Data literacy, visualization and the challenges of data integration

About this episode:

Chris Van Doorn, founder of C-Data, discusses the importance of data visualization, data literacy, and the challenges of data integration. He shares insights from a recent project and emphasizes the need for a good understanding of business requirements in data projects.

Key Takeaways

• Data literacy is essential for both data users and consumers.
• Understanding business requirements is crucial for successful data projects.
• Data integration involves understanding the business processes and facilitating them with technology.
• The role of a business analyst is becoming increasingly important in data projects.

AI-generated transcript

Matthew Stibbe (00:01.455)
Hello, welcome to Behind the Data with CloverDX. I'm your host, Matthew Stibbe. And today I'm with Chris Van Doorn from C-Data in rainy Tilburg in the Netherlands. And he is the founder and owner of the business. Welcome, Chris. Thank you so much for being on the show.

Chris van Doorn (00:17.232)
Thanks for having me.

Matthew Stibbe (00:19.855)
So I'd like to start by learning a little bit about C-Data. Tell me, what does the business do?

Chris van Doorn (00:25.841)
what we do is help to understand our customers how they can use data in a better way to optimize their processes and to use data to visualize it so they can try to make a better company of what they're doing.

Matthew Stibbe (00:44.719)
And what sort of clients do you work with?

Chris van Doorn (00:48.272)
That's very diverse. It's in health, it's in automotive, it's in marketing, it's in retail. Data is everywhere. So problems with data and challenges are also everywhere.

Matthew Stibbe (01:02.159)
I've heard that we are almost every business now as a data business on some level. Is that your experience as well? No.

Chris van Doorn (01:09.168)
No, no, they, they, perhaps they should be, but, and, a lot of them want to be, but before you can consider yourself to be a data company, it just requires a little bit more than Excel.

Matthew Stibbe (01:26.639)
Yes, yes. Well, our mutual friend Pavel Najvar at Clover says, if you are using an Excel spreadsheet, you failed, which I always think is a good way of thinking. Anytime you're moving data with Excel, you've failed. 

Chris van Doorn (01:41.264)
I'd like to make a short, a small addition to that. If you're only using Excel, you failed because you will still use Excel in any situation. Even I do.

Matthew Stibbe (01:55.823)
That's a good qualification. Thank you. And I'm sorry, Pavel, for putting words in your mouth. So you've been running C-Data for a long time. Tell me a little bit about your personal journey and how you got to that and kind of what is your world of data?

Chris van Doorn (02:13.264)
Actually, I just stumbled into the world of data. I studied aeronautics and space technology in Delft and failed to finish the study because of personal reasons. And then I stumbled into IT, like probably everyone in the late 90s, to start solving the Millennium Bug and other challenges we had.

from there, I learned, structuring data basis for customers, programming, in a small company. And, I turned in, to a bigger company, beginning of the, of 2000. And from there, I specialized in DB2 and in data structures, helping programming and, departments to structure the data better, starting with the first visualizations in business objects and then on in Cognos. Later, Tableau came along in 2012. And in 2012, I also ran into CloverDX for the first time. And both companies had an opportunity to partner with my own company, which I had by that time.

I saw the opportunity to provide end customers with a total solution for structuring and preparing the data to be able to visualize it in a simple way.

Matthew Stibbe (03:52.783)
Well, that's an interesting thing because visualization is often an end product of a well -organized data-driven company. But to be able to visualize in Tableau, what do you have to do to your data? What are the things that people are struggling with with data visualization?

Chris van Doorn (04:01.072)
It is.

Chris van Doorn (04:11.824)
Take the time that you need to visualize your data, multiply it by five, and that's the effort you need to put into preparing your data. And it's not just connecting to and cleaning and structuring. You know, we're trying to get information from all the underlying systems that we use in our operational processes.

And that data just isn't structured almost always isn't structured in a proper way to visualize on. So you have to operational data model on one side and you have the analytical data model on the other side. And to get from the operational data model to the analytical data model that needs a lot of work. And it first thing that you need to do is.

Determine for yourself. What do I need to be able to report on because that will be the basic? Input for your analytical model and once you have that you can try to see do I have that data on what? Schedule do I record that data? Are there gaps in it? What's the data quality and all those things come into play? Yeah.

Matthew Stibbe (05:38.223)
It's an interesting idea that there are two different data models conceptually, an analytical and an operational. How data literate do you find people? How can you educate yourself to understand things like an operational and an analytical data model?

Chris van Doorn (05:56.4)
Let me start with the first question. How data literate do you find people to be?

That's a very broad question. Let me focus that into two different answers. The first one is overall data literacy is way below the level it should be. And that's not even for people who are using data, but also people are consuming data. And if you look into modern media, the amount of graphs, data, figures, numbers that you see in BBC News or Dutch media or even on Facebook, Twitter or other media streams you might follow. It's overwhelming and people. I really believe that people do not have a broad enough knowledge in data literacy to be able to really place those figures and graphs in a correct way. So it is almost like getting your driver's license, you should have a sort of data literacy license in schools or when you start working somewhere. That's the first part of the answer. The second part is data literacy among people who use data is a bit better than what I just mentioned, but it's still below what it should be. And the people that do have the data literacy are the data scientists, the people that are working like a data architect. And those usually have problems to relate to how businesses see their processes and the data that you can use in those processes.

Matthew Stibbe (08:01.295)
That's, I think a very profound insight. And I'm not one of the world's data literate people, but with clients, they want lots of reporting and analytics. But when you ask them, well, what do you want reported and how do you want that? They struggle even to express that in a clear way. So how do we go about educating ourselves to, well, not yourself, you're an expert at this, but how does a data illiterate person like me educate themselves about data?

Chris van Doorn (08:31.536)
try to not call myself an expert because some other people see me as an expert, that's flattering, but I try to keep educating myself because expert level requires a constant to even reach it, which I don't think I'm still at, but nonetheless, to reach it requires a constant keeping up with trends.

and developments and you have to keep educating yourself. Where do you start? Well, I have a training for you if you know that, just kidding. The links in the bio. No, well, joking aside, there should be some sort of training and you can go online and look up data literacy trainings.

Chris van Doorn (09:26.192)
The visualization tool I use, Tableau, has a basic data literacy training. It does require some basic knowledge, which might make it a bit too advanced to start with, but there is a lot of information online you can find. Even if you just understand what does data actually mean, why do we use it? How can I interpret a simple bar graph or a line graph?

When can I see that it's manipulated, like missing axis or just showing the top of a graph or those kinds of stuff. That would help.

Matthew Stibbe (10:07.983)
That's an interesting thing for me and others to go and explore. Let's move on to another area. I'm very interested in digging into a particular project of your choice and really understanding the story of it and not only the technology that you use, but also the engagement you had with the client and the processes you went through and what you learned from it. So perhaps if we could start by

thinking about a particular project, tell me a little bit about what you were trying to achieve for your client with this.

Chris van Doorn (10:44.752)
One of the most recent projects that I'm still active on is trying to get all the data pipelines and data exchanges into a system that is on site for the customer. So they actually have a lot of external parties that arrange their processes.

which means they do not have control over it, which means that if something goes wrong, they don't know until a couple of weeks, perhaps even later, we've actually seen that. And it's small, not really important business processes like invoicing and ordering of products or parts. So...

you know, the crucial parts that they don't have any insight in. And that's where we are, where we came in to get a grip on analyzing the processes that are available and to see which parties are included and show the business why they should be in control. And,

The technology we use of course is an integration platform, data integration platform, you are connecting to so many different products and different types of documents being Excel sheets or CSV files, but also APIs and databases to get the information from. And just getting the overview of all the...

parties involved and all the processes involved together with the business people that took a couple of months alone and then we still had to do all the work. So yeah.

Matthew Stibbe (12:50.639)
So let's start with that. So can I just understand a little bit more about the objective? Was it the case that you wanted to, the systems might exist in different places in the cloud or whatever, but you wanted also to have a local copy of the data or you wanted to bring the systems in -house as well?

Chris van Doorn (13:10.736)
Some of those systems were already in-house, but they weren't able to build their own integrations between those systems. So they turned to external parties who built those integrations. Then you lose two parts. First of all, you're not in control of the architecture and the functionality that they are delivering. Plus they charge a lot of money. And they had the integration platform in -house.

but they didn't know how to use it correctly. so when we came in, we said, you know what? I know we're an external party party too, but, you own the product. You should have people inside your organization that, care for that product, for the platform. And that is an ambassador for that product. They need to walk around.

And here's something I want to exchange information between those systems. And they need to go like, we have a data integration platform. We can do that for you. When do you want to have it ready? And once you have those ambassadors in house and in initially that's us. The external party that comes in to build the data integrations and get the streams flowing and optimize the processes, make sure that the business sees they're in control.

Provide them with a dashboard that they can see that's green or process ran correctly today. There's no harm for you when it comes to financial or operational flows. But after that, it has to be internal people. It really has to be. Otherwise you still won't own the platform and it will not give you all the benefits that you thought it would. And for us, that's always our main objective: to empower the customer to be able to build it themselves.

Matthew Stibbe (15:14.607)
It seems to me, as an outsider, that there are different levels of capability. There's sovereignty or control over your data, and then there's control over the integrations and the pipelines that join them together. And at the highest level, there's control over the ability to build and maintain those pipelines. So with this project that you mentioned, what...

Chris van Doorn (15:25.456)
Mm -hmm.

Matthew Stibbe (15:41.711)
Tell me a little bit about the engagement you had. How did you, during those several months, how did you understand where all that data was and understand, you know, build that map of the landscape?

Chris van Doorn (15:54.096)
Talk to people. That's all. So they have an IT department that is at their central headquarters, but they have so many locations in the Netherlands, in the Benelux, in Germany.

Chris van Doorn (16:19.088)
Not everybody knows what the IT department can do. So you don't need to focus on the IT. You need to focus on the business. So talking to business executives, hearing the opportunity, we have a problem. Come talk to me, come talk to us. We'll try to understand you. In the end, it's about the data for us.

But not for the business user. For the business user, it's about his process. He needs to be in charge of his own business processes. And it's either selling or customer satisfaction or invoicing or logistics or that's his main focus. And when you can help them solve a problem in their area, they will know when where to find you.

And that's exactly what happened because we had a backlog of about 60 or 70 days when I said, guys, I can't run this on my own anymore. So I need to fly in someone else. That was okay. They estimated about six weeks and after six weeks, the backlog was a hundred days. And a couple of months later, we're still at a hundred days backlog. And, because the work keeps adding up, it's not like we're not delivering. On the contrary, we're still delivering, but the additional work keeps growing, growing, growing. They see the advantages.

Matthew Stibbe (17:58.895)
It's fascinating. As a space enthusiast, I think you'll like this. I remember Verna von Braun said, going to the moon is easy, but the paperwork is really difficult. I think when I talk to people who do what you do, the technology bit is, it's not easy, but relatively it's easier than the business.

Chris van Doorn (18:12.688)
Your bit, the communication bit, the understanding, the education, you know, training people, all of that, all of the work around the work is hard. And it takes, so in the end, how many days do you think you're going to end up delivering on this project?

Chris van Doorn (18:38.512)
It is.

Chris van Doorn (18:46.576)
depends on how many days they will still add. At this moment it feels like a never ending story. And of course it will be a story with an end, hopefully a good end, but it's looking like that at this moment. I reckon we at least need until the end of the year to clean out the most...

Matthew Stibbe (18:48.079)
Yeah.

Chris van Doorn (19:15.44)
challenging and most pressing items. And who knows, because the company is growing and growing, they will run into new challenges and needs before the end of the year that will be prioritized over the needs that we are solving right now.

Matthew Stibbe (19:32.239)
Well, appetite comes with eating, I think. So the more they see you can do, the more they will want to do. So tell me, with this project, what's the main lessons that you've learned that you will apply in future projects?

Chris van Doorn (19:35.408)
Yes.

Chris van Doorn (19:51.76)
I think, and I already knew it actually, but this really proved it for me. You need to have a good understanding and good relationship with the business to make this work. You can have so, good understanding of, of the technology and you can be such an expert in what you're doing.

If you don't know what will help your end customer, the business, you will have no clue. You really need to understand that business. So it's not, I'm more and more considering myself not as a data analyst or a data specialist or expert if you want to. The part of business analysts.

is becoming more and more important. Being the middleman between the business people who have a requirement in their processes and the technology in the IT side that will empower the end users to run those processes. We're just facilitating with systems.

That's it. And the more we can understand what the business needs and facilitate them in a better way, the happier they are. And of course, the better a company will have the opportunity to grow.

Matthew Stibbe (21:32.255)
I think that the ambassadorial role of talking about technology to business stakeholders and talking to technologists about business outcomes, I think that's a really powerful place to be. It's very impressive. So when you're thinking of taking that experience you had with that project and all the other ones, what advice would you give to a business that was on a

a data journey and embarking and starting on a data integration project. What's the most valuable insight that you could offer at the beginning of a project?

Chris van Doorn (22:13.071)
your requirements on paper. What do you want to get out of it? And when you have a goal, talk to the people who will be building that. And hopefully, and that's where the business analyst role comes in, they will try to find out the question behind the question. Because once they do that in a correct way, you will get more out of that.

journey than you bargained for in the beginning.

Matthew Stibbe (22:47.727)
That's a powerful insight. Thank you. Well, Chris, we're almost out of time. So I just want to say thank you very much for joining me today. It's been a real pleasure meeting you and learning about C-Data. Any final suggestions, recommendations, tips, or advice?

Chris van Doorn (23:00.4)
Likewise, likewise.

Chris van Doorn (23:11.792)
I think I've put a lot of the tips and advice already in the conversation. Try to be authentic.

And if you are able to describe a solution and you know you can build it because you can, it's so much more powerful than doing a sales training.

Matthew Stibbe (23:34.351)
Amazing. Thank you, Chris. Well, that brings this episode to a close. If you're listening to this and you would like any more practical data insights or you want to learn more about CloverDX, please visit cloverdx.com. Thank you very much for listening and goodbye.