen!), here are the words:

James McCarthy
Morning, Edward, how you doing?

Edward Bass
Yeah, very good. Despite the sudden turn to rainy British summer under lockdown, I’m pretty good.

James McCarthy
Well, of course, it’s the eighth of July. Therefore it has to be rain doesn’t it?

Edward Bass
Natural – rainy season!

James McCarthy
Welcome everybody once again to the business of data podcast series. My name is James McCarthy and having spent 25 years in the tech industry, I’ve always been fascinated by the impact of tech on the world that we live in. And recently that includes the growing topic of data. Every company is now a data company and we’ll be talking about the impact of that on organisations and the people that they serve. We’re going to talk about how data should be used to improve the world we live in as well as the risks and moral challenges surrounding data. When it comes to consumer privacy and security. So if talking about this business of data sounds like your kind of podcast then you’re in the right place. We’re set for another interesting conversation today. In today’s episode, we’re tackling the fascinating subject of social insight and social listening. Everyone knows how big a part social media plays in consumers lives. Because of a series of data breaches, though, and the likes of the Facebook and Cambridge Analytica scandal in 2018, any use of consumer data in a meaningful way can be perceived as a bad thing. I thought it’d be interesting to find out whether social insight gathered from billions of people’s public rants on social media is a force for good and how it can help organisations stay relevant and in touch with the views of society, using insight gathered ethically and with the right intentions. I’ve tracked down Edward Bass, who is an expert on this, the founder of a business called entSight. Edward makes a business out of working with a wide range of organisations to put the science of social insight and social listening to good use. I’ll be asking Edward how this works, what can be gained, and how we understand the ethics around the whole subject. Once more, we’re recording remotely due to the pandemic and I apologise in advance if the audio connection isn’t perfect. We’ve got loads to cover, so let’s get cracking.

Right then, Edward, I’ve got loads of questions on this. But let’s start with the obvious one. Can you explain to us what is social insight?

Edward Bass
When we talk about social insight and social listening as well, social intelligence is another term that’s often used is generally talking about using analysis tools to explore conversations from social media. Say for instance, something like Brandwatch or Pulsar, you can dive into more open platforms, ie ones that you know have the data more available such as Twitter or Reddit, blog posts on blogs, such as Tumblr for instance, some Instagram data, quite limited Facebook data, shiney social platforms like Weibo, Wibu as well. It’s effectively taking data from all of those, analysing it and using it to understand human behaviour and preference.

James McCarthy
And is this just data that’s just available over API’s and anyone can grab it, or do you have to do something with those platforms in order to get access to it?

Edward Bass
Well, the platform’s generally tap into API’s or use web scraping. But also they’ve constructed their own kind of boolean based tools, and rule based categorisation technologies just to sort of help you really find literally a needle in a very, very big haystack because one of the areas where social data differs from other data sources is that it is mostly noise you know, you’re tapping into a huge wealth of conversation even if you’re just looking at something in a very, very specific subject. So it’d be a little bit like being at Glastonbury and trying to zero in on a conversation specifically about Wensleydale cheese, you know, you have to, you’d have to do quite a bit of running around – what it’s effectively doing, it allows you to use keyword analysis and even things like imoji analysis, sentiment analysis to zero into that conversation, and as you can probably gather that’s quite not given the fact that it is very noisy, and it’s multi-language and, you know, things like sarcasm, for instance aren’t necessarily easily identified. So there’s quite a lot of work involved in order to get it and the tools whilst they’ve made some very good steps to become more easy to use, still require a lot of work going in, but then as a result of that, you can get some very open and honest and unique specific insights out that wouldn’t be available for other research methods.

James McCarthy
And that’s really interesting, you make the point about Wensleydale cheese so presumably Wensleydale is a place as well as a cheese, what you’re talking about there is extracting the topic so identifying the cheese from the place for example really probably daft example!

Edward Bass
That one literally I just pulled out the air it’s not even a particular type of cheese that I’m a fan of. Yeah, I mean, I’ve learned I’ve learned this over the years getting close to a decade now working with these tools. That kind of separation of language, that kind of detail is often the hardest part. So to give you an example, you know, if you’re working on a product or something to do Mars, you can get rid of quite a lot of mentions of chocolate bars. And then there’s things that you just don’t know until you suddenly know them. So a few years ago, I was working with a project for BMW, and entSight just to give some background began in the entertainment sector where there is a wealth of conversation to explore. And we were doing analysis around Spielberg movies that also related to automotive and there was some very strange spikes in the data around a particular time periods that didn’t correlate to movie releases, and I found out that Spielberg is also the name of a racetrack somewhere in middle Europe as well, which I had no idea about before. So those kind of anomalies that you have to go in and tidy them up. But of course, I couldn’t have done that. I could have created a boolean programming to say not you know racetrack without having the prior knowledge. So there was a lot of filtering, there’s a lot of contextual understanding – one of my favourite examples that I use all the time, but it really drives the point home is if I’m doing a project about medical behaviours and response and human health, and I’m doing a project about teenagers response to video games, the word sick is going to take on very different connotations. So you need to have a good understanding of topic and understanding human behaviour, a great understanding of language and semantics and then you’ll probably about halfway there.

James McCarthy
So it’s not just the data that you’re getting it’s the intelligence, the enrichment that you’re putting on top to understand what it’s about. It’s interesting actually, because our software is we’re not here to talk about it today, but our software is data discovery software that works on people’s data on the inside of their organisations rather than social data, and we have very much the same thing. So we have topic extraction, and we use natural language processing to do stuff. But you’re right, you have to understand the data set. And you do get some quite interesting anomalies and things that you need to go into. So it becomes a bit of science, doesn’t it really, at the end of the day?

Edward Bass
Absolutely! There’s been a lot thrown at AI, which is, in most cases, just machine learning to figure this out. And actually, whilst the basics can be spoken to, I think that still the more complex analysis takes a human input. We also within entSight tend to work with Google Search data and critically, most actively with survey data from the likes of global web index, to kind of rationalise some of what we’re seeing, and sometimes we’ll do that in advance of doing the social media analysis because we need to understand the What? before we understand the Why? if you like, or we need to understand the question before we understand the answer to get very kind of Hitchhiker’s Guide to the Galaxy on you – It’s very much that, you know, it is great looking at loaded conversation, but actually, you don’t understand from the quantitative behaviours first, you don’t necessarily have the guidelines to figure out what you’re seeing.

James McCarthy
Agreed. Okay, you mentioned Facebook earlier, I’m going to ask this now. So we’re talking about social insights. Sounds to me like that’s mainly publicly available because you’re scraping stuff that’s in the public domain anyway. But how do we distinguish between what happened with the scandal involving Cambridge Analytica where social data was used, you know, when let’s not go into the detail of the how that worked, but using people’s data has a bad reputation or is gaining a bad reputation. So how do we differentiate and distinguish between social insight for good versus social insight that isn’t ethical?

Edward Bass
Yeah. Okay. So first of all, it’s fair to say, I think that like every tool set, it can be used, social data can be used for good and it can be used for bad, very subjective terms, of course. But, you know, ultimately, there’s nothing inherent in the prices of the software itself, that is either good or bad. It’s just the human being to the using it and their agendas. And I think really what happened with Cambridge Analytica and is still happening with Facebook is really down to a degree of naivety that exists in the platforms themselves, they are still very much new platforms. They’re born as much of the original internet was out of a very much idealistic, fairly pure bits of the hippyish West Coast ideology. And I think really what happened with analytic and Facebook is they simply never expected that it would be co-opted in that way. And everybody was quite happy as they often are, to just look at the revenues coming in and not question, you know, not ask questions like, it’s quite interesting that we’re seeing quite a lot of revenue coming from St. Petersburg, for instance, that kind of thing is I think, when you’ve got that it’s an entirely new format, so we don’t know how it can be abused or not abused. And then also, what’s happening is actually generating quite a lot of revenue as well. They turned a blind eye to it ultimately, but this is still going on. I mean, you know, you’ve got an embargo at the moment from major brands around Facebook and hate speech. When Jack Dorsey much to his credit, and certainly something that’s improved my perception of Twitter has pulled all political ads from the platform and just said, Look, we’re going to come out of that we’re not going to do it. Whereas Facebook is still continuing to do that, to allow that and to, you know, use things like freedom of speech. But it is naivety I think it’s, you know, secondary to that, because it’s a new technology it’s not going to be self managing. And the people running that technology and responsible for that technology are always going to be lightyears ahead of anyone in government who could potentially control the impact of it. And then when you’ve got a situation, even crazier where governments or political entities are then basing their political campaigns on that platform, they might not want to, you know, they might not want to rein it in, so there’s not as well. But to flip it over, there’s a lot a lot of good that can be done with this kind of understanding as well because you are ultimately tapping into very honest assessment of human behaviour. So I’ve personally worked with Public Health England, for instance, to help around you know, understanding what the barriers are to people giving up smoking- a really interesting project, like Freud’s with entSight recently, we’ve done work with brands to really help them understand more about the sort of perceptions around sustainability and social good that their audiences have, rather than it all just being oh, they’re gonna respond to sustainability. Well, sustainability could mean a number of things. So you know, understanding the specifics of that but really, to wrap it up knowledge, you know, knowledge is power and power can be used for good and bad and I think there’s just been more instances of the bad being reported because it’s newsworthy, yeah, I think it’s the responsibility of people like Facebook and Twitter to rein that in and to create a better public view of themselves. I often believe that Facebook’s PR is terrible, really and it doesn’t really do itself any favours – it comes across as kind of belligerent and arrogant, for the most part. But ironically, this is the thing as I mentioned earlier, Facebook isn’t one of the platforms that the likes of myself or anyone useing brandwatch or pulser or audience, we don’t have access to that data – it is relatively locked down. And Cambridge Analytia got access to it, really via something of a loophole. And it’s worth saying as well, a lot of what Cambridge Analytica claim to be able to do, they didn’t actually do, because it wasn’t this other platform. So they were somewhat hoisted by their own petard by being a bit too boastful about some of the bad stuff they were doing. So there you go. That’s, that’s my long answer.

James McCarthy
I think what I liked about that, and I’m not I’m not an expert in this like you are but you know, my sense is that if you’re gathering the publicly available data anyway, rather than dipping behind someone’s API and pulling out personal data, which is a bit different, then the individuals that are creating that data must know that it’s public, and therefore it’s slightly different, isn’t it from an ethics point of view, because if you’re in Facebook and you choose to message and post and whatever, then you know that that’s publicly available, you are making that available. And that’s something that the consumer has to recognise, the individual has to recognise their responsibility to only say what represents them, etc. So I suppose that helps too.

Edward Bass
I also think that it’s really important to understand that no one is working with individuals data. It’s not like all, you know, a survey, or even a focus group can’t operate on the basis of one person. The real value of using social data is operating at scale. And, actually, you know, I mean, everything we do with entSight is anonymized. When we create a report for a client, we have made it a rule for the team not to actually use identities. Because we don’t have permission to do that, unless it’s a celebrity or a brand, in which case they’re in the public space. But, you know, Hannah from Glasgow suddenly finds out that she’s being used in an Ogilvy presentation for Coca Cola, she’s probably not going to be that happy about it. So even though there’s nothing wrong, that’s not seen as being – oh, they could say, oh, well, she’s on Twitter, but it doesn’t matter. What matters is that we’ve got sort of a public contract in place there and we need to be speaking to that. Again, I think this often comes about because it’s technologists, who tend to be very, very focused on the end results but not on the implications and impacts. That tends to, it tends to disregard some of those but on the other human level, we have to, you know, we have to just just be a bit more sensitive.

James McCarthy
Yep. Which is, which is ethics in play rather than ethics on paper as well, it’s the decisions that we make when they are managing that data that matters. I guess, something you said was really interesting…

Edward Bass
Remember, it’s human beings by that date, remembering essentially, you know, yes, even though you’re looking at a load of information that’s been rolled up from maybe conversations of 5000 people, it is all coming from individuals, ultimately.

James McCarthy
Yeah, but it’s interesting the use of it is aggregated. So in any case, it is coming from individuals, individual human beings who should be respected, but also actually the value to an organisation or whatever is more around the aggregated data in any case, which is an interesting point.

You mentioned how, in social people say what they actually think, rather than how they might respond to a survey. That’s a really interesting point. So I’m a marketer and in my world, you know, there’s a really interesting stuff around how people, when they respond to surveys, they answer questions based on what they think they should answer, and what they think their view should be. In other words, they answer questions very rationally. Whereas in reality, when they make decisions, actually go and do something they do so emotionally, and therefore what people say they’re going to do is very different to what they do, which is, you know, it’s borne out in things like polling data. So if you remember the the general election A few years ago, where the pollsters got it completely wrong, because what people said they were going to do in terms of voting intention, they then turned around and when they were standing in the polling station did something completely different. That’s quite interesting and I suppose social data, therefore, is is very useful as a source in the sense that it’s what people actually think in the moment, rather than what they might answer if they’re answering for more questions on something.

Edward Bass
Agreed. And I do think, you know, the more visible the person makes themselves, the more that there’s a performative aspect. So I think, you know, when you’re looking at conversations from influencers at the end of the day, they are trying to project but generally, when you’re just tapping normal conversation from normal people, you’re seeing a much more honest view. And also, you know, I’m, I’m gonna be very open about the research profession, especially traditional research, probably not going to be very popular, but ultimately, a lot of research is carried out to verify ideas and concepts and what I’ve always liked about social data is it tends to throw a spanner in that work, which is very good at actually forcing people to go back to the drawing board. And in some cases, you know, we’ve had to be quite tough with some, especially brand clients, about their expectations and what they wanted to see versus what they’ve seen. I’ve had to say that there’s any number of research companies that you can go out there and pay to just rubber stamp your ideas. But ultimately, you’ve come to us to be transparent and honest and, you know, we don’t we don’t work with data because we like pretty graphs. You know, we’re interested in human behaviour by me personally, I’m interested in, you know, actual truths and not just because there’s some kind of crusade for the truth it’s because if you have an accurate insight about human behaviour, any strategy based on that has a higher chance of being successful, because it’s reacting from the truth and marketing has struggled with that.

James McCarthy
I think you’re right. I mean, it shouldn’t be research shouldn’t be used to ratify decisions. But it very often is. The other reason people go down the route of research is to get very forensic into a particular audience and a particular scenario that relates to a particular product, or usage or whatever. Now, my question to you, therefore, as a marketer is when you’re using social insight, how granular are you able to get on behalf of a brand, you know, on a certain topic, I’m really interested in how you would then apply it, if that makes sense.

Edward Bass
There’s really a few factors involved at any successive granularity there. Assuming there is a good amount of conversation, varied conversation available, there’s potential to do it. So we’ve been tasked with products in the past that aren’t generally talked about and you’re at something of a loss to be, you know, I mean, hypothetically speaking, this isn’t thankfully one – If a toothpaste company wanted to come to you and understand experiences with toothpaste, it’d be very idealistic and somewhat foolish to believe that Twitter and Reddit is going to be full of conversations talking about their teeth. Brushing experience doesn’t exist. However, if you’re looking at conversations around interactions with entertainment, which is one of the reasons why we work in that space, because there is a wealth of data to work with or automotive experience or travel or brand perception or, you know, sustainability and environment or politics as we learn in Cambridge Analytica, then yes, you’re you’ve got much more than opportunity. So there’s the volume of data available. You also as I infer, you have to put the time into it. I mean, you can get very granular. But getting granularity and maintaining scale can be quite difficult, especially in, you know, specific regions as well. So if you’re only looking at, you know, people in Ontario, for instance, you might find, but it’s very much dependent on the existing conversation that exists. Without that you’ve got nothing to work with. And I think that that’s been a hard realisation for some organisations working in that space is that it’s, you know, we’ve had even relatively well funded and media driven campaigns, even ones involving celebrities. We’ve had brands come to us saying, Oh, well, you know, what was the conversation around this? I mean, people have watched it on YouTube, we can tell you that but no one’s actually been talking about it because it’s an advert, and let’s be honest, you know, people don’t! They’re inspired by ads, you know, investigate and buy and even advocate maybe adverts but they don’t generally talk about the content in the way that they might talk about Game of Thrones or, you know…

Exactly. How many adverts honestly are very good adverts? So, I think if you if you do do a very good advert and you do generate that if you if the brand or the organisation does do something to stimulate conversation, then yes, you’ve got a good opportunity there for granularity

James McCarthy
All right. Okay, so let’s reverse back out. And by the way, you know, you were saying before we started recording this that you might get an Amazon delivery? I’ve got my wife downstairs telling off my 12 year old son for not wanting to do any more schoolwork, so it’s all happening here in our in our new world. I hope nobody can hear it on the podcast! I think he’s done fairly well, to be honest. But anyway, there you go. So I just wanted to back out from the, from that sort of granular question I had to the more big picture. You mentioned politics, obviously, there’s an awful lot going on in the world today, you know, Black Lives Matter, we’ve got obviously a massive pandemic, that’s, you know, fundamentally changing the way the world is operating, and all of those things merging together as well bouncing off each other. You talk about, you know, brands and organisations in general, you know, and how they can use social data, I’m assuming that then in their sort of big picture discussions, trying to keep in front trying to stay relevant, particularly when the world around is changing so quickly. Social insight must represent a better source of information for those sorts of things.

Edward Bass
Yeah, it is. I mean, there’s there’s always been a fascination I think from the brand side and research side, looking at social data as being kind of trend analysis, which I agree is really a good use of it. Especially when you have shifting paradigms that we’re going to see for the next few years as well. And it’s, you know, we’ve got a major US political election coming up as well, which is always, always is going to spin the dial around as well. You know, when I think about how much the word Trump has dominated every word cloud, even in unrelated subjects, it’s taking me not to want to say negative stuff about the guy, but we’ve come to regard that almost concept as a virus because it just corrupts the outcomes, abrupt conversation and it is conversation to a single point so yes, I think that trend analysis is very, very important, especially right now. And where there is actually a need to understand progression. What is often overlooked, though, and not factored in is historical data, and really much of the work that we’ve done in the past five or six years has been not just looking at the here and now but how it contrasts and benchmarks to the past. But how is conversation on a particular topic from a particular audience changed over the years – the fact that everyone’s talking about, you know, the fact that everyone’s talking about Black Lives Matter right now, of course, it’s relevant, but there are other spikes in conversation. In Black Lives Matter going back for years, it’s been a terminology and the hashtag that’s been in the public consciousness for a while now. It’s returned to the public’s consciousness at scale. But there’s a lot that can be learned from the impact of it right now, from looking at it historically, so I think there is definitely you know, social insight very, very good at that. But there is also a need to tap into historical data, just to contrast and understand, and understand how a trend actually grows and evolves, where it’s going.

James McCarthy
Okay, so let’s apply an example to how an organisation might do that. So how can an organisation own an environmental or social challenge, you know, based on their audience and their perceptions and beliefs and then use social data to to enhance that?

Edward Bass
Okay, yeah, ‘m actually going to refer not by brand name because I’m NDA, but to a couple of projects actually that we’ve been working on in the past few months. And going back also to the idea that I mentioned earlier, when we talk about environmental sustainability and social good, these are very broad terminologies at the end of the day, they are not single concepts, they’re not encapsulating everything that, you know, might even be associated to them. So you have a brand that has an audience and in one case, we have a brand that we’re working with quite a large brand, who were prospectively looking at developing a new product focused on more sustainably orientated, younger audiences. And also more recently, we’ve been working with a brand that has an established audience and established product, but is trying to understand in detail a lot more about what that means. They know vaguely that their audiences would buy eco friendly products, they know vaguely and we’ve even ourselves verified with data from the likes of global web index that you know, they believe brands should be more about more in terms of social good and sustainability, but they don’t know the specifics. And they need to know the specifics, especially in fashion, because otherwise they could be imagining that their messaging is kind of vague. Or in the case of I’m not going to name the brand, but there was a major fashion brand whos collection earlier this year or towards the end of last were criticised by the press and public and fashion experts alike for being effectively kind of hollow and performative because they were doing this kind of ‘Oh, it’s sort of environmental type fashion show’, and everyone quite rightly was like, but you’re not actually, you’re just making a bit of a play about the broad attorney, sort of a cliche. It looks like you you’re just trying to jump on the bandwagon, and then you’ve got no real dedication to it at all. And it’s also worth bearing in mind that the more affluent customers tend to be the most savvy, and therefore they are not the ones that you need to be alarming or annoying. So if I look at the case of the most recent work we’re doing, where it’s an existing brand who is trying to understand, it’s been a case of understanding, say that their interests, at a broader level and then drilling down into social data and identifying things that we know that the older audience themselves is very interested in charity and volunteering around social good and sustainability, environmental causes. So we’ve explored that in more depth to get into the specifics of well what, you know what particular causes are they likely to be most interested in and we also know that almost is a reflection of themselves, they expect the brand to be acting out in this kind of volunteering as well. So we’ve suggested that you should use your social and content channels to engage, to help to help guide them to charitable volunteering, what they could do, you could actually use your marketing resources to be more proactive in sharing organisations that might be looking for volunteers, for instance. And at the moment, what we’re really getting into is what what are the kind of key you know, is it is it things like single use plastics? Is it is it sweatshops? Is it you know, the localised environment and into the fixing and mending in that case as well? So it’s, it’s really that case of getting into the detail we know that the audience’s perspective In a bucket marked sustainability and sits in the bucket, Mr. Associate good, but we are not going to get. And again, if I’m honest, given how I sort of do good at those are, yes, we could run surveys out to look alike audience or even their own audience about that. But the chances are they’re going to respond to those questions in a survey how they want. So they can say yes, of course. Yes, of course it is. But they’re not, you know, with at least this many social data. We actually know what there…

James McCarthy
is back on my other point, isn’t it, which is, you know, the poll data, you know, what you think you should answer when you’re asked versus actually what you’re going to do when you’re standing in the polling booth, which is the same thing, isn’t it? So how you’re actually going to respond is up to you and it’s personal to you, but you’re more likely to be honest on social rather than responding to a brand who’s asking you a question.

Edward Bass
Yeah, it’s worth mentioning though, that because we do work with other data sources, when we see that data overlapping, where we see insights matching up, we get quite excited because we know that we’re looking at a truth, we know that it would be too much of a coincidence. I’m going back to the automotive examples, we were analysing search dates around automotive and entertainment partnerships, Mercedes and Jurassic World, BMW and Mission Impossible a few years back, and we were matching up search data. So people searching for those those associations with the kind of social conversation, and then also some sort of broader survey data around perception of automotive and entertainment. And we were seeing just amazing overlaps in terms of the search, there was a kind of a delay, but yeah, it was a delay that you’d be able to attribute into fairly normal human behaviour. So it’s when we actually see verification across different data sources, and that points to another point, I think using social data in isolation, and honestly I did for many years is great up to a point. But there are avenues and aspects of human behaviour that you can’t verify or understand just through social. So for instance, just to quote something I’ve talked about, an IBM event last year, no one goes on social media and describes their customer journey that they have to buy a holiday. They don’t go on Twitter and say, Oh, yeah, well I searched and then I might ask some friends. No one does it and it would be boring and you know, no one’s going to use their Twitter feeds to just instruct the void of people, the global audience about how they bought their holiday to tenerife. But on the other hand, they will respond to questions about that customer journey equally, trying to find out a particular audience’s preference for a celebrity or entertainment property, or even a you know, a stainability cause that might be very difficult because you can have – imagine it’s a celebrity and you asked an audience of 500 people, ‘who’s your favourite film star?’, you can’t do that via a survey because the amount of answers would have to be able to run into the hundreds. And people are going to look at the top 10 names and just pick one of those to move on to the next question, so that they can edge closer to their for free Amazon voucher or whatever it is. So it’s the kind of information that you need is very, very much and again, when there is a wealth compensation on social it’s much better to tap into to that, as I said, around entertainment or causes environment, politics…

James McCarthy
Or more about experiences that people are talking about. So when it comes to their holidays, their favourite bit, or whatever it might be.

Edward Bass
Yeah, exactly, exactly. It’s, you know, the experience, absolutely but you know, something like customer journey, we’d always go to global web index for that, because they have the data points that are collected in an appropriate way to help to analyse them.

James McCarthy
And I suppose there’s organisational data that that organisations may have. So some of those brands may have data that they can gather from how people have interacted with them, you know, during customer journeys, I suppose that they can then map on top of these things, and you can start to build a better picture.

Edward Bass
I think this is where a lot of really good data science work is being done at the moment is around customer experience and internal and external data because you need data scientists to do that kind of work because it’s working with huge, sometimes noisy, sometimes complex, sometimes very distributed across the organisation, data set. So yeah, it’s that that’s where the real time is. But yeah, bringing that in and I’m constantly interested in the interaction between email, social, CRM, and how behaviours sort of own behaviours might match up, or even differ to social media behaviours. We recently had a conversation with a client and they were saying all the emails do really well, but but at the same time, we were talking about what social media is very good for brand discovery, whereas your email is really dealing with your ongoing relationship with a consumer. So it’s you know, it’s important to understand where the data plays out what is relevant?

James McCarthy
Yeah. Amazing. All right. Listen, Edward, t’s been great to chat today. Really fascinating topic. We could talk for hours about this I think!

Edward Bass
I know normally do!

James McCarthy
Thank you so much for your time and for sharing your knowledge. Who knows where the world is going to find itself next but whatever happens, I’m sure that social insight is going to have a huge role to play in helping us steer us along in the right directions, so thank you very much. What else have you got planned for today then in the rain?

Edward Bass
I’m off to darkest Norfolk with my partner to go and see some friends for social distance catch up.

James McCarthy
Hopefully the rain will clear up and I think there’s a test match that they’re trying to kick off in Southampton as well, so who knows how that’s gonna happen, but English cricket is back!

All right. Thank you very much. All the best.

Edward Bass
Thank you and really good to talk.

James McCarthy
And there we have it. I really enjoyed that conversation, an area of data I’m not an expert in but very much of interest to me and hopefully you found it interesting too. Thanks so much for listening to this episode of the business of data with guest Edward Bass from entSight. This podcast is brought to you by Exonar, delivering world leading software for data discovery in organisations. If you enjoyed today’s conversation, please subscribe to more from wherever you prefer to consume your podcasts. And that’s it from me. I’ll see you next time.