In this second Tag1 Team Talk episode about the Insights Supply Chain Framework, we examine this structured approach designed to provide a clear strategy for organizing your data teams, eliminating ad hoc collaboration, and ensuring insights flow seamlessly.

Discover the fundamental decision every organization must make: whether to keep all data experts in one centralized team or spread them across the business in a decentralized model. Learn how organizational context, not size, is the critical factor for success and why this framework is essential for building robust career paths that attract and retain top data talent.

Tune in to explore the framework's core principles and set the stage for our next discussion on the Data Organization Matrix (DOM) and the impact of viewing data as a core competency.

Listen now to get the definitive strategy for structuring your data team!

That's why Dr. Duru Ahanotu, the leader of Tag1's data strategy team, created the insight supply chain framework. It brings structure to how you organize your data professionals and teams to leverage your data and enable insights to flow through your organization.

This is the second episode in our series on the framework. In our first episode, we provided a general overview. Today we're diving deep into a critical decision. Organizations have to determine whether or not to keep all of their data experts together in one team, a centralized approach, or spread them out across the organization into different business units.

A decentralized approach. Picking the right structure for your organization is fundamental to your success with your data professionals and your data. In upcoming episodes, we're gonna explore the data organizational matrix or DOM, and whether data is or should be a core competency of your organization and how that's going to factor into your decision to be centralized or decentralized.

We're also gonna do episodes on how AI has been impacting this framework, uh, and we'll dig a little bit more into the data maturity curve, what it is, and how you can progress along it. Thanks for joining us. Let's get started.

​

[00:01:50] Dr. Duru Ahanotu: Thank you, Michael.

It's good to be here.

Uh, Duru brings tremendous academic credentials and real world experience to help organizations address one of the biggest challenges that they face today. And in today's episode, we're gonna showcase the real world experience and examples of how he's applied this framework across his career. But before we dig into the insights supply chain, I wanna give you a little insight into Tag1.

Tag1 is the number two all time contributor to Drupal, the world's second most popular content management system. For nearly 20 years now, we've been the architects of the open web, leading the creation of the software and best practices that powers millions of websites and hundreds of thousands of organizations worldwide.

We're a full service strategic partner, applying that same architectural expertise across technologies and throughout your organization. From discovery and design to building and scaling complex applications, we lead AI strategy and implementation design and manage infrastructure and architect large scale web applications across a wide range of platforms.

We're trusted by industry leaders, including Sumitomo, NTT Data, the European Patent Office, and the American Federation of Teachers to solve mission critical challenges and build lasting solutions. Check out Tag1.com to learn how we can help you. So, Duru, for folks who missed our first episode or who need a refresher, um, we talked about the origin story, an overview of the Insight Supply Chain Framework.

Um, can you walk us through, just at a high level, um, what is the problem that your framework is helping organizations address, and why is it fundamental to solving those challenges?

And within that, I also, within that framework, sorry, I created just a very structured approach to understanding how data flows through an organization, going from the raw data to the insights and the different kind of roles that need to be placed along the insights supply chain, um, to support decision making in the organization.

[00:04:32] Dr. Duru Ahanotu:

You know, what data, should be a core competency or is a core competency. Um, you know, uh, we don't wanna get too deep, but, uh, just a sense of, what that is and, and how it might impact your decision making down the line.

That means that data is central to your organization's success in either supporting the products and services that you deliver to the marketplace or the data in itself, is the product or the service that you're delivering to your customers. And so once you've answered that question, you'll know how to place yourself on that data org matrix and decide between centralizing and decentralizing the skill sets, and then also centralizing or decentralizing the teams that, uh, those skill sets are staffed with.

[00:05:53] Dr. Duru Ahanotu: mm-hmm.

[00:06:00] Dr. Duru Ahanotu: So this is the simp, the simple, uh, the simplest, uh, matrix. And what it's saying is on one axis is the centralized, decentralized, uh, skillset.

So you have to decide, um, whether you want to hire, basically, think of it this way, you want to hire a bunch of generalists as your data professionals, or you want to hire, um, specialists as your data professionals. So a generalist will basically be able to function just about every role along the inside supply chain.

And then specialists will actually have very specific slot. Where they, uh, deliver, you know, their, um, expertise along the insight supply chain. And then along the other axis is, well, how are you going to organize, uh, those folks? Um, and again, you can centralize them or decentralize them. And then each of those specific points in that matrix tells you exactly how then people in teams will collaborate when they are working in the insights supply chain in support of, again, the business, uh, and supporting decisions, um, that the business needs to make to drive value for its customers.

[00:07:14] Dr. Duru Ahanotu: Yeah. So there's two things to do. One is the, you already have a bunch of data professionals and you're trying to, you know, figure out whether you need to restructure them. The other one is you're starting fresh and you want to hire, right?

So it depends upon your, where you are on the data maturity curve, essentially. So let's take the, what's perhaps the, the more common case, which is you already have people who are doing data in an organization. Um, and particularly if you haven't thought much about it, you'll have scattered skill sets across the organization, right?

So there might be some data, um, skilled people in sales, and some folks in marketing and in operations, um, and or even data folks who are specifically supporting the executives and so on. So in this case, what you need to look at is, okay, am I serving these, uh, data professionals well, in the way that they're organized and are they.

Are they set up to succeed in, again, delivering the products and services of the business? Um, and how do you answer that? Well, you answer that by saying, okay, um, let's take a, a simple example, which is, I have one product, one service. In most cases, you're gonna wanna centralize the data team because in that scenario, if you have one product line, one service line, you need to be able to standardize all of your data processes and your metrics.

And the best way to do that is to have all those data folks in one place. Um, and so that would be the centralized decision because you need everyone aligned on this one, let's just call it semantic model of how the business translates into data structures. Or you could think of the other way how data structures translate into the business.

So that's one decision point. Um, now the other one is, okay, again, same company. But you look around the company and actually you have different flavors of this product or business, and they're different business units. And those different business units are very distinct from each other. They don't overlap.

Um, you know, they may use the same hr, they're probably gonna use the same hr, maybe even the same sales folks. But the products in themselves are very distinct. And then the data models that support that product are very distinct. And in that case, centralizing may not be the way to go because you need folks who are very close to those business units and you need them to be able to collaborate very, uh, tightly.

And the standardization that they're gonna do is more along the lines of the business units or the product verticals and not across the entire company. So in that case, you need to enable the different business units, um, a way of having close contact with their, um, data professionals. And so in that case, you're going to decentralize, um, your data teams and then leave it up to the business units to figure out whether they want to organize their folks as specialists or generalists, it's all gonna depend upon their use cases.

But is it say? Like, size of organization might matter, you know, your perspective like, um, are, would people look at this and make different decisions based on their organization and context?

Now, of course there is loose correlation. As businesses grow, they tend to have, um, distinct divisions and all that kind of thing, but it's really the context. And I like to stay focused on what is it that you're delivering to the marketplace? What solutions, what products are you delivering to the marketplace?

And how do those look? And so the data, everyone, particularly the data folks, need to be closely aligned to how you're delivering value to your customers. Um. The size will impact sometimes the opportunities you're able to give to your data professionals. Um, so for instance, in a larger organization, you might be able to co-create complete career paths for a even a decentralized structure.

Um, if you're a very small company like a startup or whatever, um, the career paths will look a little different. They may not be extended, but there may be expanded roles and responsibilities, uh, that you can give to people, um, as the startup is in its early, um, phases of growth or the small companies in its early phases of growth.

So, long story short, it's the context first. Uh, then everything else falls into place from there.

So when you think about how you, you know, implement this, how you leverage data resources, how you use 'em in your organization, you know, that's a really important factor to consider. You want to be in a position where you can not only attract great talent, but retain them and provide them with a career path.

And I just thought it was super cool that your framework, um, you know, no doubt because a data professional is something that's important to you.

[00:13:00] Michael Meyers: Um, but it's something that organizations really should be thinking about and I, I just thought it was great that, that that was something that you highlighted and mentioned because I, you know, it's, it's critical.

So I, very early on, I was a strong advocate, particularly when I was building out large centralized teams. Mm-hmm. Um, I was a, I was a very vocal advocate for building out those paths for my, uh, for my people, for my teammates. And I've carried that forward, uh, since then.

Mm-hmm. You know? Mm-hmm. And, and that could be a double-edged sword. I, I dunno how much you wanna get into that today, but, um, you know, the reality is, you know, that, that people are, are, you know, are using tools and applications more and more throughout their day. You talked about

[00:14:11] Michael Meyers: You know, sales, marketing, finance, like everyone is, is, you know, using one or more tools to collect and synthesize data.

Um,

[00:14:22] Michael Meyers: So, I know that you're talking, you know, about, um, data professionals, um mm-hmm. But, so like, how do you reconcile that? Where like, I, I know, know that I would call them data professionals, but there's certainly data users and they're decentralized by nature because they're everywhere.

[00:14:41] Michael Meyers: does that factor into this?

There are data users in there, but nobody's going to build their career on being a data user. They're gonna build their career as being a marketer, a great salesperson, a great operations folk, a great HR person, and so on. I'm talking about in this particular case. So the data professionals in the organization, their role is to enable the success of all the data users in the organization.

So again, going back to insight supply chain, you can look at this as I'm going to centralize my data team. They're going to create standards that apply across the organization. And when, for instance, the sales folks are trying to define things like, you know, revenue and um, uh, conversion rates and all that kind of stuff, you're going to support them in that.

The data pipelines that are created to, uh, store the data that they're collecting from their sales calls and sales, um, uh, sorry, their accounts. You're gonna make sure that all those pipelines are robust and that they don't have to think too hard about all of the, um, you know, technical complexity that might go into data governance and security and all of those kinds of things.

They just help them focus on the immediate problem they're trying to solve, which is, for them, data is just one tool that is going to get the job done. Whereas for the data professional, the data is the tool and that tool I am, um, distributing across the organization to help everyone do their job better.

Um, so me as a data professional, my career depends upon data excellence, you know, centers of excellence in data, um, all the things of data, but then the sales folks and all the other business units, their career paths are still built on marketing, sales, and finance and whatnot. And they are going to collaborate with the centralized team to make sure that that data tool is as strong as all the other tools that they use to get their job done.

[00:17:30] Dr. Duru Ahanotu: So when I, um, joined Yahoo, um, I, the data, let's just call it the data professionals are very decentralized. And so I joined a product team. And in those days it was a very exciting time because, uh, Yahoo was going to build, uh, well, had built a social media, essentially a social media stream called Vitality.

And so I joined that team, um, as a, oh, I can't even remember exactly my title was, but I was a data analyst type person. And what I did is I came in and all of these product managers and program managers had their own spreadsheets, tracking, you know, all sorts of metrics for their different products and programs and then.

Uh, I think it was once a week they would all get together in a meeting and then the entire meeting was spent with all of them going around and sharing their different spreadsheets. And yet we were all in the same product team. Um, and so my role when I came in there was to relieve them of that because these folks weren't data professionals.

Data was just one of these onerous things that they had to use to create all these reports. Mm-hmm. So when I came in, I took all of that off of their, uh, shoulders. I went to every single product manager and program manager in the group, found out what they did, what they needed to report, what were their metrics, and then I consolidated, and then I found the commonalities, right, because there was overlap in terms of some of the metrics that these folks were producing.

And so I integrated everything into one whole. Um, and I made one big spreadsheet that was integrated. This is sort of a precursor. Um, back then I was going to move it to a database, but before I could do that, Yahoo decided to centralize all of its data professionals. And so I was plucked out of that product team and put into a central team, and I joined a bunch of other people who had been plucked out of different verticals.

And we were put together, um, and we joined basically what had been a more technical oriented infrastructure type data team. And now we were a comprehensive. Um, data team. And in that role I started to grow a team. The, the SVP at the time of the data team asked me to build out a more complete centralized team that would serve all of Yahoo's product teams across the globe.

It took almost a year, but I, um, hired out, I think it was a team of almost 20. Some of them came from, again, teams that had been, you know, people had been plucked out from different places. And then I hired to, to fill that out. So it had data scientists, what we call business analysts, um, and, and what we called at Yahoo Insights analysts.

And you might see that in my insights supply chain. I had that insights analyst at the end of the supply chain. Uh, anyway, so after I built out that team, lo and behold, uh, the CEO decided, you know what? I actually want to decentralize my data professionals. And off we went, uh, scattered to the winds yet again.

Um, interestingly, uh, just as a quick sidebar, it was a different model. You know, usually you're told from the top down where you're gonna go. We were actually all given the opportunity to go to figure out where we wanted to go in the organization. Um, so all of my team members, um, and colleagues. Kind of had natural places to go because they just went to the teams that they were working with.

But me as the manager, the leader of the team, I actually had an open opportunity to either stay in that centralized the, the remainder of the centralized data team, or find another team to join. I decided to join marketing. And the interesting thing about marketing is that it was one of the last functions that actually had, um, a purview across the whole organization.

So they still had to work with all sorts of different verticals or, or product units in the company. And so once I was in there, a few of my team members came. I absorbed some other folks that were already some data scientists that were in marketing. And so we created our own centralized marketing data science and data engineering team within marketing.

Um, but we only served the folks within marketing, right? So this is a centralized model. Each of us were specialists. But we were in one team called marketing and then other verticals may have had other data, professional structures. And so I looked at this experience and I said, you know, along the, each of these two steps, I never really knew what the strategy was or what the reasoning was except that the leader at that time.

just had a particular preference for how they wanted to organize these, and I said, there's gotta be a better way. And so using that experience and my experience actually working at other companies before that, I came up with this insight supply chain to provide a framework for thinking through how to organize, uh, data professionals.

And I stuck with it ever since, and I've of course have modified it over time, including in today's AI driven world, I'm now incorporating the lessons and principles of AI into the insights supply chain as well. So starting that was a very long answer to your, your question, but you know, I've, I've worked a long time and I have a lot of different, uh, experiences that have been related to the Insights Supply Chain.

[00:23:14] Dr. Duru Ahanotu: Yeah. So. So my, my kind of flip answer to that question is, we know from school, um, in engineering math course, whatever, there's always a brute force wave that's going to work and it will get you there.

And it's called brute force for a reason, because you're gonna be sweating, panting, and you'll be exhausted by the end of it. Um, and meanwhile, you've used up resources, time and energy and attention that you could have applied somewhere else. So you're not gonna be as effective as an overall data professional.

You know, just pounding and pounding, pounding to make things work. This is why structures, principles, algorithms, formulas are, are very useful because they cut through, uh, the need, uh, for brute force. So, yes. Uh, In fact, organizations will default to some sort of, uh, format if they're not purposeful or intentional about how they organize their data professionals.

And I believe the, the default that will happen is that people will just sort of trickle. To what whatever team or organization is the loudest, um, about the need for data, uh, resources. And so you'll just have people in teams that have, you know, that have been grabbing at the company's attention the most.

Is that gonna be the most effective? Maybe by just pure random luck, you can stumble into the right form, but why rely on that when there are principles and structures and frameworks like the insights supply chain. Mm-hmm. Um, you know, to make, make, to smooth out everything and make it a very rational process.

So again, this is self-interested because I wouldn't wanna work in a place where the company is just organizing and leading us by brute force.

[00:25:09] Dr. Duru Ahanotu: Right.

And, you know, the organizational shift and upheaval that's created by these changes, um, you know, uh, the amount of, you know, uh, energy that is lost, time that is lost. Mm-hmm. Um, and then someone comes in and makes a decision and, and it's just like, so I, I appreciate the fact that you can brief force things in certain ways at certain times to some degree, um mm-hmm.

But also, you know, the idea that you should really think through. These decisions and not make them, you know, haphazardly, you know, make them very thoughtfully. Um, you know, it's, it's good to hear that in your experience that this has a definitive benefit.

Yep. Yeah, and I'll

Again, it, you know, I'm very biased towards supporting the development of data professionals and supporting their development through a career path in a properly structured data ecosystem. Um, and so that motivation also is embedded within the insights supply chain. Right. Um, and I think that when people are operating in a system where the framework is clear, that should tend to build uh, clearer career paths and more motivation and enthusiasm about working in a place where you know that your contributions

are making an impact, um, where the business is finding greatest value. And I think in this day and age, you know, I've seen a greater and greater awareness. You know, I've followed a lot of podcasts and whatnot, and I've seen over the year, the last few years, the greater awareness telling and counseling younger data analysts.

Look, don't make your job all about creating reports and dashboards and focusing on the technical mechanics of, you know, organizing and analyzing data. You need to understand what the business is about. You need to be connected to where the value's being created in your business. Uh, and again, the insights supply chain will give you sort of a visibility into where in the organization do I want to plug myself into, or how can I make myself most of value in the organization in this new era where data analysts, their career longevity depends upon, you know, the more the insights, uh, analyst part of the supply chain insights supply chain.

Mm-hmm.

Well, thank you so much for, for digging into the centralization versus decentralization more. Uh, I look forward to our future episodes on this.

It's been great to dig into this deeper. Uh, we're gonna talk about Data Org Matrix coming up. Uh, and you know, how data as a core competency impacts the centralization versus decentralization decision. So we're gonna get into this further. Uh, we're gonna talk more about the, uh, Data Maturity Curve, which has come up a bunch of times and mm-hmm.

Uh, get into how AI has been factoring into your thinking. Um, folks, you can check out, uh, these and other episodes at tag1.com/podcast. Uh, we'd love your feedback and insights. It's always great to hear from you. You can reach us at info@tag1.com, uh, and subscribe so you don't miss future conversations.

Uh, special thanks to Tracy Cooper, June Gregg for producing today's episode with input from Hank Vanzile and Cassey Bowden. Until next time, take care.

​