Podcast | What data means when convenience is new loyalty – and what it means to give data an ambition.

Jean
Shin

Director, Strategy & Content | Podcast Host of Mobile Interactions Now

22 min Podcast

In this episode, we check in with Bas van Leeuwen, Marketing Director at Gen 25. Combining his firsthand experiences in the data science academia and technology industry, Bas brings a unique perspective and valuable insights into how companies can create customer loyalty from the data they already have.

Podcast transcript

Jean:

Bas, welcome to the show. I am thrilled to have you on the show because I've been really wanting to talk about the role of data and how companies can use it better, now that we know a little more about data, to help them really reshape some of the customer journey they have, and fill the gaps they have in their customer experience. But first, just so that we get to know you better, can you just tell us a little bit about yourself?


Bas:

Yeah. I'd love to Gina. Thank you very much for the invitation and having me on the podcast show. My name is Bas van Leeuwen. I'm the marketing director at Gen25. Gen25 is a software and consultancy agency. We primarily focus on implementing Salesforce and AWS and more or less our mission is to improve customer interaction for companies. So companies have a better interaction with our customers and, well, enjoying a better customer experience, and also an employee experience.

My background is a combination of more or less human technology interaction and marketing and sales. So I've been in a combination of both. Before joining Gen25 earlier this year, I've been a business development director at JADS, which stands for the Jheronimus Academy of Data Science. Which is a data science university with a task to really create failure using data. So not so much only fundamental research, but also applied research, and how can we use data to challenge all different topics from deforestation to a better customer interaction, to better pricing models, to even seeing how visitors of museums act and spent perhaps some additional money within the city center. Before being business development director, I was in HR and IT, and I'm also in different starting agencies. Also, I think that the main topic within my career has been human technology interaction in any case.


Jean:

So from the very basics, what is the role of data in delivering the type of experience that today's consumers are really wanting and demanding even?


Bas:

Yeah, I think what's important in customer experience and even what consumers expect, the sentence "convenience is the new loyalty" is one I really like. If there is an abundance of choice, then people go for the easiest routes. You have to think about with online shopping, internet, easy access to a lot of different opportunities for consumers. There's an abundance of choice. And if there's a lot of options to choose from, and they're more or less comparable, then there are two main triggers in which people act on most products. So to say in their shopping experience. If the product is the same, let's look for the best offer, so to stay, but also, the most easy experience and the most easy way of obtaining the advantage that consumer wants. We're going to France this summer with our family, and you have to order a specific environmental sticker for your car.

You have to fill in all these types of data with their registration number of your car and your license plate. And I know that this data is available at some points by filling in my license plate of my car. It can automatically generate all the other data which is needed to complete this request for this environmental sticker. So instead of me buying the sticker for three euros on some governmental website, I went to a commercial website, paid 15 euros because then I only had to fill in four different numbers or some additional information. And it got me my sticker.


Jean:

You paid your convenience premium.


Bas:

Exactly.


Jean:

And that's three times higher. Wow.


Bas:

Five times higher.


Jean:

So 3 to 15, yes.


Bas:

Yes. I'm not being efficient in terms of money, but I'm being efficient in terms of time. So if it really saves me a lot of time, if you have this data, which you can use for customers to create a better customer experience, it's really valuable, but you also have to be aware that data in itself doesn't want to do anything. Data is just on some server being collected and it has no ambition, it doesn't want to do anything, it's just sitting there. So getting more data, it doesn't get you anywhere. You have to really get it working into models or CX experience and get the real value chain up and running.


Jean:

I love what you just said. And I think it's first time I heard it that way. Data doesn't have any ambition, but you're very much right. But the thing is, the way you are looking at it is..I got this sticker before, the data is sitting somewhere, you might as well just use it rather than having me repeat that and do it all over again. The lack of data is not the problem here, we are seeing every interaction is creating some kind of data, whether that's captured and used or not. But what are we actually learning? I mean, from it, having this data all around us, are we teaching ourselves to be more ambitious to come up with something?


Bas:

Yeah. You see, when I see in main directions, I think there are two directions  a company can take. The first one is companies saying we have to do more with data and we also have to do more with AI. So let's collect a lot of data because if we have the data we're rich and knowledge and we can improve our interactions. I think the other route is perhaps more interesting, collecting smaller amounts of data and doing experiments. So doing a lot of different experiments with perhaps smaller target groups, but learning from those experiments and scaling up the experience from there. And I think we have to let go the idea of for data you can get a hundred percent right. You're never going to get a hundred percent right and that's okay.


Jean:

I mean, come to think of it. That's how science works. It works on the probability, right? The experimentation that is going on, which I think is more interesting rather than just having data, I think those days of just talking about having a lot of data, data lake and all those things is entirely different category…we are interested in, how do I make Bas’s experience less painful when he’s getting that sticker in that situation. For me, what is more interesting is in that particular scenario. What you’re talking about in terms of getting the probability right so that we can confidently say, okay, if that happens then we do this…are we getting any better at that?


Bas:

Yeah, I think we're getting better because, especially big data. I think it was in 2013, you had to Viktor Mayer-Schoenberger who was a professor at, I think it's Harvard, but you have to perhaps correct me on that. But he wrote a book on big data, which really kicked off the whole data hype, so to say.

First we thought…a lot of companies and organizations, and also myself, we have to collect a lot of data, because if we have data, everything will be fine. And then we learned getting a lot of data is really costly. What is the problem you actually want to solve? And that’s the really interesting starting point. How are we going to use it? You probably don't want to use data. You want to improve your customer experience, or you want to be more efficient in some production chain or you want to improve the value chain. So think of that challenge at first and then think of the business challenge, which is the one to solve or the problem you want to solve. And then think back to towards which data is important in that case. And then you have this, perhaps smaller, amount of data, which is needed, which you can use to run experiments on.

I think the competition is really important because competition keeps you sharp and makes you better, right? If you're in a competition, you think about your own performance and you think about a way to increase it. If you're a government organization, providing the ability to buy a sticker, in this case, is a goal which is well achieved by launching the website and giving you the opportunity to do so. So there's a green approval and where everybody's happy.

Wisdom comes in different steps. You know, first of all, you want to provide people with this opportunity to apply for a sticker. Then you have this process up and running. It's in some cases also the experiments and you have to really learn from, okay, now we've got this up and running, the minimal viable product, perhaps you can relate to that, and think how to improve it. But if you're not in a competition, you're most likely to run your MVP and think it's our definite product. If you're having a website, which has a bounce rate of about 80%, which is just Google analytics, and you know your product is really popular, but your bounce rate on your website is 80%, something's going wrong with your user experience. You should improve your user experience because I don't think 80% bounce rate on a product is a healthy percentage, right? It needs to go down.


Jean:

That really jives with what I learned too, a couple of episodes ago, I was talking with somebody who's working on insurance claims automation. And this is a typical situation where people are stranded, straight after a car accident and calling the insurance company to report the accident. And right upfront in IVR, they give you an option to do self-service. And, and in this case over WhatsApp, and once the process moves to the channel…basically what used to be consultation with a specialist who has legalese kinda document in front of them going through every question in that legalese kind of way. And the customer has to answer on the other side. And sometimes they even send a specialist to the site to take pictures later, and a whole to do.

But the thing is, once you decided to do something about the process since the process is too painful for the customer, there’re different options to make it better. In this case, they ended up doing it on a chatbot-enabled self-service over WhatsApp. And once you change it to that interaction, it's no longer a specialist going to the legalese. They have to be very, very selective with the questions they're asking. I would love to see more of this, but in order for that to be sufficient enough to replace what used to be handled through a specialist interaction, much has to be supported by data. Are we seeing more of this?


Bas:

I think there's also a challenge there. I think it's really great that we can use data to improve customer experience, but also to make the questions better. So I'm not asking for the answers you already know, or the answers you don't really need in this case with this insurance company. Right? So getting it more into the core questions she wants to ask. I know about this investment company in the Netherlands, they invest into smaller businesses and it can be a hassle for companies who are growing rapidly to get new funding from traditional banks, because they have to a lot of procedures and talk to a lot of people to get the funding in and grow as a company. Right. And especially if you're a new company for, let's say three years old, and you don't have books, which go back a 10 year time.

So you can see there's a revenue increase over longer periods. Getting funding from a bank can be even more difficult in some cases. Okay. It doesn't mean it's not doable, but there's opportunity there to perhaps make it more easy. And there's this Dutch company who say, okay, if you want to apply for funding as a smaller company, you can apply with us using our website and just fill in some forms and we'll grant you an application or not. First of all, I thought to myself, okay, we're talking about, I don't know, even a million dollar plus investment on a company even more, how do you do it? So I talked to the CTO and he said, we're just looking at, is the company existent? How's the website built? How LinkedIn profiles built on, gather all this data, not so much on a human interaction, but technology driven models to gather and scrape all this data. And we build a specific profile and we say, okay, this company is trustworthy and the risk is this percentage. So we can grant the funding. The thing is that it's working so efficiently that if you fill in the forms, you can get your reaction back in like 15 seconds. But if people fill in the forms and they get declined in 15 seconds, we tend to think it doesn't work. It can work in 15 seconds.


Jean:

That is awesome.

I love the fact that we’re learning and applying it, but at the same time, part of me is a little bit like, okay, isn’t it borderline manipulation. Is that the future we would have to actually kinda accept…


Bas:

Yeah. I think you're true there. I think the way to go there, if you look at integrity and manipulation, that's always a challenge, right? Especially with using data, because using data, you have to accept that you're never a hundred percent, right. It gives you an estimate of what is most likely to be, but it doesn't really give an assurance that it's going to happen in this way or that way. I think if you're looking at scientific research, now there's a lot of research being done also on transparency in AI, right? So if there's an artificial intelligence model, can we understand why it's making specific decisions? Otherwise, we're just going into a funnel.

Maybe it's about the experience. When you're in a restaurant and you're ordering your food, right. And you have your main course. I don't know how it is in most countries, but in the Netherlands, you have to wait for, I don't know, 40 minutes, an hour to get your main course, which is fine by me, because I can chat away, I can drink some wine. I have some interesting conversations with my wife or with my friends, but they can make it faster. They can make my main course in, I don't know, five minutes even. But am I being manipulated or is the user experience for me the most valuable one? And I think in the last case, I think it is so, and it doesn't, I think it's manipulation if you, in some ways, change the outcome without knowing of the one receiving the outcome. And in this case, the company doesn't change the outcome. They just put in a small delay to get the outcome towards the specific customers.


Jean:

Let me ask this. I love how you can provide that kind of perspective…to think about where to draw the line, but it will be amiss if we don't discuss the privacy issues related to using data to make this experience fast and efficient, is there a general guideline as to what to do, what not to do?


Bas:

Of course, up to the standard of the legal boundaries. So to GDPR in Europe or even different regulations in the United States or in Asia, there are different regulations around the world. You have to really think about the legal boundaries, but I think it's important to say those are legal boundaries. You don't have to go all the way to the boundary to get your process up and running. Think, first of all, about what's the real outcome you want to, to go to, or what's the problem to solve. As I mentioned before, and then what data to collect and not per se collect more so don't get data greedy, because it gives you two risks. First of all, if you collect more data, then there's a risk of costs involved because collecting data, using data, storing data is more costly enterprise. It's more, it's more costly to collect more data than to collect less data. So think about the claim data to collect. The other one is that if you collect more data, you have to really think good about security. And I think that's also a smart way to go. A lot of companies are looking at a data lake at this point or solid systems to collect their data because there's a lot of different loose applications, which collect data in some form, which aren't per se, a hundred percent waterproof when collecting data. And they can really be a risk to your company because, for any company losing data is not something you want to have.


Jean:

In terms of geo locality, where the data center is situated, do you see any pattern in terms of, because me being in Germany right now we have a lot of companies like Lufthansa where they would not trust cloud storage of customer data, do you see any trend?


Bas:

Data is considered to be owned by individuals. So me as an individual, and you as an individual own our own data, and we have the ability to share a part of the data with companies. And if we think they shouldn't use it anymore, we can get the data back. And there's even a fine risk for companies if they don't use data in the right way. If you look at the United States, the usage of data is less regulated towards individuals. I think companies are in a stronger position using data. They have more opportunities to use data in comparison to Europe. So you can say in the United States being really rudimentary in the selection, data is owned by companies. In Europe data is owned by individuals.

If you look at China, for example, which is also a big data-driven economy in a lot of different ways. And the government is really an important player in that in part, and a lot of big tech companies are also well co-owned by the Chinese government. So, and they're using this data without too much restriction to build a profile of individuals and to create a customer experience. I think they think suits best. So you can say in some ways in China, data is government owned. So you have company owned, individual owned and government owned.

If you're looking at SaaS, cloud solutions compared to on-prem, traditionally on-prem was considered safer. But you have to be honest. If it's connected to the internet, they really are being challenged by a lot of different potential intruders, which don't take into account there's a fence. They just go into to the internet and try to access your specific environments. But if you're a part of a bigger environment, like the bigger corporations, like a Salesforce or AWS, which have a lot of security measures in place, you're bound to be safer than building your own secure environments, or you're really going to have to invest extremely heavily into security measures and find some sort of training methods to keep those people up to standard, also. Because I think a lot of companies are being attacked hundreds of times a day via the internet. So getting the security up and running is really important and it's best to defend yourself when you're in a group versus when you're alone.

Sign up for our newsletter

This article is part of the Connections Now newsletter. You can sign up here to receive it every month.