Glenn Vanderlinden: If you wanna start with personalization, if you wanna to start with data or upgrade, whatever you have, start with use cases, verify what you need in order to bring them to life, be specific about the requirements, take into account privacy consent, data minimization. But the user story is so important because otherwise you might end up with data that has no use case.
Kailey Raymond: Hello, and welcome to Good Data, Better Marketing. I'm your host, Kaylee Raymond, and today we're discussing why the boring bits of our jobs are actually the most important. When it comes to technology implementation, it's easy to get awestruck in the beginning with a shiny new toy. We often wanna jump in headfirst and play around. When it comes to CDPs, despite our best intentions, this approach can lead to poorly adopted use cases. Instead, we need to focus on the structures, processes, and definitions. These steps may not be the sexy part, but they are necessary in driving the right outcomes. Only by building the right infrastructure at the beginning can you adapt to the ever-changing use cases and evolving business requirements. On today's episode, I'm joined by Human37 co-founder, Glenn Vanderlinden. We discuss bridging departments to create a consistent customer experience, focusing on processes to drive results, and making the most of customer data with a flexible enterprise.
Producer: This podcast is brought to you by Twilio Segment. Segment helps every team access good data. Data that's real-time, clean, and accurate. The result? Relevant customer experiences that drive real revenue. From creating more engaging customer loyalty programs to optimizing your ad spend, Twilio Segment helps over 25,000 companies turn customer data into tailored experiences. Learn more about why everything is better with good data at segment.com slash good data.
Kailey Raymond: Today, I'm joined by Glenn Vanderlinden, co-founder and lead solution architect of Human37. Glenn has over a decade in analytics experience. Previously, he held roles at Semetis, including executive director of technical operations and services, and worked as a business analyst at Deloitte. Today, Glenn helps organizations drive business results by leveraging customer data. Glenn's an expert. Glenn, welcome to the show.
Glenn Vanderlinden: Hey, Kailey, thanks for having me and thank you for pronouncing my name in a very cool way actually. Vanderlinden, that was good.
Kailey Raymond: Well, I want you to pronounce it back to me now so I know how wrong and how off I was.
Glenn Vanderlinden: I like it a lot that we say Vanderlinden, which is the very Flemish way of saying things, right? Sounds cooler in English.
Kailey Raymond: I like it. Oh, not many things sound cooler in English, I will say that. Glenn, I know and I shared a little bit about who you are and what you do, but in your own words, how did you get to where you are today as the co-founder of Human37?
Glenn Vanderlinden: It's a very good question. I would say by accident almost, and by circumstances. So like you did the intro, I started my career at Deloitte, which is at least in my head it's where one of these companies where people go when they don't really know what they wanna do, which is the result of studying something like economics, which typically is also something that people study when they have no idea what they wanna do in life. So that's two things in a row. I learned a lot of my time in Deloitte, but I did understand it was not my type of culture environment type of work I wanted to be doing. So I started to do some reorientation and I ended up at the company called Semetis, which was founded by the first two Googlers in Belgium. And their objective was to bring performance media advertising to big brands or to the market, combine it with back in the days what they were calling web analytics or digital analytics. And that's basically how I got intrigued by everything that was data. Because back in the days, advertising was actually something that was still required a bit more analysis.
Glenn Vanderlinden: It wasn't like a performance max or all of these automated things yet. We actually needed to pull a bunch of Excels, figure out what was the best setup, and then recalibrate all of our bits. So that's where I kind of got hooked onto the optimization of things, and I naturally stumbled into analytics. I stayed there for eight years, and after eight years, we kind of figured like, okay, what are we gonna do next? Because we were fairly young to be in the positions that we were in a company that was sold to a big group. The market was changing. Advertising was not the most, let's say, interesting part of the digital sector anymore for us. And we noticed that a lot of companies were struggling to understand what the value of the data was that they already had or they started capturing. The whole movement of first-party data was actually starting at that moment. At that moment, also COVID hit, which meant that all of a sudden the world was your market because people didn't really care anymore if you were in Brussels or if you were in London. And so we kind of set out for that adventure to start our own company with some colleagues that I had back at Semetis back in the days.
Kailey Raymond: Very cool. So you've, well, first of all, the fact that, you gained your inspiration at first from Excel spreadsheets is beyond me but congratulations for that.
Glenn Vanderlinden: Yeah, the world was different back then [0:05:40.3] ____.
Kailey Raymond: Yes, definitely. Glad that we no longer have to pull audiences building Excel spreadsheets and uploading them into our marketing automation tools, hopefully. But exciting to hear that you've worked with a ton of different companies and kind of noticed this trend of things that they've been finding challenging or what direction they're moving in. You mentioned first party data being one of them. I wanted to get your take as somebody who is advising a lot of folks on their strategies in this industry about some of these big trends that you're watching that are impacting customer data, customer engagement.
Glenn Vanderlinden: I think there's a bunch of them. I don't wanna touch upon AI because I think at some point it's going to be unavoidable in this conversation. So let's save it for then. I think a lot of people get, or got at least scared. In the last, I would say, 24 months, because of the whole messaging about third-party cookies will disappear, third-party data, the introduction of concepts like second-party, first-party, zero-party, all those type of things. And usually when there's confusion, people go start or start going in all kinds of directions. So we've seen all kinds of, let's say, demands from businesses to help them out doing things that actually made very little sense in some cases. But what I do believe is that the whole requirement, let's say, of the industry to refocus on first-party data, let's call it like that, and zero-party data is actually a good thing. Because working for first-party data or working to obtain first-party data or zero-party data is way harder than it is for third-party data, right? Which means that you actually need to have some value exchange, which means that your product or your service needs to be better.
Glenn Vanderlinden: I believe that it kind of, was sometimes easy to capture a lot of data in the last decade or so because everything was allowed and all of a sudden everybody had loads of data and was shipping it left and right and hoarding it. But now you actually need to work in order to get data from your customers, in order to generate meaningful interactions, and in order to actually start building a relationship. So the result of that, I feel, is that the relationships with customers, or at least the hope I have, is that relationships with customers get valued more because of that again.
Kailey Raymond: That makes a lot of sense. The value exchange that you're talking about of, perhaps this consumer demand that they require something in exchange for this information that they're providing a company. I'm wondering if you have any thoughts on any of that, like the consumer behaviors that might be behind that are driving and catalyzing these industry trends towards first party data. Any consumer trends on your mind that you think are kind of pushing us towards these new data trends?
Glenn Vanderlinden: Consumer trends, it's hard for me to interpret it that way, right? So I'll try to rephrase a bit of the question, if you will allow me. I think it's more like, what are we seeing that is driving the way that businesses are moving forward? And one of that is like, a lot of organizations says that personalization is key, but I think that personalized servicing or personalized offerings are more important than anything. So what we see a lot is that a lot of the big organizations that have been doing business the way they have always been doing it are now being challenged by organizations that put service or product or whatever it is at the center of how their entire organization has been developed. They're also the type of organizations that have for the very first time somebody responsible for customer data or somebody responsible for at least bridging the gaps between the different silos, rather than customer data sitting either with IT or either with marketing or anything like that. And so those are the things that we see. We see a lot of the big organizations are struggling with that, but it will take a bit of time, let's say years or decades for them to suffer enough to kind of swing around the big boat that they are.
Glenn Vanderlinden: But that's kind of what we're seeing. So a lot of organizations are looking into how can I bring better services. I think that's the main thing in general. And they articulate that by saying leverage first-party data and personalization. I'm not sure if that's actually what they mean. It's just a scratching the surface on how that is represented, right? But that's mostly what we see that's happening. And as a result, it impacts the entire tech stack, right? For the first time, people need to talk between departments in a lot of cases in order to get a consistent, coherent customer journey or service or experience for that customer. Which also means that a lot of the stacks or technologies need to be bridged, need to be ripped and replaced or are being thought through all over again. And the interesting part is, I feel at least for the very first time, it's done from the entire customer journey perspective. What can we build? Or what should our infrastructure look like in order to facilitate an experience that's not necessarily linked to a department, but it is linked to how we wanna to serve the customer across customer departments or across the company departments.
Kailey Raymond: This is an excellent point that you're making. And it's something that we've been playing around with a little bit here at Segment. You've been talking about the stack and how teams are kind of oriented and how silos might naturally be created because of that. We've been playing around with this concept that we're calling the customer engagement stack. Which is slightly differentiated from perhaps other stacks because it does, in fact, center the customer in the very middle of it. And it makes sure that you're bringing together different departments and different buying committees and different stakeholders and making sure that you're putting that customer profile and the customer at the center of the problem rather than the problems of the marketing team. So it's like a slight nuance, but hopefully it kind of gets to that idea of breaking down those silos. And I think it speaks to this larger trend that you're talking about, which is really like this flexible enterprise, this need to adapt, this need to be interoperable, this need to be, should we say the word, Glenn, composable? I know that was a big trend that customer data platforms were talking about for a while. Do you have any take on the direction that we're heading with the flexible enterprise, interoperability, flexibility, adaptability, whatever name you want to put on it?
Glenn Vanderlinden: It's interesting that you call it a flexible enterprise the first time I hear it. Like last year was a lot, especially like the first half of last year was a lot about packaged versus composable, right? And that was in the CDP space. The more you make it tangible, the better or sometimes the better you can define it, the harder it is for people to understand, I feel sometimes. On the other hand, I wasn't necessarily happy with the first half of last year neither, because all of a sudden everybody was a CDP. And as a result, like it doesn't. It's not really clear what it is or what it isn't, right? So if you think of it, or if I think of it, the last, let's say, 18 months, kind of what happened is there was the debate around composable versus packaged. I don't think that's necessarily a debate anymore because it was very much versus rather than they can coexist or they're both two ways to get to the same goal.
Glenn Vanderlinden: I don't necessarily think one is better than the other. Everything depends on the context and the use cases and the teams and the people and everything you're trying to involve. It's not clear what we're trying to tell the customer or what problem we're trying to solve for that customer. So I guess that bundling everything together, or at least the concept coming closer together and rather than saying, well you did this part, you did this part. And like, we stay away from each other, but trying to fuel each other's part of the customer journey, or at least the engine that powers that customer journey. I think it's a good thing because it puts again, the customer back at the center. Yeah.
Kailey Raymond: I like that. I like that concept of, essentially this is kind of like the technological guardrails to make sure that you can actually serve your customer in the way that they deserve to be served. Is that right?
Glenn Vanderlinden: Yeah, yeah, maybe that's a good way of looking at it. Then again, it's like the whole privacy aspect, right? Especially with the whole thing that all organizations are opening up budget for AI. Whatever, LLMs and all these fancy things, but very few organizations are talking at least explicitly about privacy, about consent and about data quality. And I feel like this conversions, like if you could add in those layers in that converging stack or those converging components, like it's gonna double benefit the customer because everything's going to be built in or at least nicely tied together.
Kailey Raymond: I'm glad that you're bringing up this concept of privacy. I do think that GDPR completely changed the game as it relates to privacy. And now we're talking about cookie deprecation. Who knows when that's actually gonna happen? But you should probably think about moving towards first party if you aren't already. And I wonder if it's because a lot of the forward-looking kind of platforms that are being developed today are developing consent, privacy, you know, all of that within the ethos and within the product itself.
Glenn Vanderlinden: To some extent, yeah, I would agree. The interesting part for me is that you can't necessarily buy compliance through tech. And it sounds a bit weird because practically you could, right? You could say like, here, I buy a CMP, whatever, and everything will work. The thing is, it's just a piece of tech facilitating workflows, let's say, and facilitating flags on top of users and do's and don'ts for specific populations, right? A lot of organizations forget that it's also about the process constructed around these technologies, because every technology piece with regards to privacy can be customized, which is the whole purpose, right? There's different legislations. There's differences in what you can do in specific sectors and you can't, and differences how DPOs, even in the same sector, apply certain elements. And so it's more about this entire process. And one of the things that we came across very interestingly was, for instance, like a super simple example about which is about process and less about technology when it comes to compliance is like imagine that everything is a CDP, an analytics platform becoming a CDP or branding as a CDP because it can now shoot data to any destination, right?
Glenn Vanderlinden: Including one of your advertising platforms. Well if you think about the data being captured in an analytics platform, well you probably had analytics concerns, but do you actually have consent to send it to the marketing platform? Technologically, the platform would facilitate it, and technologically it would potentially allow you to put on some guardrails, but the whole process of when do we capture, how do we inform, how do we get all of that set up, is something that lives outside of the tools and can be baked in based on configuration or based on code or based on something else. But it's something that is often forgotten. And one of my expectations is that a lot of data will end up in the coming, let's say, 24 months in a lot of platforms where it was never supposed to be in the first place because it got piped through from another platform that got a different type of consent in order to get it there in the first place. So it's kind of the difference between like, I buy tech that allows me to do things. Some tech has built-in features, other have less, but what is the process? And it's for me, the process that needs to be talked about first and foremost, and then the tech will be configured based on that.
Kailey Raymond: That is an incredible take. I hadn't really considered a lot of that in making sure that you're right. Almost every single tool that you're buying today has some sort of integration to somewhere else that you can send data back and forth. But ultimately, making sure that the processes that you have internally are strong and robust and continue to put the customer at the center is the most important thing. The technology itself is just the route to get there.
Glenn Vanderlinden: Most of the data these days ends up in a data lake anyways, right? And then from the data lake, it could go anywhere. People model for a million things. They build audiences for a million things. But you don't want that data to end up in places where it should not, even though for a lot of customers, it would be invisible, right? You're still doing stuff that you should not be doing. And so being respectful of the customer and thinking ahead and working with your legal department in order to make visible what could happen and what scenarios are where things could be forgotten and process that out and build it in from the start, I think is good practice and best practice, especially with everybody now going crazy over AI and doing all the craziest thing, pushing their data left and right, and using it for personalization, wherever they can find an opportunity to personalize, means that you could surface some very weird behavior or your data points could come back at some point where you were not expecting them to come back, even though you said no for that specific purpose.
Kailey Raymond: So interesting. I mean, you've brought it up a couple of times now, so maybe we can dive into that AI conversation. And really what I think you're talking about is the simple concept of garbage in, garbage out, but it also is the idea of making sure that the data that you are using is consented first and being treated correctly with privacy implications in mind first before you are implementing it into kind of these models. Is that right?
Glenn Vanderlinden: Yeah, and the interesting part is I read this or I saw this white paper come by like about the new role of the chief AI officer. And all that stuff. And I read that X amount of budget for marketeers or CMOs was like liberated for AI and Gen AI and all that stuff. I'm honestly still waiting for the chief data quality officer or the chief customer data officer or something to show up first. Because otherwise, it's like you said, garbage in, garbage out. The only difference is it's gonna be at like an unprecedented scale. You're gonna be bad at the scale that you could never have imagined, because you're automating everything. So that's kind of my main worry. And it's almost, I don't wanna be a pessimist, right?
Glenn Vanderlinden: I don't wanna be negative. It's a major opportunity. I just think that like, we're kind of jumping a couple of steps and some of the things that are POC or proof of concept value are in widely in production. And the thing with proof of concepts is once they're in production, they're rarely revised to be upgraded for production scale, right? So that's kind of what you take with you in production. It's gonna become a legacy or like burden over time to kind of fix all of that stuff. I find it very interesting. I find it kind of worrying too, but there's a lot of work to be done there at least. The good thing is that people always want what is shiny, but in order to get there, they need to go back a couple of steps and polish all of that up. So I guess that everybody's excited about like the automation, the AI and stuff, but at some point people will realize that they need to take care of the boring parts between quotes, right? In order to make that actually work for their business.
Kailey Raymond: It usually boils down to the boring parts, honestly, is process and structure and knowing what your use cases are to be able to actually get value out of the data that you're collecting. Why are you collecting that data in the first place? It's interesting with AI, I really feel like we're like Darwin's finches and we just like haven't developed the beaks yet to be able to understand how to use this technology. Like humans, it's too fast for humans to adapt to the speed of technology coming out. But I like your concept of like, it's okay maybe to be slower in the Finch generations and be able to do it the right way.
Glenn Vanderlinden: Yep.
Kailey Raymond: Yeah, this is an interesting concept.
Glenn Vanderlinden: The thing is why I'm saying that it's, I feel like it's okay to be a bit slower, especially for the people in marketing, right? It's because some people are gonna hate me because I'm saying this, but it's only marketing, right? I mean, like you're personalizing some content, you're personalizing some images, some texts, some messages to some extent, right? And I agree, it works better. It works better in most cases, right? Personalization usually works. But it's not like we're using AI to its full potential already. And so maybe marketing is a playground we need to do or we need to play with. The thing is, it's always immediately going directly to your customers. So if you do it wrong, well, it ends up at a customer. And since it's AI and automated, you're probably doing it wrong at a scale.
Glenn Vanderlinden: So I would say like the difference that it will make, we're talking about percentages. So it's not like we're missing, I feel, a mega huge leap to like, okay, this should have been in production yesterday. I'm more careful with like, okay, why don't we just spend time on, like you said, what's the use case? What's the user story? Let's work on that to figure out which data points we need, make sure that these data points are polished, they're clean, and that we have a process for them to be clean in the future as well. And then let's roll things out. That could also be an incremental way of actually putting it in production rather than spending like, I don't know, two months only polishing data and building process on that. It could also be use case driven. And so you bundle the boring parts with the exciting, shiny output of it.
Producer: This podcast is brought to you by Twilio Segment. Great customer experiences require better data. With customer profiles that update in real time whenever new signals emerge, like when a customer adds new shoes to their cart, and a suite of best-in-class privacy features that ensures you're respecting your customer's data privacy every time, Segment allows you to make good data available to every team so that they can deliver personalized customer experiences at scale and at exactly the right moment. Cue that cart abandoner email. Julio's Segment helps over 25,000 companies power their most important objectives with data they can trust. Are you interested in learning how to build stronger customer relationships and greater profitability with a complete understanding of every individual customer? Learn why everything is better with good data at segment.com/gooddata.
Kailey Raymond: I'm wondering if you have any advice for marketers, data analysts, like folks that are kind of really touching a lot of these use cases every day of how they might be able to avoid over automation as it relates to AI. Do you have any like gut check or like checklists that you kind of abide to of this is going to work and this isn't? And here's the thing that we need to make sure that we're checking before we do this.
Glenn Vanderlinden: The thing for me is like, to what extent should you automate everything? Because you wanna own the relationship with your customer, right? I always find it intriguing that people wanna outsource and automate everything, including all of the touch points they have with their customers. I think that automation should influence part of how you address it or part of the story you tell. But you should still own the way that you compose that interaction somehow, because it's your business and people like to do business with people and not with machines. And that's kind of what throws me off a lot. When everything is automated, I feel like I'm talking to a machine and I don't wanna be talking to a machine for specific things, right? When I have a question about my UPS package being delivered, yes, I don't care that it's a machine because I want it to be fast.
Glenn Vanderlinden: But if I want some personalized service from a shop, I might not want that to be a machine. I actually want that to be informed by a machine because you wanna have an idea of what my profile is, that I like technical gear more than I like any other gear. But based on that, I should have like genuine interaction with kind of the soul of the company as well, which is hard to capture in a fully automated flow, I feel.
Kailey Raymond: Let's dig in a little bit more there of like, you know, what good data might actually mean, how humans might define good data, how good data is being used by those bots that we're talking about. I'm wondering, Glenn, if you have a definition of good data.
Glenn Vanderlinden: I asked this question a couple of weeks back to... Like we had this debate at the office about good data or something along those lines and it was actually very intriguing because it's a very simple question and it's very hard to answer. And one of my colleagues said, well, it depends on the context. And I'm like, that's such a consulting answer. But it's actually kind of true. Because you don't know if data is good or bad. A data point is not good or bad if you see it's on its own. You need to look at it in its totality among other data points. So I would say good data is more consistent data. In the sense that if you have an issue with your data and it's persistent, at least you know it's persistent, right? Because if some of your formatting of your data is not the way you want it to be, but it's not for all of your other data points, let's assume a data format or something, then it's still consistent and we can still work on that in the totality of the data set.
Glenn Vanderlinden: So I would say consistent data points, good data consists of consistent data points. Another attribute I would give is fit for use or fit for the use case. There's a lot of companies, I believe, that have data, but they don't have a use case. So the question is, is that good data? Well, it's hard to tell if it's good data, right? Because it's just data. Nobody knows what to do with it anyway. So you can't tell if it's good or bad or if it's actually useful or not. So the data that you have should be fit for a use case. At that point, it's useful data. Because it also means that you're only capturing what you need, which is like a principle of data minimization, which I like a lot. Which also is linked to privacy. So taking into account capturing only what you need to capture for the use case in a consistent way.
Glenn Vanderlinden: Those are kind of like the three keywords that we've touched on so far, which would for me be an attribute of good data as well as like consented data. It kind of ties back to fit for use, right? If it's a marketing use case, you should have the opt-ins for marketing. If you have an analytics use case, well, the data should be consented for that. But I think I would say if I would summarize it, it's like consented, minimized, fit for use, and consistent. Yeah.
Kailey Raymond: I really like this because what you're really teasing out here, this use case part of it, has to do with activation, how this is actually being put into use, how it's showing back into your customer base through their customer journey, through whatever personalization tactics that you're creating for them, making sure that you're keeping privacy in mind. And so I think that a lot of what we're talking about today is really the boring stuff, which is like... Yeah, sorry, but you have to do a lot of the upfront work to define your use cases, your policies, your processes, before you actually start unleashing a lot of this out into the world. And you can do it incrementally, of course, like start with one, then build and grow in terms of your use cases.
Kailey Raymond: But it's something that takes a lot of time and thought and consideration. And perhaps is one of the reasons why it is a slower boat to turn, especially with really large enterprises, is gaining that consensus and breaking down those silos and making sure you're all on board with those use cases could take a little bit longer.
Glenn Vanderlinden: Yeah, it's true. And I don't wanna sound super conservative, right? But I think like to some extent, we kind of, I'm trying to balance out like the hype that exists about all the stuff you could be doing, especially with AI, because that's driving that whole frenzy, right? The culture. Where people have heaps and heaps of data. Like I got an email from a brand that my data got leaked a couple of weeks back because they had a... They had a leak, they had a data breach. And the thing is, I like this idea of banks got robbed because they had the money. Companies get robbed because they have the data, right? And so minimizing your data and thinking about the value of your customer and tying it to a use case and should we have this data point? What's the use case?
Glenn Vanderlinden: It sounds very boring and very conservative, but I think it's going to be one of these things that inevitably we'll have to deal with. And I think it's gonna become a necessity anyways. I'm just trying to sometimes in a conversation, try to pull it forward for saying, Hey guys, people or organizations that, and I didn't invent this quote, it's from a book, but organizations that hoard data generate their own risk. And so being mindful of that, because you're actually dealing with very sensitive data in some cases, right? If it's not minimized properly or depending on the type of business you are for your customers. And you also wanna make sure that you protect that, right? It's not because you have a consent that you should be storing everything everywhere and shipping it to like a million platforms that each represent an additional risk, right?
Glenn Vanderlinden: And then while there's the additional point of like not storing data for the sake of storing data, which is like the carbon footprint of it. I read, I think it's the Economical World Forum or something that released a paper saying that storage of data or data centers in general, I think it was, have a major footprint, which is, I was surprised when I saw it, expressed in percentage of contribution. But half of that footprint is actually generated by data that is single use. So it was stored once, used once, and then never looked at again. So it also contributes to that aspect of like, global warming, all those type of aspects. But on the other hand, it also just, if you think about capitalism, what we talked about earlier, it's about reducing costs. Like why should I have 10 years worth of useless data about my customers while it's generating risk and racking up costs? I could just get rid of it. It would make everybody's life easier, including the legal department being on your back. So like thinking about the use cases and what are we storing and for which purposes is one of the things that we need to think about more going forward.
Kailey Raymond: I've never heard the companies get robbed because of their data. Like that is so, yeah, duh, like, of course, but it's so, it's such a good way to highlight the value that data has today. And I think it's a really astute point to make as it relates to data minimization as well. So thank you for bringing that to my attention because I probably will be stealing that, Glenn. This has been, I would say not gloomy, but like there are some cautions, there are some risks involved with what we're talking about. But I wanna get to the other side of this. I'm sure you have examples of the way that some of your clients or some brands that you really respect and appreciate are leveraging some of this good data, this consistent data, this consent of data that we've been talking about to build great customer experiences and programs. Do you have any examples of things that get you excited that you've seen in the world about good data use?
Glenn Vanderlinden: Yeah, I think one of our favorite use cases from one of our clients is a organization that is in sports events. And what they basically do, their customers have digital wallets, which basically are the applications, right? Users who are being tracked in there are consented, so we take care of that. But one of the use cases we have is that during one of their events, we could have food waste, or the risk of food waste exists, because there's multiple places where you can eat, there's food preparation that's being done during the event, and at some point during the event, the prediction might happen, or the case might happen that we predict that there's an overage of food that we need to get rid of and or that won't be consumed. And as a result, everything goes into the trash can, right?
Glenn Vanderlinden: What we're experimenting with is actually pushing messages to people's digital wallets or applications in order to tell them, hey, you're at this venue. Here's a coupon to get actually some food there at a 30%, 40%, whatever reduction. And at the same time, what we're doing, we're doing two things. One, we're making sure that, well, one, we don't waste the food. So it also has an impact on the environment. Like it's battling food waste, let's say. And on the other hand, it's making sure that we can kind of experiment with couponing. Can we activate people to try new things? Can't we activate people to try new things? And all these types of things. So it's one of the interesting use cases that we have. There's many more on that as well, because a lot of the use cases we talk about in marketing is usually digital.
Glenn Vanderlinden: But the advantage with customer data is that we can actually track a million things, right? It doesn't only have to be a website and application. People who opt in on beacons can be tracked. Passes for events can be tracked if we have opt in all these type of things, right? And so we can tie it together into a better experience for those end customers while taking into account how do we optimize flow, et cetera, et cetera. One of my personal use cases is actually by Zalando. I think they're German, the big retail in Europe or e-commerce in Europe, selling clothing, sneakers, and all that stuff. I tend to be the guy who likes a pair of sneakers or jeans or something. And like, whenever they're done, I will rebuy them. And so what Zalanda does very smartly is they just say, hey, you bought this product six months ago in this size and you kept it. And so it's almost like a one-click add to basket for me. Which has a lot of value, right?
Glenn Vanderlinden: It's very simple, very stupid, because I know that in Vans, my shoe size is 42 or something. In Nikes, it's 41 and a half. And so I don't have to order two pairs. I just know like, okay, I like this pair. This is the recommendation. You know my database on this pair. I'll buy this pair with this size and then I could just get it shipped at home. So those are like the more, I guess, product or marketing user experience related use cases while the first one was a bit more broader on like what can you do with data in general.
Kailey Raymond: I like both of these. The second one I'll address first, which is... It's like, a food delivery app. You probably order the same thing or similar things from the restaurants that you get delivery or pickup from often. And so making that easier and reducing the friction for you to be able to do that in clothing is brilliant because let's be honest, like I do the same thing. A lot of people do the same thing. They purchase the same clothes or maybe in a different color often and over time. And so that's like a really brilliant way to be able to build loyalty within your customer base in a really simple, easy way. Which is also tying back to your other one.
Glenn Vanderlinden: If you think of that, if that second use case, it's not only about customer friction, right? Because if you think of it, imagine how many packages that organization sends. Because typically what people would do, they were like two pairs and two pairs of jeans and like two pairs of shirts because they're not sure about the size. And so it means that if you're in two sizes, 50% is gonna get shipped back anyways. Which means that some other customer is not gonna have that item in stock, which means that transport is being paid for going to the customer and potentially going back from the customer. So even from a cost perspective, if you go back to capitalism and efficiency, it makes a lot of sense to implement that. It's not only about increasing conversion rates and reducing time to checkout. It's also about what is my logistical cost of inventory, of logistics, in order to basically get the right item delivered to that customer.
Kailey Raymond: Which is also tying back to your first one, which is really interesting. It's both a loyalty play as well as an efficiency and a logistics play. You could be building loyalty within your customer base by cross-selling or up-selling them food at a discount, which they might not even realize you're doing because of food waste reduction. It's also this really interesting thing where, I think to your point, a lot of these experiences are digital, but you're bridging the gap between digital and physical by bringing it into the real world of events and showing something that is just truly a cross-channel experience. You're sitting in a stadium, you're having food from a vendor that's physically there, but you're getting this message from an app and it's all, it's killing two birds with one stone. It's hopefully reducing food waste, you know, helping the environment, but also, yeah, putting more money back into the pocket of that company.
Glenn Vanderlinden: Yeah, and there's like, this is, I feel like this is already one of the more, like the cooler use cases, right? But if you're in sports event, there's very simple use cases that we're putting in place, which is like users who have a subscription to all of the games, but don't show up for X amount of sequential games, which means that we're left with empty seats in the stadium, which is not what a club wants, right? Because they want you, if you're very honest, to buy a shirt or do something else and like spend some money while you're there. And so identifying those users and showing them or onboarding them onto the resale market for tickets and showing them like, hey, if you go through this flow, you can actually open up your ticket for somebody else to be there. You make some money and we have somebody in the stadium, right?
Glenn Vanderlinden: And again, it's about efficiency, but it's also about like, hey, maybe the team didn't play well that week, you don't know, or the last month, and people don't go to the stadium anymore, but somebody else wants to take their kid to the stadium or something. At least that person gets the money back, the other person gets a nice night out with the son or daughter. And in the meantime, the club is happy because the seats are filled. And so it's all about like, how do we allocate based on, or what's the best next move almost, if you will. And that's where AI a lot of times comes in, right?
Kailey Raymond: That's so true. And I just saw for the very first time, this exact use case that you're talking about of resell within the same app ecosystem, which normally goes to another platform, a reseller, and they're getting your data, they're doing business with you. And so keeping it all in one platform obviously has a lot of good effects for that company. And I never considered what they were doing when they were showing me that, and it makes total sense now that you're sharing back to me what this strategy is. It's so smart. And it's something that I think is gonna be extremely disruptive to the resale market. And I'm curious to see what happens. My last question for you, Glenn, before I let you go today. And I think maybe I know what this answer is gonna be, because we've been talking about a lot of the work that goes in up front to a lot of these sexier, more interesting use cases. But if you had any steps or recommendations that you might give somebody that's looking to up-level their customer data strategies, what would they be?
Glenn Vanderlinden: And I know I'm sounding very boring. It's not the purpose. I'm very excited about all this, right? But I would say always, always user story, use case or user story. And a use case can be multiple user stories, right? But try to be very precise in what you're trying to achieve because it will force you to go through the entire process anyways. And I'm a big believer in processes, because typically processes outlive people in an organization, which means that if the process is designed well, like the rest will at least go to a certain standard or work according to a certain standard. My idea would be always like, if you wanna start with personalization, if you wanna start with data or upgrade, whatever you have is like start with use cases, verify what you need in order to bring them to life, be specific about the requirements, take into account privacy consent, data minimization.
Glenn Vanderlinden: But the user story is so important because otherwise you might end up with data that has no use case, right? And it's what you called correctly activation. I feel like data is only valuable once it is activated. And activation can be many things. It can be used for analysis. It can be in your weekly dashboard or in the report or analytics platform. But be specific about it. You should not have data that could be useful at some point in time. Because having could in your sentence is not a good strategy when data is involved. At least that's my feel. So I would say start with user stories.
Kailey Raymond: That's beautiful. I think that it really does all come down to the process that you're implementing that keeps the customer at the center. Glenn, this has been great. I think it's been honestly a needed reality check. So thank you for dropping your knowledge today.
Glenn Vanderlinden: You're welcome. Thank you for having me.