Listen on SPOTIFY | APPLE | GOOGLE
Guests
- Erin Kelly, SVP of Enterprise Solutions & Strategy at Kraft Analytics Group (KAGR)
- Francine Klein (Sr. Solutions Architect, Vendia)
In this episode, Erin Kelly and Francine Klein talk with host Tim Zonca about the differences between analytical and operational data, the challenges and opportunities inherent in sharing each (and both) at scale, plus the game-changing promise of connecting these separate worlds inside enterprises and across business networks.
Show notes
- Erin Kelly on LinkedIn
- Francine Klein on LinkedIn
- Kraft Analytics Group
- Slalom
- Vendia
- Joey Chestnut, competitive eater
- Microsoft Dynamics 365
- Salesforce
Transcript
Tim Zonca 0:07
Greetings from the team at Vendia, and welcome to Circles of Trust, the podcast for leaders across all industries committed to speeding up innovation at scale, making a profound, positive impact on business and the world. I’m your host, Tim Zonca. And we’re about to dive into a conversation with longtime experts in the data space: Erin Kelly, Senior Vice President of Enterprise Solutions and Strategy at Kraft Analytics Group, and Francine Klein, Senior Solutions Architect at Vendia. In this episode, we’ll talk about analytical and operational data, and Erin and Francine share some insights and dig into these two types of data and take a look at are they on a collision course or not? And what does this mean for companies that rely on both types of data? Alright, so let’s jump into it. Erin, Francine, welcome to the show. It’s great to have you.
Francine Klein 0:53
Thanks for having us. Yeah.
Erin Kelly 0:54
Thanks for having us. So excited.
Tim Zonca 0:55
So it’s been a while since the two of you work together at Slalom. Tell us a little bit about each of your backgrounds and what you’re focused on now. Let’s start with you, Erin.
Erin Kelly 1:04
Yeah, absolutely. So in my role at Kraft Analytics Group, affectionately named KAGR (a little bit of a play on the compounded annual growth there), I lead all of our client-facing delivery for both a strategic consulting arm, as well as a data and analytics platform perspective. So, kind of, live in the business x tech x data intersection on a day-to-day basis, and we service at KAGR, we’re across all major professional sports leagues and industry players. So folks like NFL, NASCAR, Amazon, and others we’ve worked with in terms of how to utilize data to drive great fan experiences.
Tim Zonca 1:46
Great, thanks! Francine, what about you?
Francine Klein 1:49
So prior to joining Vendia, my career was in consulting, and I was actually on Erin’s team in the data analytics space, building out different data strategies, and really understanding where companies were with their data vision. Then, I went and had done it myself at a company, and by doing so, really knew the dichotomy between the operational and analytics data. And so, by stumbling upon Vendia, and seeing that there’s really a data fluidity problem and a data sharing problem, as well as this separation of these worlds, I jumped over here to make it real for our customers. And so my role is to understand how customers are doing data sharing today, and really using operational and analytics data, and building better solutions built on Vendia with that.
Tim Zonca 2:32
Thanks! Yeah, one of the things that I assume is really interesting in your roles is that you’re working with customers all the time. And so I assume that spans just a really broad spectrum of customer maturity around their dating sharing practices where, you know, some of them are really mature and doing impressive things and others, you know, they have a lot to do to catch up — the opportunity is large for them. I’d love to hear from you, what are some of the most common themes that you see are trends across those areas of maturity, depending on where they’re on on their journey?
Erin Kelly 3:11
Yeah, absolutely. I’ll grab this. And so the sports and entertainment industry is pretty interesting, I think. In many aspects from a data and analytics perspective, it could be anywhere from five or ten years behind some other industries like financial services, insurance, what have you there. But it’s pretty exciting. From both my experience prior to joining KAGR with Francine at Slalom and other pieces like that, to really help the organizations kind of climb up that maturity curve and ladder. And so we focus a lot on, you know, many sports and entertainment organizations are what we’ve called “data laggards.” So really, that bottom of that maturity curve. And how do we actually get them over that initial chasm where organizations may have had an initial data platform, they have their operational systems in place, but they’re not really connected, and they’re certainly not driving business value in those pieces there. And so as, as folks move up that maturity curve there to more data pioneers, one of the big things for us that we’re talking about all the time is that strategic application of data: “So great, you’ve got the technology, you’ve got the process, you got the integration in place, but are you thinking about the right ways, and the use cases in the right time to be able to activate appropriately and really drive the value?” Because, at the end of the day, if the business value is not being achieved, then the investment in that tech data stack really isn’t gonna get you where you need to be.
Tim Zonca 4:29
Yeah, that makes sense. What about you, Francine, any trends that are particularly impactful that you see across that spectrum?
Francine Klein 4:34
Yeah. And Erin made some really good points around the fact that people have this operational data for really targeted activities versus then using that [data] for analytics and insights. And so how do you actually embed those insights into some of those workflows? And what’s always really interesting is that the thing that’s in those, it’s all the same core entities, right? We’re always talking about a fan or a customer or a product. And so the reason why these things become arduous is there’s so many stops and starts. And the capability of actually creating those datasets and the operational workflows, right, create the new customer or create the product purchase versus the insights generated all that. But it’s the same customer and those two things. And so it’s the stops and starts that makes sure that those two customer lists are actually the same. And that’s some of the crux of the issue that we see, and where those two worlds really should be colliding.
Erin Kelly 5:27
Yeah. 100% I think one of my more favorite buzzwords out there right now is really this “data mesh/data fabric.” It’s like the buzzwords of the years past around the lake, the swamp, the lake house, everything water-related. Now, it’s above the surface and more connected in terms of both fabric and mesh and what does that really mean to be bringing analytical and operational data together to go do an action, deliver a great experience, drive a competitive advantage for an organization? And I think folks are starting to aspire to some of those more seamless moments, which I think, you know, I think we’re going to unpack those today here in terms of what that could look like at the edge of analytical and operational. But again, yeah, it’s balancing out between those laggards and then trying to get them up [from the data laggard] to the pioneer, so that we can even talk about the “fashionista” terms of mesh and fabric and everything else like that.
Tim Zonca 6:29
Well, you know, to help us get there, I’d love to have the two you dig into… you both use the term operational and analytical data. And so can you talk a little bit about you and your customers? What are those terms? What are the differences in those data types? And if you can provide some examples of each, of what you see in each of your customer bases, I think that’d be really useful for our listeners to get a good appreciation of those different terms — especially what they mean to you and your customers.
Francine Klein 7:01
I can give it a go, and Erin, I’d love you to help massage it.
Erin Kelly 7:04
100%.
Francine Klein 7:04
So operational data: Think of it as the events, right? The actual thing that happens in the workflow. You create the customer, right, you made the deal, you made the purchase. So, for us, right, it’s shopping online. We go, and we click things. And so that information is operational and, from a data perspective, from a technical perspective, you actually need all the information about me and all the information about a product, right? I pull the product, the price, you know, comments, etc. And so from a data perspective, you can think of it as the full row for a single record. And translating that from an analytics perspective, you don’t need all that information. You actually want to look at all… you want to sum up the calculations of price, you want to sum up the activity history for single columns — but across lots of rows of data. And so that’s really the crux of it, at a core level, [it] is wide or long. And then there’s layers on top of that.
Erin Kelly 8:03
You’re spot on there, Francine. And I think one of my favorite statements around analytical data is [that] it’s used to make business decisions. It’s very crisp and clear there, where, at the operational level, it’s transactional data. To your point, [Francine,] it’s recording — at this point in time — what is happening in that specific business process or workflow or other pieces like that. And that’s kind of the beauty of getting them closer together, in terms of how do you use the information around the trends and the aggregate that is used to make business decisions and deliver that to the edge where operational transactions and other pieces are both triggered out to a consumer, an operator or whatever, and then also collected?
Tim Zonca 8:42
I’d love to hear some examples of each, but before we get there, is it fair — based on what you just said — to kind of assume that the decisions being made, let’s say, on the analytical side, you use some examples of like, there are people like looking at stuff saying, “Hey, we’re gonna make business decisions here.” Whereas are the decisions being made on the operational data side, because of transactional, more made by computers? Like, show dynamic pricing based on these mini-decisions on the data while someone’s, you know, looking at the merchandise or what have you? Is that a fair assessment?
Francine Klein 9:25
That a very good question. You actually picked up on something where the analytics and operational worlds have started to merge a little bit — dynamic pricing. And so yes, right, it’s the workflow operations, the immediate workflow transactions [like] a deal booking an invoice payment, right? A credit card swipe. But dynamic pricing actually is the start of the operations and analytics world, and it is an example of the tip of the iceberg of what really can happen. And so for that to happen, right, you have to actually look at some trends. You have to do analytics off of trend history. And so for that to have occurred, someone, some systems have to have taken tons of tons of operational data, put that into a warehouse, looked at some trends, and then, therefore, from those insights, embedded that into that operational workflow. And so one of the things that’s just started is that spot capability. And so analytics has embedded real time and operational workflow. But those insights are probably/likely delayed, right? It still is the massive amounts of stops and starts to get that information, generate those trends to then incorporate that. And that’s I think, where Erin and I are hitting at is, those stops and starts shouldn’t happen, right? That just can start going away with some more of these systems coming in place. And the reconciliation that needs to happen to make sure that you can incorporate those insights in the right place, to the right customer, is where the need for the data fluidity is. [It’s] where you need that aligned customer, you need to align product. Otherwise, you have all this transformation and logic that’s really complex.
Erin Kelly 11:01
But you’re but you’re spot on, Tim, in terms of technology being the true enabler there. If you’re feeding it with the right information and the right triggers and the right business rules and controls, then it really does become, then, a pretty great differentiator from an automated, efficient standpoint — in terms of those those true workflows that are hitting the operators, the customers, the products, the things that matter and are driving the business on that day-to-day basis. You know one of the supplemental… just some examples of analytics use cases: So we, at KAGR, focus a lot on the fan experience. And to us, we think about the fan in the overall fan funnel — whether that’s acquisition, engagement, retention, or whatnot. From that standpoint. And so, you know, some of the interesting use cases that we’re tackling on a day-to-day basis with many of our clients, you know, from an acquisition standpoint, it’s really focused a lot on performance marketing and our targeted campaigns. So is a team or league talking to the right fan with the right product, being a ticket, retail merch, engagement through the right channel? And how do we actually think about that on a day-to-day basis? And, you know, we have clients that have immediate tickets to sell for a game on Sunday; we also have clients that are sold out and are really focused on driving broader fan engagement. And so being able to use a combination of analytical data to understand the target and pull out those specific customers — and then activate actions through any sort of digital channel there (your more traditional of email, CRM, sales, sales, call center type pieces, but also, you know, lots more digital channels that are coming online, specifically in the sports and entertainment space) — I think one of my more favorite examples that we’re in the talks of right now is really around what happens in-venue when fans come in? How can, by the scan of your ticket when you’re entering the venue, can you get identified and get personalized beverage offers, or fan sweepstakes, or other pieces like that can go out to the right person at the right time, and deliver an awesome fan experience? You think about, you know, how can you use data and analytical data and operations to drive memories? That’s what we like to do on a day-to-day basis. And that’s what we like to work with our clients on.
Francine Klein 13:21
And so, with that, the type of information that you guys have to pull right, it’s not just operational in nature. It’s also the speed at which you have to embed the accurate information based off of if they’ve already done something, the analytics can’t be delayed, and also has to be incorporated, probably quite real time. Right? You don’t want to get them the recommended beverage when the game is over.
Erin Kelly 13:45
Yeah. 100%. Exactly. We’re constantly building up behaviors, trends, understanding, who do you look like? Does Francine Klein prefer this hot dog or this hamburger or whatever, even if you haven’t purchased before? And, you know, I think it’s just it’s really exciting in terms of the power of the data and the information and the partnership between analytical and operational to do that in a swift fashion. And, you know, that’s where the super cool terms like data mesh, fabric, and all these other things that venues and operators are thinking about come into play. How can you actually create a more integrated ecosystem or business network, if you will, to connect all those pieces?
Francine Klein 14:31
And actually, to play on that example, but not to belabor this too much, but imagine you don’t have information from the ops perspective and the analytics, but you keep recommending the same hot dog,? Like “I’m full. Stop recommending that. I’m over it.” Or we recommended it two weeks after the game is over.”
Erin Kelly 14:48
(chuckles) Right. You’re not Joey Chestnut.
Francine Klein 14:48
Right. You gotta update that stuff. Or like you know, the customer, right? I’m Francine in one system and I’m spelled the wrong way in another [system]. How do you know that’s that same “Francine” and that you know something else about me? Because I bought a shirt, and maybe I’m not gonna eat a hot dog because it’s gonna get messy.
Erin Kelly 15:07
Yeah. 100%. You know, one other thing that’s really interesting, and I’m sure it’s applicable to all businesses — but certainly in the sports space and certainly coming out of the pandemic — is that the face of the sports fan has changed a lot. And so behaviors and trends and historical patterns that existed prior to the pandemic have really gone out the window. The new normal or whatever we’re calling that these days has been established. And, you know, there’s an interesting element that you have the operational data on: What is the customer doing or the lookalike customer doing? But what we’re also pulling in is a lot of different macro trends around consumer behavior, consumer spending more broadly, lots of economic-social pieces there. And how can you continue to learn more, inform more, get to a better set of actions at that operational layer, through not only the data that you can collect on a day-to-day basis, but the information that you can enrich just with the world around us. And, you know, really try to push the innovation forward, which is what we love to do.
Tim Zonca 16:05
I want to touch on, actually, some of the data sources in just a moment. So thanks for bringing that up. Before we do, Francine, I want to get your take on some examples. But Erin, I love that example. Just because at least personally, I’ve kind of seen some of this at play over the summer. I’m a Chicagoan. And so, like you, [I] grew up just loving baseball. I live out here in Portland, Oregon now, which doesn’t have a baseball team, but we’ve fallen, my kids have fallen in love with the Seattle Mariners. Because I think they’re an example of a group doing such kind of compelling work, and just kind of continuing to bring us back and give a great experience. So, really, that example really resonates with me. Francine, what about the kinds of customers…? I mean, you work with customers to do all sorts of stuff from auto manufacturers to pet healthcare providers, I mean, so the examples are extremely varied. Any favorite data or use cases that, to you, really highlight some of the trends that you see happening in the market now?
Francine Klein 17:14
(joking) You’re asking me to say a favorite of some of our customers, Tim?!
Tim Zonca 17:16
I think, you can leave out names or something, but you know, you see the kinds of things that they’re doing. And it seems to span a really broad breadth of kinds of customers. I think that’s what seems fascinating.
Francine Klein 17:35
Yep, one of the ones I’m going to pull on is actually what I think might not be the most moving, but I think it’s going to actually really simplify some of the things in which people [as consumers] might be pigeon-holed in the wrong way. And it’s someone who’s trying to build basically a IDX graph based off of all different information. And so, think of it as, like a real-time, accurate, “no fly list” based off of information from anything you might do — from credit card processing, to your credit history, to other places like your mortgage payments, and so forth. And so, right now, right, your credit score is quite delayed, right? it’s not only delayed, and it’s really hard to argue with. And so, say you want to, you know, oh, my goodness, like dispute something that’s on your credit history report. You have to call this firm, you know, you have to call three different firms because they each have a report about you. And who knows what kind of phone tree you’re gonna go through, and then you call each one of them. And if one of them has not updated it, like, they’ll get the information from the other one and basically overwrite. And so you’re stuck! You’re basically stuck with this credit history that you might not agree with, and say, like, “Hold on, this is not right. This is inaccurate.” Who knows what the reason is? And so, [this customer is] trying to create this, kind of like this “no fly list.” But it’s not a no fly list. It’s everything, right? It’s your credit history, it’s your payment history, it’s all of these things in one. And you have insight into it, and can, [in] real time, provide input to say like, “This is accurate, this is not. I want to argue this.” And it goes immediately to each of the different parties that are using this information to generate their own scoring mechanism. And I think that will really kind of open up the opportunity for folks to not feel basically pigeon-holed and labeled in a way that they feel like they don’t have a way to fight against.
Tim Zonca 19:24
Yeah, interesting. In both of your examples, you touch on just this idea of the kind of data sprawl. And I think it seems to be such a given where, like, there’s no future where, you know, the data is not more places across more stuff and looks, you know, harder, harder to access harder to share. So, I’d love to have the two of you touch on that. What sorts of data sources are common? Paint a picture of that because it seems like you know, anytime you hear an example it’s like, even way more varied than someone like me would have originally guessed. It seems like therein lies some of the snafu and complexity. So I’d love to hear about some of these sources. And what makes it hard to share that?
Erin Kelly 20:16
I’ll grab the ball first. So, yeah, from what we see on a day-to-day basis with a lot of our clients, the top three sort of sources (and again, we focus a lot on understanding the fan and the fan behaviors and the fan engagement)… the big sources for us coming in are the transactional layers, which are most common. Ticketing is obviously up there in terms of how people are actually transacting and purchasing. Obviously, there’s retail up there as well. The other pillar is that kind of more CRM, that traditional information around interactions and behaviors and calls and services and all that good stuff. And then kind of that email marketing or whatever that engagement and touch point is, and that’s where we start in terms of building up kind of that single view of a fan throughout a variety of touchpoints. And then it gets really fun. And then it gets fun in terms of what else are they doing in terms of an experience, but online? How are they interacting with the loyalty program, the other pieces like that? I think, you know, as sports organizations are innovating pretty rapidly in terms of increasing the number of touch points they can have with said fan, and eyeballs, and all those other things that are happening there. And, you know, I think then we’re talking a lot about, even in that cookieless future, which is probably, you know, coming sooner than later, you know, how can sports organizations and industry players just create that reach? So that could be anything from digital. And how do you transition? Or how do you convert the unknown to the known and be able to figure out more patterns and behaviors from that standpoint, you know? And then we focus a lot on your more traditional demographic appends and additional information that we can pull in from a variety of places. But you know, at the end of the day, we’re living in the sports and entertainment space. And so it’s really all around that ecosystem on how an organization is interacting with the fan, whether that’s initial acquisition, or all the way through purchase, and then retention and loyalty from that perspective.
Francine Klein 22:23
Yeah, Erin, what you had pointed out in that example is the thing that I keep thinking of is, if you’re talking about this fan, right, and you obviously want to augment their experience with different information, operation, analytics, and data sources that you’ve talked about, whether that information is still about the fan — it’s touchpoints with them with the marketing system, or touch points with them and calls, or touch points with them and things that they’ve purchased. Right, so there’s like a financial system and marketing system support, you know, support help system. There’s still that fan that some, you know, they’re each disparate systems. And so they don’t know that, you know, in System One, I might be you know, Francine at Gmail and in System Two, I’m, you know, Francine K. And one of these things, from a data perspective, that I just keep seeing — no matter what — is like getting back to basics of how do you actually know that this Francine is that Francine to actually create that unified vision?
Erin Kelly 23:17
100%. It’s actually like, as you were talking about that, it always seems to come back to the basics. So as much as the technology is continuing to evolve, if the guiding principles or the basics aren’t there, then it’s not going to be realized. And so it really does come back to an increased sophistication on, it’s one person — how do you connect all those informations? And when you do that, right, that’s when you can light up all the great, super-great experiences. But, you know, it’s that balance of complexity, the place where technology continues to move us is so exciting. But it’s only as much as how much stitching and controls and other pieces that you can put in place to be able to capture that. You know, we talk a lot about just kind of change management and data. I think that a lot of organizations have kind of nailed the nail on the head. And again, I’ve been in consulting for a long time prior to joining KAGR. So I saw a lot of change management, I think, but when you think about change management from like an application or a process or even an org structure standpoint, I think organizations typically do that pretty well. Or they understand what it’s going to be like to move from a Dynamics to a Salesforce or from you know, an org structure one way to an org structure the other way. It’s not easy, but there’s some blueprints there. I think change management around data and how you can move up the curve with data is really complex. There’s a lot of different muscles that folks, you know, almost don’t realize, I think. You know, we talk a lot about when you’re putting all that stitching, all that data together to be able to then actualize it, you’re putting a huge spotlight on, potentially, a lot of upstream or other issues within the organization that need to be addressed to in order to true really use your data as an asset. Which I think is what we’re all talking about here. It’s that data is the valuable asset that’s going to drive the business or the process or the other pieces like that. And, you know, it is a different element around change management. And so, you know, Francine and I know (we’ve been in multiple discussions back in Slalom days around change management and all the importance there). But yeah, there’s probably a few coffees or wines to be had there, in terms of unpacking the solution to change management for data.
Francine Klein 25:26
Well, it’s funny. Oh, go ahead.
Tim Zonca 25:28
Go ahead, Francine.
Francine Klein 25:29
It’s funny, you were saying about the back to basics and so forth. You kind of hit something that I was thinking; it is [that] we can talk about the data side with operational analytics data coming together, and to your point, one of the things we always say [is], “Well, can’t you just do this? Can’t you just give me information about the customer?” …We can’t “just” because there’s, like, there’s complexities. And so the first layer is the data of ops and analytics coming together. But exactly what you said is, the next layer is, you know, even if you think data mesh —, whether it’s a tech system that’s distributed or teams that are distributed — and that the whole concept of mesh is the autonomy, you can’t give too much of that autonomy because then you’re just recreating that issue of the of the same data accuracy across each of those autonomous systems and or teams. And then their potential divergence of op processes that give you data that you can rely on.
Erin Kelly 26:21
Yeah, it’s really fascinating. And it seems like, you know, in that example of the mesh, the further you get into the mesh, the challenge of something happens is, is almost even — [it] could be potentially worse. That you’re sharing more out there than that moment of or the difference between doing it great and doing it not so great is even smaller.
Francine Klein 26:45
You still need something that keeps the mesh within the web.
Erin Kelly 26:48
Right. Yeah, like, “I like this. We’re going to pull this one.”
Tim Zonca 26:52
You know, both of you touched on, especially you Erin, and you said something along the lines of, you know, when you access data that you know, is within your organization, it seems like, part of the set of challenges is, you know, you’re accessing data, but it looks like, increasingly, that that data isn’t in your organization. And so it’s not, you know, access includes this notion of permission, sharing, you know, cleanliness, I can’t go back in like, to your point, for instance, and be like, “Hey, I’m going to tell my partner over here to go clean up their stuff.” So it seems like that’s a whole other smattering of complications. I’d love to hear what are the common problems you run into with customers as they want to, you know, access data, but it’s actually not. It’s coming from some other set of partners. You know, any kind of gotchas and then practices that you see the best people doing as they walk through that set of problems?
Erin Kelly 27:54
Yeah. 100%. I mean, it’s very common to not be a technical challenge that you’re facing. That you can build the patterns, you can understand those pieces, but it very much is around, like, I feel like there’s a few things that we face a lot in terms of how are you defining and understanding the data? How are you going to be using that? What’s the best way to bring that in? Because I think (and Francine and I can attest from years past) you can land data very, very quickly. What are you going to do with it? And how do you put enough discussion about that up front without boiling the ocean (or going back to whatever water example you want to have there)? But again, you know, understanding those pieces there, and then, I do think that there’s an element here of, like, looking around the corner. What should you be thinking about a few years from now? How do you think about those pieces like data privacy and preference centers? And things like that are what we’re talking about a lot in the sport space right now, in terms of where did a fan enter into someone’s web of understanding? Is that at the vendor level? Is that within something that the organization ran itself? And then how can all those little privacy policies actually stitch together to a whole, kind of, governance and [the] controls that need to happen there? But I guess, you know, Tim, going back to like, what’s the secret sauce, or the first step, or the best practices there? You know, I do think that there’s a moment up front, with the right kind of business and tech leaders in place, that [they] need to sit down and really talk about what’s the overall intent? What’s important now to understand the baseline understanding of the data and get the right kind of initial privacy process governance pieces in place and, at least, set a pattern. We talk a lot about patterns. You’re never going to solve everything. Data is naturally always moving and it’s very imperfect. But what are the patterns that you’re going to start to build with said vendor or other networks like that? So there’s a common understanding about the overall use case, the overall intent, and then data that flows in terms of the foundation of the discussions that you can have.
Francine Klein 30:06
It’s really interesting because, 100%, they need that common definition. And so with that, right, everyone has a common definition of what is it that we want to share? And how do we share it? And what rules do we allow? And so while people want to come to the table, if there’s some trust, right, they want to work together and do so. But there’s not complete trust in that network.
Erin Kelly 30:27
Right.
Francine Klein 30:27
They’re still each to their own. Autonomous. And so the big issue in the space is how do you make sure that there’s this data network among these departments in which they still have full autonomy on their data, but they’re all speaking the same language?
Erin Kelly 30:41
Yeah, it’s an awesome challenge. I feel like there are moments each week, and you all may have this too, as you meet with customers that you have those ahas where, “Oh, this language is not the same. Or we’re not actually talking the same language. Or you’re defining something that’s this.” And, you know, how do you level the education that is out there around data, in general, both within organizations and universities? Like, you know, at the beginning, we talked a little bit about ourselves, but, you know, I was a computer science geek in high school, computer engineering major, I came up through those pieces. But, you know, that was a long time ago. So there’s certain elements here: How do we broadly continue to raise the overall awareness? And, you know, it’s just really interesting, and I love some of the advancements that technologies like blockchain, and other pieces have forced into the world. It’s become a common vernacular around how to be securely, you know, sharing information, the contracts that you need to put in place. But how do we actually take some of those as interesting buzzwords and actually layer them into this intersection of operational-analytical. Especially when you’re talking about outside of the organization. Like you mentioned, it’s not just about sharing internally, it’s about sharing externally.
Tim Zonca 31:59
I want to dig into something you just said, Erin, on like, educating. You know, the customers that we work with, and especially around kind of coming back to the these two terms and ideas of the difference between operational and analytical data, you know, I’ve been in the enterprise IT software space long enough that, you know, I remember the days of like, OLAP, and OLTP, which, you know, were really like, bifurcated. Different markets and different sets of tooling to help people with their, you know, OLTP, like, the transactional. Or as we’re kind of talking out here, you two use the words like operational data, and then the OLAP, like the analytical. And, you know, one of the things we’ve kind of touched on is these two worlds seem to be colliding. And so from an education perspective, the difference is in analytical data versus operational? How much do those differences matter to the people that, you know, you’re working with and educate? And how do you walk them through to the degree that there’s a difference? Like, how much does it matter? And what do they need to know about those differences?
Francine Klein 33:13
That’s a really good question. …Let’s almost take our data hats off for a second. And if we’re talking to anyone who wants to…
Erin Kelly 33:20
(chuckles)
Francine Klein 33:20
…Right? (chuckles) How dare I? …If we’re talking to anyone who’s not in data (like my entire family), and we talk about insights that you might want, and you say the word “data” or the thought of “OLTP” — I may as well be singing the alphabet. Even saying “operational.” Like, what do you mean by that? And so even to have to say, “Well, you think these things…” — they just think, “Hey, I want to do things. And I want insights.” It doesn’t even matter that to deliver those insights, there might need to be a massive analytics infrastructure and data moving stops and starts between them. And so, even if we as non-data people think about things, we think about them holistically. But the systems because of, again, where we all came from and how you store data, they were very different. And the access patterns were different. And the development capabilities of teams are different. There’s so many things that are the reasons why they’re separated. But if those technologies can start to collide, then the conversation and the, Erin, to use other buzzwords and blockchain, the consensus among business networks can collide, right? And then the team skill sets developing on them. And so then basically, the systems and the data, the data stores, the people building on them, the business network collaboration can catch up with, you know, almost the common [or lay]person’s dialect around “Hey, I just think of data as all one thing.”
Erin Kelly 34:40
Yeah, it’s really interesting. And I think you’ve made some really good points there, Francine. And what I had in my head right out of the gate, when Tim asked that question of like, does it really matter at the base level between analytical and operational from a true education standpoint? Because, yes, there are nuances within there. But when we think about where the education is falling down, it’s around the operations, the controls, the ways of working the process. The things that are around data in general, whether it’s an aggregated number of 5% for a conversion rate, or a number of one for a, whatever that might be from that data bit that’s picked up from the CRM or something like that. I think at the end of the day, it’s really around how do you understand the data for what are you going to use it for? How? What is it? What should this be? What is it describing, whether it’s an analytical number, an operational number? And then what’s important, from the organization, to both keep master (what’s driving the underlying business model from a data standpoint) and then what are the key security infrastructure process governance controls that need to be put in place? And I think we’re just lacking so much education around those pieces that it’s almost like the analytical versus operational doesn’t matter right now. Because there’s so much around the process and the people-side that needs to elevate, and then it gets real. I think if we could establish some of those just understandings at that level, just out of respect for data as an asset and how that’s looking, you know, then I think you can get into some of the nuances too.
Francine Klein 36:22
So yeah, that’s a really good point because do you remember when, back in the Slalom days, when we’d sell and people would say, you know, an example is your conversion, or I want to look at my analytics of my pipeline velocity or something. And I would joke and say, “Absolutely, let’s select all from this really cool table that doesn’t exist.” Like, right? The data does not exist. The process doesn’t exist. You don’t even have a CRM, so to your point, we’ve got to understand these analytics and these insights, but the education of “There’s data. And process. And just commonality of what is this information that you want to collect? And the process for the accuracy of that information?” [is] so that you can then answer these great questions, and then make your process better.
Erin Kelly 37:03
Yes, 100%. Whether that’s a calculated metric or a collection of data at the edge or a trigger of information that is put there, but what’s the overall kind of process in the workflow? I couldn’t agree more. Yeah.
Francine Klein 37:14
And we expand that, right? If that process is across different networks, and vendors, or ticket sellers, or, you know, different data sources for credit history, that it’s hard enough IN walls. Imagine ACROSS the walls of just even that process and language. If you have a shared definition of what it is that you want to share, and just something that actually makes the data flow, and people have their autonomy. I mean, it just opens up. The next set of questions that you have to answer right is obviously the business. The thing that makes it hard, right? The business relationships and so forth. But we can remove the tech side of it. Then we can really focus on the value-add of the business strategy.
Erin Kelly 37:56
Yeah. 100%. Remove the noise. Yep.
Francine Klein 38:00
Yeah.
Tim Zonca 38:02
One of the perceptions I have that I’d love to have either of like debunk — or if you say, “Hey, no, I think there’s something there” — is that there’s been a lot of fascinating things happening in the analytical data space over the course of the last few years. You have vendors like Snowflake and Databricks that are, at least in our tech circles, household names. And they’re, you know, it’s a process in analytical data that many teams have really started to build. And so, I think the thing I’d love to have you do, you can either either debunk or say, “Hey, there may be something here” is, it seems like, there now will be the wave that will kind of come alongside, which is now there’s going to be this explosion of operational data. And so you have all these analytical people in a great spot to be like, “Hey, we can now extend our practice to include these other sorts of things.” So the first thing I’d love to hear is, is there anything to that? Do you buy that? It’s kind of high level, you know, [share] your impression? And then if, if so, what can those folks do to prepare or to start to take advantage of what should be a wonderful opportunity to do really cool things over the course of the next five, seven, ten years?
Erin Kelly 39:25
I love this question. And I’m 100% with you in terms of the level of technology and innovation on the analytical side is so high. …The funnel or the roadblock here, or wherever/whatever term you want to put in, is really like how much can that actually get to the operational side in terms of both collection as well as activation? And I think, you know, I think it’s an interesting area in terms of where businesses are investing. So, if they’re investing only in the strategy and analytics, the data tech space but not in the sales marketing operations space, it feels like a huge missed opportunity. Because the investment needs to be made in both areas of, “Hey, we’re elevating our ability to see, to collect, identify, analyze, and activate data in a pretty rapid fashion. We need to have the business process business applications, the sophistication within the functions itself to be able to actually take advantage of that.” And I think, you know, it’s almost like the rising tide raises all boats. The boats all have to go together. You can’t just have that from an innovation standpoint now. It puts more pressure and I think, in a fun and a good way, that keeps Francine and I on our toes in terms of how do you help the strategy analytics — that data tech, analytical side of the business — actually articulate the benefits of the investment at the functions on the operational side? So how do you kind of paint that future state vision in terms of where the value of analytics can drive the overall business? And not just a business decision, but really a business action and an activation? And then, what are the true investments that are needed on the operational side?
Francine Klein 41:15
Yeah, Erin, you said it perfectly. It’s now a visual of the rising tides, that the data flowing and rising, right, that the business process that generates the data, the analytics that [process] uses, and trying to find trends in the data. And so, to me, hitting up the analytics space has grown and its ability to churn through, right, it’s made it really, really easy to go through the mass amounts of data that exist in ways that it couldn’t before. Before, it was just slow. And you have to do a lot of different transformations. And you have to store it in different ways and different formats that data might come in, were actually a way that slowed things down. And so analytics has made it to say like, “Okay, well, data is gonna be in different forms. Let’s figure this out. We can kind of handle/you can handle buckets, and parse through data and buckets; we can handle data that’s in tables and rows and whatnot, and go through it faster. But one of the issues that they had is the operational data was either slowing them down, to Erin’s point, right, because it wasn’t accurate, or the it wasn’t really coordinated. And so, sure you had this really powerful system that can turn through but the sales and marketing and finance data wasn’t actually all the same information, or the data that you need from another company that you are trying to partner with isn’t in your house or isn’t even aligned. And you guys aren’t speaking the same language. And so how do you get that in? And so the analytics world is trying to make a bridge to the operational world by creating connectors, hey, it’s gonna be easy for you to get the data in great like so then now, that’s where the data lakes came, like, look, we’ve made it really, really easy for you to send us all your operational data, but it missed the mark. And what Erin is hitting is you really actually need the strategy that says, we’re going to speak the same language. And we’re going to actually share information across the ops and the analytics and our partners, so that the analytics actually has timely, accurate, connected data to then have insights. And then back in the process.
Erin Kelly 43:10
Yeah, 100%. 100%. And one thing that you mentioned, Francine, is not only within the organization itself, but outside the organization. And so a lot of pieces that we’re thinking through with with our clients is like data collaboration. How do you how do you actually formulate the right relationships in terms of, you know, data movement, data fluidity between between organizations? And what does that actually look like? And very often, it comes down to really articulating business value. So what is the value of my data as an organization to you, a vendor? And what is the value/what value can you, vendor, provide to us? And then how can we actually expand the pie together? And so, you know, it comes back to that, like, overall business strategy and other pieces like that. And, you know, we talk a lot around, give-get, but again, this is it’s, it all starts to elevate into, okay, now, the next piece is that those contracts need to be written in a different way to make sure that you have the right privacy and controls, but also the right measures to allow data to flow in the right ways back and forth from organizations.
Francine Klein 44:18
And so that’s a really good point. Because if they’re sharing data, right, you don’t want to share it one way you want it to be mutually beneficial. So if they’re giving data, they also need to give the, you know, data information back. And so how do you make sure that that back and forth and bidirectional nature to it is there with, like, there’s some trust? Right, you’ve done contracts in place, but there’s, you know, they don’t want to share it with everyone. So you need agreement of who sees what, alignment there. And agreement of how you’re going to use that; if they’re going to use it. They’ve created a network and an agreement of how you’re going to use this data for an aligned mission. You have to make sure that they do so.
Erin Kelly 45:00
Yeah, that’s interesting, because what I was just thinking through, Francine, is whose job within an organization is that? Right? You know, I think very often it happens right now in isolated silos based off of the business use case or the need. But, you know, is that the next wave… I’m sure there’s some really cool, innovative Chief Data Officers and other pieces like that, that are that are stitching it all together, but it’s like, it’s a marathon, not a journey, not a sprint, I guess, in terms of getting those pieces. And, you know, are the right people actually coming together to make an informed decision for today, but also for tomorrow? And then think right on the corner?
Francine Klein 45:39
Because people are sending data to and from. But if the businesses aren’t strategically coming together… it’s almost like we need a Data UN [United Nations] across each different group to say, “How do we collaborate together on this mission by sending data to and from, but on the insights that we’re going to generate and the experiences that we’re going to do? And how do we hold ourselves accountable, right, so we don’t do bad or that there’s not one bad actor who’s using the data, you know, for harm or against the mission that they’re doing together?
Tim Zonca 46:09
I hear in both of what you , you touched on, like, how the operational and analytical data on the tech side are coming together. Erin, I love how you highlight, “Hey, whose job is this?” So, it’s evident that it’s still early days of the coupling of these sorts of two different data types. So assuming that there’s more fluidity across those two boundaries or, you know, those boundaries over time, what sorts of doors do you think it will open for organizations?
Erin Kelly 46:47
Lots of them! (chuckles) Go ahead, Francine.
Francine Klein 46:47
One thing that comes to mind when the ops and analytics start colliding is the obvious, right? The obvious in terms of the stops and starts, the delays, the layers of infrastructure that people are not focusing on the business value because they’re focusing on just moving data to and from these systems and reconciling. So like some of the obvious ones that we’ve all seen in terms of the total cost of ownership. How do we focus on the insights? But now, what are those insights that people can start focusing on, and when we’ve now removed that technical and the team layers, right (you know, the different types of teams and processes and so forth), it almost feels like data becomes more of a natural dialogue. And so we can talk about data and the way that we think about anything else. We think about the fluidity of like, we can talk about data and the way that non-data people might think of it as it’s all just information. And so we can then start really thinking about experiences that feel more human, as opposed to “Hold on, I have my sales process. And then I have my analytics over here, but it looks at sales process.” And so we’ll just almost feel more human.
Erin Kelly 47:57
Yeah. So it’s like, I guess, as you’re explaining, I’m thinking it’s adding, contextualizing the insights. It’s adding the information. It’s adding the other pieces that are required to take, again, that 5% conversion: Is that good, is that bad? What are the levers that I can pull to actually drive that activation or increase in that? And then what does that impact going to look like down the line and, kind of, raising the overall… I like vernacular, or the common language. From, like, a broader business model/business architecture standpoint, thinking about customers, products and services, operations strategy. For us, customers like deeper understanding, deeper engagement, you know, the relationship they can build getting the right product at the right time to the right customer, and doing that in a way that seems authentic to whatever that mission is for the organization. From a products and services standpoint, being able to make more nimble decisions on what product is right, what services, when is the cost per acquisition for a particular product outsized and you need to move on; helping to inform some of those decisions and not feeling like this is the finger pointing game, but this is data-informed information of how you’re going to actually translate that into the business side. And I think operations stuff really speaks to itself in terms of efficiencies and other pieces like that where you can start to really move the dial and move the lever, in terms of bringing the the worlds togethe. But yeah, I think it’s pretty exciting. And just thinking about the world as even a consumer or outside of an organization — not necessarily as a consultant or service provider or product, but as a consumer — I see the future of it just feeling better. [I see] it feeling like more authentic, less hassle, less noise, you know? “I don’t need to give you some much more about myself, but the organization’s actually doing a better job of connecting it.” So that it is less hassle. And, you know, there’s enough noise in the world, let’s use data to actually calm some of that, so that the real things can take place.
Francine Klein 50:13
And it’s funny because, you know, we saw the art of the possible, both in our current roles and also at Slalom. I can’t help but think about backing up fingers to keyboard, you know? I came into the coding world. And so these are the possible [and] really, really awesome things that I think are gonna open up. And one of the things we talked about with back to basics that I don’t think will ever go away is 101. We’ve got to describe the entities: What is the customer? What is the product? What is an employee? And once we get these, like, I guess, nouns as the entities in a non-data word… Once we define these nouns appropriately and [agree on] a common language across everyone of what this noun is, then all the other things that come off of the noun, all the verbs and events and activities can come through. And so I can’t help it always go back to basics of like, we’ve just got to have the same definition of noun.
Tim Zonca 51:09
Yep. Yep. Francine, you mentioned just the notion of like the art of the possible. And so in the final questions as we start to wrap up, as relates to the art of the possible, if each of you could get anything you wanted in the next — let’s keep the timeframe kind of short — six to twelve months or so that would help unlock something in this data sharing space as you work with customers, what would it be?
Erin Kelly 51:35
You can go ahead, Francine.
Francine Klein 51:38
I think I kind of hit on that. But I think that there’s this concept, and I don’t know the right word. And I know that someone much better at branding is going to come up with a much better word. But it’s, like, a Data UN, where these people come together and figure it out. Right? We know that data between the walls of a company is hard, right? 100%, each company struggles with it. And it’s something that there’s a ton of tools and processes and procedures to do so when approaches. And we’ll continue to try to solve that as people fix their processes and evolve them over time. But I think I think the next thing, right, as we keep doing that, we’re going to always have to put away our “laundry,” right? Our laundry is always going to be continuous. But I think the next thing is as data across companies, and this concept of this UN, where companies have to kind of work together for a shared goal, but some guidelines that kind of keep them in line to do so.
Tim Zonca 52:34
Yeah, it’s interesting. Yeah! I’ve heard the term “data alliance” from some folks kind of popping around. That sounds really similar to kind of this vision you have. Yeah, just a better word. Erin, what were you…?
Erin Kelly 52:45
Yeah, I was just gonna build on a little bit with what Francine was saying, and then then I’ll answer the question. But the build on is like, historically, when we think about setting up data organizations, there’s traditionally a few different models (there’s not an infinite amount of models). It’s a centralized model. It’s a decentralized model. And then it’s a hybrid model. And typically, when we enter organizations, or back in the days of Slalom, it’s super decentralized. Everything is running all over the place — Wild Wild West. Sometimes there’s that gut feel of, “Okay, centralize first, and then control.” But the control is never giving the right information out at the right time to everybody, especially as the organization matures. And to your point around that Data UN or that data alliance, what is the new hybrid model between businesses and process and vendor management — and all those pieces — that is the best practice? [What is] the blueprint of how organizations should be thinking about data on a day-to-day basis? And history can show and teach us a lot around how some of these pretty complex situations might be able to actually have some patterns that are there now. [I’m] not saying it’s easy, but it’s just an interesting piece to pull on. And, you know, I think from just, to answer the question in terms of the next six to twelve months and how to unlock and make it maybe more bulletproof and less scary for organizations and open that up… I think I’ve got to double down on the security, the governance to control like those pieces there. I think we, especially in the sport space, there’s a lot of talk around data security, data privacy and other pieces like that. To me, 100%, those are absolutely forefront both for KAGR and for the industry. We’ve got to be able to break that down, though, into bite-sized chunks that can be configured. Controlled. Other pieces, like beyond just the “Don’t send anything” — it’s a how do we actually send this in the right way? What are the right governance and controls that need to be in place? And I guess it’s kind of tangentially related to that Data UN. What are the right blueprints? There’s a lot of pieces out there in terms of either legal policies or other pieces that no one has really stepped in and solved with “Here’s a couple of ways that this thing can actually happen.” And I think there are some organizations that I’m sure are doing it better than others. How can we actually learn and establish those pieces? And again, it dovetails back to what we were talking about: The education. The more information and the more impact and use cases that we can have around this whole process controls perspective, would be so fantastic. I think, for all organizations, and particularly in the sports space, and really making sure that they’re protecting, you know, all assets. And from a data perspective.
Francine Klein 55:39
We need a model alliance to start practicing it with The Youth.
Erin Kelly 55:42
(laughs)
Tim Zonca 55:43
(laughs)
Erin Kelly 55:43
That’s like the modern day Debate Club. We should be injecting that, Francine. And like, what’s the next data governance, data sharing perspective? I mean, hey, there’s probably some really cool ideas in [with]the youth of whatever apps are being used. I’m certainly not cool enough to be using all of them, though, I am a technology lover. But I’m sure there’s some really interesting insights that some creative minds could put together on how to solve this problem.
Francine Klein 56:19
And they don’t have the weight of the systems and processes and nomenclature. They could just think, “Well, why would I want to have to wait?” There could be something there.
Erin Kelly 56:28
There could be something there. Next podcast, Tim.
Francine Klein 56:31
(chuckles) Yeah.
Tim Zonca 56:32
Well, thanks to you both. I appreciate not only your time, but just you both sharing your expertise with our listeners. So thanks to both of you, Erin Kelly and Francine Klein, for all the real talk on real time data sharing. I appreciate it. And thanks to all of you also for listening in. If you’re interested in learning more about the various organizations, products, research, etc. mentioned in any of our episodes, just visit vendia.com/resources/circles-of-trust for all the links. And when you’re ready to keep the conversation going, download or stream all of our episodes. You can get them on Spotify, Apple Music, wherever the top streaming services are. And if you have a point of view on the challenges, power potential of real-time data sharing and you want to be a guest on Circles of Trust, email us at [email protected] or DM vendiaHQ on Twitter and mention Circles of Trust. And thanks again for joining us. If you like what you hear, take a moment to drop us a few stars or a favorable review and share with your colleagues. Until next time. Thank you, all.
Get Circles of Trust in your inbox
Join the Circles of Trust community to get real talk on real-time data sharing right in your inbox. Or subscribe now on Spotify, Apple, or Google Podcasts.