Less Chatter, More Matter: The Communications Podcast

#72 What comms professionals need to know about data (ft. Roelant Vos)

• Mel Loy • Season 1 • Episode 72

Share episode

By 2025, it's estimated that the volume of data or information created, captured, copied and consumed worldwide will reach 181 zettabytes. That’s the equivalent of 4.98 quintillion movies, or 78 times the length of the Milky Way.

Now you may be thinking: firstly, that’s a heck of a lot of data - and you’d be right; but secondly, you may also be wondering why this matters for a communications podcast or to a comms professional?

Well, that would be because as communications professionals, data surrounds us everywhere we look, and it is a critical tool to make sure our work is the most effective it can be. And that is what today's episode is all about. 

In fact, we’re so sure of this implication to our work, that we enlisted the help of self-confessed (and proven) data extraordinaire, Roelant Vos from Agnostic Data Labs, to talk us through what data is, where to find it and how we can be using it as comms professionals. He also shared some of the challenges with using data that we need to be on the lookout for, alongside the tools that we can leverage to make it even easier for us to get ahead of the game… including AI!

So don't get left behind and sit back, relax and enjoy this chat with the wonderful Roelant Vos.

Links mentioned in this episode:

Say hi!
Follow me on LinkedIn
Find out what I'm up to Instagram
Check out my website
Ask a question

Mel:

Hi, and welcome to Less Chatter, More Matter, a podcast about all things communication without the waffle. I'm your host, Mel Loy, and in this show, I will give you short, punchy, practical communication tips and insights You can start using in your communication practices right away. I'm a former corporate communication executive who happily took a redundancy, started my own business and never looked back. These days I use my 20 plus years of experience to help guide organisations of all shapes and sizes in how to communicate more effectively. I'm wife to Michael, cat mum to Cookie, aunty to 12 nieces and nephews, a yoga teacher, and a group fitness fanatic. I promise these episodes will always be short, sharp, and helpful, so let's get amongst it. By 2025, it's estimated that the volume of data or information created captured, copied and consumed worldwide will reach 181 zettabytes. If you're not sure what zettabytes are, well, 181 zettabytes is the equivalent of 4.98 quintillion movies. To put that into other terms. If each one of those movies was a DVD, and you stacked them on top of each other. That stack of DVDs, it would be 747 trillion kilometers high. Well, roughly 78 times the diameter of the Milky Way. Why does this matter? Because as communications professionals, data surrounds us everywhere we look, and it is a critical tool to make sure our work is the most effective it can be. And that is what today's episode is all about. Hello! And welcome to Less Chatter, More Matter, the communications podcast. I'm your host Mel Loy, and on today's episode, I am delighted to bring you an interview with someone... who might seem a bit unconventional for a comms podcast. I'm speaking with a data geek. My guest today, Roelant Vos has been a consultant, trainer, software vendor, and decision maker in the corporate world. And over the years, Roelant has observed data management from many different points of view. Throughout his career, the common theme has always been a passion for data automation and cogeneration finding and using repeatable patterns to interpret data. Roelant has always felt that this is the key to making data solutions easier, more manageable, and truly flexible. And he is currently writing a book about these topics called Data Engine Thinking, and has started a new company, Agnostic Data Labs that provides tools that can help implement data solutions using these ideas. In this interview I asked Roelant to give us the one-on-one in plain English. On what data is, where to find it and how we can be using it as comms professionals. He also shared some of the challenges with using data that we need to be on the lookout for. And some simple ways to use AI tools like Gemini or Chat GPT to help us get the information we need from data. Without bugging the IT team. And if you're still wondering if this is for you, well, if you want to remain relevant as a comms pro, then you need to know where you can find data, how you can make sense of it and how to use it to your advantage. This is what is going to start setting you apart from the rest of the pack, as we enter this new AI driven phase of our profession. So don't get left behind. Sit back, relax and enjoy this chat with the wonderful Roelant Vos. Roelant, welcome. Welcome to the show.

Roelant:

Hey, thanks. Thanks for having me. Hi Mel.

Mel:

Hello. So you're joining us all the way from where today?

Roelant:

Uh, it's actually local in Brisbane, so not as warm as I was hoping it would be, but yeah.

Mel:

And where do you hail from originally?

Roelant:

The Netherlands.

Mel:

Beautiful. The home of poppies, I believe. So tell us a little bit about you. What do you do? How did you come to develop your expertise in this space?

Roelant:

Yeah, well, um, it's, it sounds a little bit cliche maybe, but you know, as a kid, I was really drawn into computers and gaming and coding and taking computers apart, assembling them again. So for example, as a, as a kid, I would, um, I would take a notebook, like a paper notebook and then write code so I can later key it into the computer. So really, really drawn in at a, at a young age. So it's no surprise that at some point I started studying computer science in the, in the late nineties. And. It had really small subjects about data and, and, uh, interestingly, it still is to this day that the, the, the focus on data is relatively small in the computer science world. What I did learn there, I really liked because, um, there were techniques and concepts that really focus on, on how can you take something. Like, like, like data and organizing in such a way that it's just a bit easier to, to handle, for example, a process like normalisation, which is really an algorithm to structure data efficiently. And I, I really liked this idea and the design and I wanted to, to do that. And I did.

Mel:

So what do you do now with data?

Roelant:

I am mostly focused on, I guess, the term data management in general, which is a very broad... principle of, of how, within a company or within an organisation or context, how we define a data element. Where's it coming from? Who should be worried about it? How is it used? How can we combine it with other bits of data to draw meaning from it? And it really means that we have to find people that, um, worry about it, right? So we need people that, that, that, that own it. And then we need to make sure that all the processes and the systems around it generated in the way that it's, it's really useful. So it's, it's like this. Morphing of the company to, to, to make the, I guess the data that is generated work for them.

Mel:

Gotcha. So let's go back to basics maybe... what is data in terms of when we think about a company like the ones that you work with, where do they find it?

Roelant:

Yeah, that's it. That's always such a hard question. Um, the way I look at data, um, I, I always call data the breadcrumbs of human activity and it's... it's something that is created as a result of something happening, right? There's a business process. There's an event, there's a transaction, there's something happens and data is created as, as a by product more or less, and that, that, that lands somewhere -and then we need to dig it up to find it and do something with it. The way I explain this is to compare this to the work that biologists and paleontologists do, for example. So imagine you're, you know, you're digging in the earth, trying to find fossils, all these different layers of the earth and you find a fossil and you think, is this a, is this a thigh bone? Is this a giant tooth? Is this a little pinky of an even bigger dinosaur? You're like, well, I, based on what I know, I need to, I can probably say it looks a bit like this. And then you, you dig further and you find more and... slowly, but surely, this, this picture starts to emerge of, of, of, of this dinosaur you're looking for. And it's, it's combining these, these different disciplines and different perspectives. They form an understanding of your reality over time. And, you know, you, you have to go back and, and challenge earlier assumptions and change your views on what you think your dinosaur looked like. And I think it's fascinating because, Imagine how often or consider how often the tree of life has changed as a result of genetics and research. And data is exactly that, right? You dig it up, you find it, and you're like, ah, that's just stuff. I don't know what it is. I've got no idea. And then you find more and more. You talk to people and you try to trace back where it's created from, which process does what, and, and this, this context, that's what gives it meaning. And that's what makes it useful. So the data is, it is the, the, the, the undisputable fact that something happens and once you understand why and how and what made it happen, then it becomes something that you can use.

Mel:

I love that analogy of the, uh, the dinosaurs and, and yes, that's make a lot of sense. And you know, from my own perspective of working in corporates for years, you know, there was a lot of data that came through from customers, for example, and their activity online, uh, and their activity in stores, but even employees as well. Is that kind of the data you're talking about as well? Like all the, that sort of information that we store and on all sorts of systems in a workplace.

Roelant:

That's right. Yep. And it's, and, and, and this, this analogy sort of works. There as well, because you, you get, you get what's recorded, but it might not be the full picture and it might not be the complete data that you get. But yeah, if a customer enters the website, it leaves tiny breadcrumbs and we collect those as data points.

Mel:

And it's, it's challenged too, isn't it? Because sometimes, uh, we make assumptions based on that data. You know, we put two and two together and end up with 16. So how do you know that you're, I guess, yeah, getting the right story out of what you're seeing?

Roelant:

I always look at it as something that's, it's an assumption at a point in time that, that will change. So what you do with data is you, you collect those events, those, those facts- those breadcrumbs- Roelant: and then you apply a model on it. You say, based on what I know, this is, this is what, what meaning I can derive from it and that, that model and, and a data model is a very commonly used term in, in this space, right? It's, it's like your interpretation, your perspective on the information that you've collected. So you start with an assumption and then you say, ah, it sort of worked or it doesn't work or like, we're not really sure. So we need to go back to the, the, the infrastructure, the, the IT environment and say, can you collect this more please? Because we, without more of these breadcrumbs, we can't really make a correct assumption, or we're not really certain that we derived the right meaning from it.

Mel:

It's a bit of to and fro by the sounds of things.

Roelant:

It's a process.

Mel:

So why should communicators like myself, you know, like PR professionals, internal comms teams, Government relations teams, media professionals. Why should we care about data?

Roelant:

So, so data is in the, in the news quite a bit. Uh, so I think as a society and everybody, we, we hear a lot of big data, AI analytics and things and opportunities that it gives and the way it changes people's lives and that's all relevant. And that's all true. Uh, it obviously needs lots of data and there's, there's, there's plenty of data to go around. Um, but I think the real value for, uh, for everybody and communication specialists in particular is that what the data says at the, at the basic level, it's the livelihood of the company. So you can argue about how to interpret certain things, but you can't dispute that something happens. And I think coming back to this starting point is, is important for communication because once you roll out these different interpretations, you can always go back to the, to the facts if you have them. And I think there's a lot of value for communication specialists to appreciate the difference and to craft around, craft your message around, you know, what perspective you take on the events that have happened, if that makes sense.

Mel:

It does. Yeah. It's almost like, uh, I guess the, the evidence that can, or the measurement that can help you go, okay, well, we should be targeting our communication more towards, for example, this particular audience at this particular time, or that particular piece of communication didn't resonate well, and we can tell that because of X, Y, Z, you know, whatever, we got back. Um, what are some of the challenges though with accessing and using data to inform communication?

Roelant:

The. Issues in general with data are often called or referred to as this container term data quality issues. And it can be varied. And the biggest issues is that it's, it's sparse. So you don't have enough to enough steps of the process to really follow it. So you, for example, you can say, you know, a customer entered my website and looked at this product and bought it. It might not do that for every customer, or it might not do it consistently, or it... might not capture the steps where the customer was looking at 20 other products because the systems need to be geared for that. So, so you always have this initial, um, initial collection of data that, that might just not be complete enough to, to draw meaning from it. So it goes back to understanding the process and then saying, Hey, if, if this is what we want to do, then we need to capture these... these events as well. And that's, that's the biggest problem because that means you have to go back to it and back to the business and say, let's just change your systems and change our way of working and because then I can really report on this campaign.

Mel:

And my stuff is the most important right now. Prioritize me. So yeah, that's huge.

Roelant:

That's, that's, that's a big challenge. So, and then of course you've got the technical barriers and the semantic differences because everybody has different... ways of explaining the same thing. So you need to align that. The thing you can do though, is that is to be proactive in that and be very specific on what it is you're trying to do. So you have this objective to, that you want to achieve, right? So to optimize the campaign, to make sure that, you know, you're, you, you test out different versions of the same text to make sure which one works best, all those kinds of things, and it'd be very precise on what is the data I need for this, what are the data points, what do they mean specifically? And then you can influence the process to, to make it happen.

Mel:

And on that, one of the challenges that, um, we actually talked about this on a podcast episode a few weeks ago is how you take, you know, data, complexity, ideas, and make it meaningful for your audience. So you might be collecting all this data and facts and you're putting together, putting it together. But how do you actually communicate that in a way that makes sense to, you know, Joe blogs. Who has never had to deal with data or numbers before, is that, how do you, how would you recommend going about that?

Roelant:

Yeah, look, one of the things, um, that is important in working with data is, is the visualisation. So it's, it's about collating or aggregating data information into a meaningful key figure or statistic or diagram and things like that. It's not really my strong suit. I'm always sort of focused and drawn towards the technical back end. But there's a whole science around how to convey complexity in, uh, in relatively straightforward terms. Right. And, and I know people that are really, really good at that. So it's, it's visualisation. So it's creating the right type of chart for the audience. things like that.

Mel:

Yeah. And, uh, I think you're right. The visualisation is so critical being able to take complexity and a complex idea and a complex set of numbers and data and put that in a visual that people understand is... it's such a core skill of communication. It's, it can be, it's something you've really got to work on. I just want to ask you as well, what are you seeing in terms of the impact of AI on your world and how we, A collect data and then use it?

Roelant:

I really like it. It makes a lot of things so much easier. For example, if you have an Excel sheet and you say, just, can you give me the, you know, the, the summary of this value transposed against this other value and this particular angle that I want to look at, it does it all for you. If you say, write me some, some code in C or Python to create this, uh, this, this algorithm that predicts the next purchase or whatever you want to do, it creates the code for you. It makes it a lot easier. And it doesn't always do it well, it almost never does it correctly, but it gets you there like 80 percent of the way. So I think it makes the, the, the barrier to entry for anybody to get their hands dirty in, in the raw data it lowers it. So I think there's a lot of value in that because then it makes it possible for people without a hardcore technical background to, to achieve those results and have those conversations.

Mel:

And obviously a lot of the AI tools like chat GPT and those sorts of things rely on data to be able to generate information back to people. So I see that kind of two way street there and you mentioned it's a good way to break down some barriers to entry for people. What are some simple ways that communication pros could get started with using data?

Roelant:

So I, I, I think that for most data analysis, some level of coding is required. So there is always this technical component, but it's also really easy to learn and things like Chat GPT and Gemini makes it a lot easier too. So if you watch a YouTube video, play around things, you know, for two hours, you can figure out how to access systems to collect data and to understand what pros are happening. And I think. That's really, that's a really useful skill for anybody to have. It's scripting, coding, those kinds of things, including for communication professionals. So the ability to automate some of your work and to go and collect some of the information without going to IT or data teams, that's super valuable. And then you can try out what these different interpretations mean and think about what kind of message you want to convey with it.

Mel:

Are there any tools you'd recommend for people to get started?

Roelant:

Yeah. If you, uh, if you, um, if you want to go down that, that route, working with, with Python is probably the, the, the easiest, the lowest barrier.

Mel:

And what's Python for those of us who,

Roelant:

uh, Python is a, is a, is a coding language, uh, open source and you can download it for free, use for free. It's, it's very easy to um, to get started with many examples of, uh, of, of how to interpret the information. But I also appreciate this is not for everybody, right? So if it's not really your thing, a really good entry point is looking at some of the web analytics tools that are available. For example, Google analytics and, and things like that. You've got free AB testing environments where you can route or have similar process follow different routes and compare which one is best used a lot in a text communication, uh, user experience and those kinds of things. So you can get up your own demographics and, and, and work with that. But doing it yourself really gives you the appreciation of, of what it means. And I think that's ultimately the goal, right?

Mel:

Yeah. And that A B testing is so, so important for, you know, I see it used for people for their websites, but even internally, uh, especially if you've got quite big, companies, , with a lot of people. Um, you know, being able to test ideas and systems and messaging with different groups of people is just invaluable. You know, otherwise you don't know what you don't know. So, um, yeah, I love that idea. Um, now you've recently written a book called Data Engine Thinking. Can you tell us a little bit about what that book is about?

Roelant:

Sure. Yep. It's, it's not fully done yet. It's almost done, but it definitely is done this year. Um, we want it to be as fun to read as possible. Spend a bit more time on the styling and the visuals and things like that. So, um, because, you know, data, it can be a bit of a dry topic. So we try to make it as not dry as we possibly can.

Mel:

I thank you so much on behalf of everybody. Thank you.

Roelant:

The whole world in, in, in desktop publishing and editorial things have opened up to me, but, um, so I'm writing this with my longterm friends and fellow data guy, Dirk Lerner in Germany. And. What it is, is it's about a, a framework you can apply to the data analysis and interpretation, um, in a way that, that supports the iterative exploration and modeling of your, data. So basically what I talked about with, you know, you need to learn more and then challenge your opinion. And so if you... if you can define this as an automated thing and then run it, then you don't have to worry about the technical side as much, but you can really focus on the definitions and the talking and, and, you know, collecting the subject matter expertise. And then the, the, the result will be automatically generated for you. And that's it. It's that concept. And. In parallel, I'm writing software that does that. So it takes a bit of focus away from the writing software takes time. But at the very least, we can demonstrate that these concepts work and we can run code to say, this is, this is what it means to be able to do these kinds of things.

Mel:

Yeah, that's awesome. Well, congratulations. Can't wait to see it, especially if it's going to be very visual. I'm a visual learner. So that's right up my alley. Now I have three questions. I ask everybody who comes on to the podcast. Are you ready for those?

Roelant:

Sure.

Mel:

All right. Let's do it. The first question is, what's one of the best communication lessons you've ever learned and how did it change the way you approach communication?

Roelant:

So I am, I really like Michelle Bowden's materials and trainings on persuasive presentations. So I went to this course and it was one of the best things that I did training wise, I think. And I really liked the way you can structure a message in very specific steps. You know, uh, starting with the icebreaker, leading with the, the, I guess the decision or the, the, the statement you want to, uh, to make, uh, prepare for any bias and that providing the arguments and all that, that, that whole framework around it and then searching back to the, to the key message. I use it for every presentation, right? So every time I do a presentation, that's, that's the structure I follow. So that, that's probably my biggest, uh, biggest learning.

Mel:

Yeah, that's great. And that actually brings us back to, uh, one of our top tips is always have a structure. It's so helpful for the person who's trying to follow your chain of thought, but for you putting together the presentation or the email, whatever it is, to have a structure helps you logically flow through the information and, um, just makes it easier to write as well, to come pull it together, I feel. Uh, what's one thing you wish people would do more of or less of when communicating?

Roelant:

So for me, it's a, so I spend a lot of time in meetings, um, sometimes all day. What I really like is when people, including myself, spent the time to prepare for these meetings really thoroughly, and then spending like maybe a couple of hours to write this decision making paper and figure out the protocols, do their research, send it ahead, and then you can make these meetings or these catch ups really, really efficient and, and it's really I really don't like the time when everybody comes together and just tries to figure it out. It's, it's so much more efficient. If you, if you prepare this and send it, send it and make sure that everybody. Actually read it when you meet up, right? You can start doing that and it, it, it saves everybody time. And I, I really liked it.

Mel:

That's a great tip. And yes, I concur meetings are often the bane of my existence. So preparation is key. One last question. Who do you turn to for communication advice?

Roelant:

Yeah, it's, um, It's pretty much everybody I've worked with in the past. So many people have been so great and so friendly. And, um, I know so many skilled communication specialists that I worked with, uh, worked with, and I'm, I'm, I always stay friends with everybody and, uh, try to return the favor as much as possible as well. So, you know, if, if somebody needs, if, if I need to look at some texts and I know a couple of people that are happy to do that, if I need to, do a presentation. People can test it. And if I need to work on the software, I know people that do the UX for, for, for living and then test, test it for me. And yeah, it, and there's so much talent and everybody's so friendly. So. Yeah.

Mel:

That's it. It's reciprocity. Right. And it talks to, you know, you can't burn your bridges and building really good relationships totally pays off in the longterm. I've seen that firsthand just owning a business. So I totally agree. Now, Roelant, where can people find out more about you connect with you and learn more about what you do?

Roelant:

Um, I've got a website, um, roelantvos.com which has a blog that talks about all these concepts in, in a, you know, arguably a bit more technical detail, right? If you want to do this yourself. Yeah, but that's, that's a really good starting point. Um, and LinkedIn. So, uh, always keen to meet up with more like minded individuals.

Mel:

Fantastic. Well, we'll pop the links in the show notes. Roelant, thank you so much for coming on the show today. We've, I've loved learning from you. It's certainly, uh, whet my appetite to go play with some data. So thank you very much. And, uh, thank you again for coming on the show.

Roelant:

Yeah, thanks for having me.