Data Wins! Inside the NFL's Analytics Strategy

Michael Krigsman: Today we're speaking with Paul Ballew, Chief Data and Analytics Officer of the NFL about the league's use of data, analytics, and AI to optimize game performance, ensure player safety, improve fan engagement, and increase revenue streams.

Our guest co-host is QuHarrison Terry, Head of Growth Marketing at the Mark Cuban Companies.

NFL data goals: improve fan experience, league operations, and team performance

Paul Ballew: At the center of the journey always is can you help organizations do two things: Connect with their customers (in our case our fans) better and more intimately. And then secondly, can you improve your operations? In our case, that involves improving the game.

As I look at the journey that I've been on, it's interesting that I've gotten to this point. I have a deep love for sports, deep love for the game, and now we're at a point where sports in particular is focused on using data and analytics, the science that comes from data and analytics, to accomplish those two goals.

Very similar to my journey in automotive, my journey in financial services and retail (over the last three-plus decades), now I get to talk about football all day long. So, on top of worrying about the technical side of my job, I get to talk about football. You think about two wonderful things going on – my passion for data and analytics, my passion for sports – put the two of them together and my days are quite interesting and, to be honest, a lot of fun.

How the NFL uses data and analytics to personalize fan engagement and customer experience

QuHarrison Terry: One of the things that is exciting about that (but also acknowledges where you've been) is in retail, banking, and now even sports on the football side, you still have to worry about the consumer. In what ways has the consumer experience part of your job changed and what differs now compared to even the previous parts of your career?

Paul Ballew: If you go way back when because, as Michael knows, I've been at this a long time, we really struggled to understand who you were and, therefore, how to connect with you in ways that are meaningful. I always like to say we want to engage with you meaningfully in context. That is the right conversation with you that you want to have with us.

Decades ago, we just struggled because we didn't have the capabilities to do that, so it resulted in very coarse things that we did. We surveyed you. We tested things coarsely. We sent a lot of things into your mailbox. Not email mailbox, by the way. Your actual postal code mailbox.

Well, today, our ability to see and know and engage with you is driven by the fact that we have the assets from a data standpoint, from an analytics standpoint, and an engagement standpoint. Think about marketing technology and all the changes in the last decade.

What the league embraced is the fact that we, despite the success of the NFL, with the world changing in terms of how content is being consumed and who owns that relationship, we have to get in front of that. We have to, with the respect and the privilege of our fans, know them and engage with them in the way they want to be engaged with.

Our largest initiative at the league (from a data and analytics standpoint) is our one-to-one program. And so, very similar to what you see with retailers or in automotive or in financial services, we're now focused on doing the same thing, and other sports leagues are getting there as well because, if you're going to be successful, you have to connect in the way fans want to be connected.

By the way, under the age of 40, they want to be connected even more differently than just turning on the game every Sunday. That's a big part of the journey we're on what I like about it is it really shows the power and leverages the power of what a data and analytics organization can do.

Michael Krigsman: What is the role then of data and analytics in getting closer to the fans and addressing those customer experience goals that you were just describing?

Paul Ballew: Our role is a significant one because, in partnership with marketing, in partnership with our clubs, our media organization, we're providing the ingredients to bake the right souffle, as I would describe it. We're responsible for bringing together that complete view.

We're actually the first sports league that has aligned with our clubs to bring all of our data on fans together in a systematic way. That's a big step forward for us because it unlocks our ability to understand tens of millions of fans systematically and increasingly pretty close to real-time.

Then in working with marketing and media and other channels, we are driving those insights into the relevant conversations to those individual fans. Then, of course, measuring and constantly learning through the feedback loops. That's extremely important because, once again, connecting with fans in a way that makes a difference in their relationship with us is how you deepen that relationship with us.

We don't take for granted that we are the biggest sports league in the United States and, you can argue, the biggest in the world. We understand that, going forward, it's about deepening that relationship.

That's what great companies and what great organizations do. And you can't do that if you don't see and know.

You can talk about how much you love your customers, but if you can't see and know and then act upon that information it's problematic. So, we're focused on everything on the data side, systematically, all the way from acquisition to governance, the usage, everything on the technology side, and then, of course, the analytics and the insights that get driven through that ecosystem so we're engaging in a way that makes a difference.

It's to the credit of the commissioner and to the credit of our owners. They've embraced this because they understand that the future has more of a direct-to-consumer focus and all the fragmentation in media that's out there, and we have to be prepared for it.

Privacy and consent: addressing customer concerns in data usage

Michael Krigsman: We have a really interesting question from Twitter, a very important question on this topic. This is from Arsalan Khan. He wants to know, "How do you manage the privacy aspects as you're addressing these customer experience goals, fan experience goals?"

Paul Ballew: One of the pillars is our data governance organization, and we have a group within that that's responsible for consent, working with our legal team and the general councils of all the clubs. We start off with our philosophy on consent and privacy, and that is: We are going to be transparent, we are going to give our fans controls to express their preferences if they choose to do so in their relationship with us, and then we're going to make sure that the privilege of having that data is used in a way that provides them value. That we're not just doing it so we can carpet-bomb communications to them. We're focused very much on making sure that the communications are relevant. That's the journey we're on.

Twenty years ago when we started going down the path of data governance, issues around privacy were very limited because we were in the early days of gathering transactional data assets on individuals. We're now at the point where it has to be one of the pillars of the overall strategy, and we're very proud of the fact that we made it a pillar, one of four pillars of the strategy around one-to-one. Every day, we spend time on it, and every day, we're making sure we're doing it the right way, not only in the U.S. but globally because it's a complex journey.

By the way, if you don't do this right, there are brand issues as well as compliance issues. Of course, the NFL spends a lot of time focusing on the brand side of it.

Governance and the democratization of data in the NFL

QuHarrison Terry: Paul, one of the questions I have around that is centered in the form of accessibility. Now, externally, you have your partners in the clubs, and they know how to manage the data. I'm not so much worried about it from that end but more so internally.

You've got a lot of partners. You've got a lot of fans. And you've got a ton of teams that you've got to manage. How do you make the data accessible internally and go about it in a way where it ensures near to real-time visibility so people have the right insights but are doing so in a way where you can with win the data?

Paul Ballew: Our approach is not only data governance, which is all about what we're doing on standardization and data quality and access controls and so on, but we're focused in partnership with marketing on marketing governance. How are we using the data? How are you orchestrating it? Then analytic governance, which goes through the validation processes.

It sounds like a lot of governance. It doesn't have to be a government-type governance program, but you have to have those elements in place. And you've got to run this with that degree of discipline.

We all believe in this theoretically, this concept of democratization of data. Create it for everybody. Make sure everybody has access to it.

But at the end of the day, you have to govern it appropriately, and that's why we've built this whole ecosystem – involving our clubs, by the way, in that ecosystem – because there are 33 entities at the NFL, and you have to be able to orchestrate, coordinate, understand what you're doing or guess what happens. Everybody wants to communicate to everybody all the time, and that uncoordinated side of it not only causes compliance issues but the bigger risk in all that is that fans start tuning you out. Customers start tuning you out.

That's not what this is all about. This is about deepening the relationship, which, by the way, leads you to things such as sometimes more is not better. Less is better in terms of your communication. That's all been built into the system we're launching and the system we have developed in the last year or so.

Michael Krigsman: Be sure to subscribe to our YouTube channel and hit the subscribe button at the top of our website so we can send you our newsletter and notify you of upcoming live shows.

A lot of your effort then is focused on that fan experience. You mentioned there are four pillars. Can you describe what those are?

Four pillars of the NFL’s data and analytics strategy

Paul Ballew: We support the league on every initiative: player health and safety, football operations, officiating, media operations, media optimization. We go across the entire league.

While the fan is a big part of what we do, all these other activities are equally as important because our commitment is to help the league achieve its strategic objectives, which is to make the game better, to maintain the integrity of the game, to make sure that our content is being consumed efficiently and effectively. It is very, very important that we look at it that way.

When you think about our structure in terms of the fan side of it, from a journey standpoint, it starts with data and, clearly, all the data management aspects of that. We then have a pillar around analytics and insights because, once you have the ingredients, the question is, what insights are going to generate and how are you going to continuously measure and track the robustness?

Then we have the pillar around marketing technology and marketing governance and orchestration. And then the last pillar is privacy and consent and the related enabling capabilities because, within that bucket, you also have identity resolution and related capabilities you have to have if you're going to do this at scale because, when you think about the data complexities that we're bringing in, it's pretty hard to govern systematically your relationship with an individual if you can't know them. And to know them, you have to see them. And you have to validate that they are them, which by the way is poor grammar, but you get the point that you go down that path.

Think of it as a full ecosystem that then gets wrapped around the needs of the organization. Then you execute off that ecosystem through your channels.

Managing and leveraging data from the Super Bowl

QuHarrison Terry: That's related to the data that you have, so you talked about the breadth of the data, but one thing I want to point out in particular with the NFL is the depth. Not only do you have a pretty well-structured information pertaining to the games and engagements and even the fans and the stadiums, but you are in charge of one of the largest media events in the world. That is the Super Bowl.

As a marketer, that is one of the most fascinating times because everyone is tuned in, right? With that, what data? I mean you have to manage the data of the Super Bowl, right? Am I correct in that?

Paul Ballew: Yeah. Yeah. Just to be clear, it is the largest event, not one of them.

QuHarrison Terry: Okay. Okay.

Paul Ballew: [Laughter]

QuHarrison Terry: It's the largest event, and your teams are—

Paul Ballew: I couldn't resist that because it is the largest event. Two hundred million viewers, by the way. And the Super Bowl had just concluded, so just wrap your head around that. No other event gets anywhere above 20 million viewers other than our playoff games and Monday nights and Sunday nights and so on.

It's a wonderful question. The breadth and the depth of the data we have is pretty amazing, to tell you the truth. It's one of the situations where we're gathering data from what fans are doing in stadiums and what they're doing with the clubs. But we also have relationships, with the fans' permissions, where we're getting the data from our partners in terms of transactional relationships that we have with those partners.

We also have a data development organization that's constantly focused on what new data assets can we acquire and how can we bring them in. So, we're blessed with actually the opportunity to have lots and lots of data from a breadth and depth perspective.

The real challenge we have is like most organizations that have been around for a long time. That data was created for a specific use case. Now we're trying to leverage it for other things to get this holistic picture.

A big part of our journey has been, can we bring the data together? Can we cleanse it? Can we standardize it? Can we deal with the ability of properly associating it with an individual?

It's pretty amazing to think about where we're at. It's also pretty amazing to think about where we're heading because we're still in the early phases. With the right permissions in place and the support from fans, what we're going to be able to build in terms of our version of a wall garden is pretty amazing when you think of the tens of millions of fans that we can see and know. Because the passion for the sport is so high and with the right respect for them, we can build very deep relationships, so it's a great question.

Yes, the Super Bowl is the largest event. Maybe the World Cup every four years, you could argue. But we're every year.

Michael Krigsman: Just to be crystal clear about that point, the Super Bowl is the largest—

Paul Ballew: I couldn't resist.

Michael Krigsman: [Laughter]

Paul Ballew: Qu and I are having a good time.

How the NFL uses data to understand how fans consume content

QuHarrison Terry: No, I think it's great. I think the thing is, in my mind, I just have so many questions about it and I want to respect the time that we have.

To me, as a marketer, that's the dream, right? You have one event, 200 million people that tune in worldwide.

You have pop culture sports, entertainment, some of the best products and partners that also coincide with this. Dissecting that data and then leveraging it to do it again the following year, and then even sharing that with your partners and your teams, that's an incredible role where you're defining, really, the future of most organizations because it's a reality now that most sports can (if they want to) obtain and build pretty large data sets. But how you maneuver said data set is probably the most important thing. We're learning from you.

Paul Ballew: Yeah. How do you leverage it? How do you bring it to life?

My favorite expression is, how do you bring it to life? And so, I'll give you a good example of this.

We've been developing an alternative way of assessing how our content is being consumed in a fragmented media world. Not just looking at traditional measurements of TV or even digital or streaming, but comprehensively.

How is it being consumed through social media? How is it being consumed through PR communications? How is it being consumed through legalized sports betting? All of those things around it.

We had to do that because of the fact that if you can get that data and understand the comprehensiveness of how your content is being really consumed, it really shapes where you want to put your emphasis, where you want to place your bets. What am I doing right? What else can I adjust in terms of my mix? That's been a fascinating journey for us to go down.

We've now validated, and we have those diagnostics to have another data set to understand what we see. We call it NFL as a platform, but it's really this ability to see how our content is being consumed and how much time an individual is spending with us.

By the way, in the month of January (in the U.S.), three billion hours – in fact, over three billion hours – were spent consuming NFL content in the United States just in one month – three billion hours.

QuHarrison Terry: You don't have a team that's that large that can ingest three billion hours' worth of data, so talk to us about your partners. I studied computer science, so some of the more fascinating case studies early on with AWS actually were with the NFL. Next Gen was the platform.

Paul Ballew: [Indiscernible]

The partner ecosystem for data and analytics technology in the NFL

QuHarrison Terry: Yes, okay. That project in particular, you all were first movers in a space that you didn't have to be. Is it because you need help processing the data? What's the idea there?

Paul Ballew: It's always this great question of do you build it, do you buy it, do you rent it? The way I describe it is that the right ecosystem now is Lego blocks.

Years ago when you were building the whole data and analytics and insight, the block as a whole, you'd get up in the morning and people think about it (as it used to be), I'd have to go to one vendor. I'd have to go to one partner. And you had closed systems and so on.

Now it's an ecosystem, and try to approach it from the approach of humility where I get up in the morning, and I think about this in the context of I don't have all the answers. We're not going to build everything in-house. So, how can we, at the end of the day, leverage our relationships with these great partners?

AWS is a really good example. Great partner with the NFL.

Next Gen Stats is a big enabler for us in football officiating, player health and safety. So, the partnership with AWS is just a good example of us saying if there's a partner that can fit into the ecosystem and help us do two things – get smarter faster, or fill a technology gap – that's how we look at it. It's really the journey we're at.

For me, going over the last few decades, I think it's probably one of the biggest pivot points we don't spend enough time on is that the ability to piece together different parts of the technology stack and that ecosystem has been transformed just in the last decade. If we just pause and take that step back because, again, even 2010-ish, we were still wedded to saying my stack had to be X, and I was dealing with this set of vendors.

Now, take a look at anybody's modern stack. The environment is in the cloud. The software tools are eclectic. They're all over the place. My analytic platform can basically be just about anything, which is what we do today. And all the piece parts are really coming together, so it's pretty fascinating.

To me, what you just asked is one of the critically important questions we as a field have to continue to stretch our thinking on because, at the end of the day, we're being asked to do two things: generate insights that are more precise and to take latency out of the equation. That's my version of Moore's law. By the way, if anybody plays around with that, I have rights to it because that was a concept I came up with. Any partner that helps us do that, we're going to be very interested in.

Decision-making based on data vs. intuition

Michael Krigsman: We have a very much related question. Again, Arsalan Khan comes back, and he wants to know, "How do you manage decision making based on data as opposed to gut feel when, in any industry, there's a human tendency among the people receiving that data to rely on their own experience and intuition as opposed to trusting that data?" As you're working with decision makers inside the league and at teams, how do you manage that aspect of it?

Paul Ballew: It's something you and I have actually talked about in the past. That is, you have to have an orientation that it's going to be art and science in established organizations and that, as you move towards science to support more and more decision making (going towards what we described as model-driven decisioning), you have to bring them along through the journey.

You have to deal with the change management, the human dimensions of this, the transformational side of this. It's one of the reasons why, within the field, you see many of us now having business transformation functions or we have translators or we have engagement managers because if you don't invest the time to do that, these organizations are established and people are established in the way they make decisions.

That old adage that "people don't object to change, they object to being changed" is true in this case. That's in any established organization, so that to me remains the secret sauce of engaging with the business partners, powering forward in a way where it's beneficial to them, bringing them along through the journey.

I can remember when I joined the league initially, the commissioner pulling me aside and saying, "Please, educate us," all the time, and he understood that. I understand it because of all the years I've been doing this.

If you look at the early legacy companies that had very large investments in data and analytics, I had one of the largest when I was at GM for a decade. We had one of the largest organizations that you would consider to be a data and analytics organization.

Now, we never got to full adoption of all the work we were doing because we underestimated the need to change and have the change management component in the equation.

How the NFL manages change and disruption in the professional sports industry

QuHarrison Terry: There are a couple of questions from Twitter, and I want to preface this question with just a precursor element. The NFL (and just sports in general) have changed.

I mean the NBA is doing this as well with the unbundling of the cable package, right? The regional sports networks, they're not making enough money, so there are possible bankruptcies that can ensue on that front, and that's going to have an impact on all of sports.

But more specifically, in this world where you're experiencing disruption, what are some of the shining stars that you're seeing on the horizon?

Paul Ballew: First, our strategy is to have optionality, and optionality because, for anybody that call the ball ten years from now, it's an interesting challenge. We've publicly stated multiple times (and we feel very good), we have long-term agreements with our players, and we have long-term agreements with our distribution channels.

That's great. It gives us time to make investments and prepare for the future. But part of the investments in preparing for the future is to have optionality in where you go and how your content is going to be distributed.

To do that well, though, you still have to make the game better all the time. You've got to connect with fans because if you own the relationship with the fans and you're making the game better, then your optionality for how you distribute is better because who knows what the future holds.

You've got to get the fundamentals right. You have to be prepared for that.

For us, we've been experimenting on a number of fronts. You saw Amazon with Thursday Night Football, for instance, had a great first season. We proved that we could disseminate content through that channel, which is a paid channel at scale with ten million viewers, roughly, every single week, and we had some weeks well above that.

That's where we're at, and we're continuing to experiment in trying to figure out where this is going. At the same time, what we're making sure is that we can control as much of our fate as possible.

The way you do that is make the game great, which it already is, but make it greater. Make it better, and then make sure you can connect with your fans, thus the investment in data and analytics is a critical enabler to us having that control of our future as much as we can.

QuHarrison Terry: With that, Chris Peterson from Twitter is asking the following question. "As far as disruptions on the horizon for the NFL in terms of either sports or control of all the data, how has the explosive growth of sports betting changed things for you all?"

Paul Ballew: It's a factor, obviously. It's something that, when you look at sports, we had to get comfortable with. You have to get comfortable with it because you don't want it to have negative repercussions on the game.

What we've done is we've continued to focus on striking the right balance, making sure that the integrity of the game is still there, and making sure that we're complying with how it's rolling out.

Obviously, it has business impacts because gambling organizations, the LSB organizations, are big advertisers, and they're promoting, and they're doing all those things that are there. From a data and analytics standpoint, we get lots of homework assignments around making sure that we're tracking and understanding, that there are no negative implications.

To me, I look at it as part of the overall puzzle of how we maintain the integrity of the game – how we help maintain because it's broader than us – maintain the integrity of the game for the future. But not a surprise that the world is there and not a surprise that it continues to grow.

It's important to fans. When we look at the amount of activity going around legalized sports betting and football, it's very, very high. We take a very large share of that in the season, a disproportionate share of the LSB activities between September and January. About three-quarters of it in sports are going to football.

How the NFL uses data and analytics to improve team performance

Michael Krigsman: Paul, we have not spoken at all about the usage of data and analytics to help improve team performance, to optimize the team, the football.

Paul Ballew: Yeah. [Laughter]

Michael Krigsman: Tell us about that. That seems pretty foundational.

Paul Ballew: Yeah, it is. Again, as I said, when you think about data and analytics in a modern sense, always think about how am I connecting with my customer, my fans, and then how am I improving my operations.

In automotive, it was about quality, efficiency, productivity, and throughput. In football, it's about improving the game, improving officiating, improving performance.

The way I always think of it is, in my side of it, my team is focused around supporting officiating and officiating analytics and rule changes. We provide a lot of support to player health and safety because of the workstreams we have there.

At the club level, they all now – every club, all 32 clubs – have analytics within their football part of their organization. If you think about clubs, they have a revenue part of their organization, which they have data and analytics teams, which work very closely with us. Then they have football teams or team members supporting the football side of their equation.

You see it. You see it in terms of football strategy. You see it in terms of probabilities being used.

Think about everybody is now amazed at how often teams go for it on fourth down. Well, that's driven by probability analysis. It makes more sense to go for it on fourth down then it does to punt.

And so, when we look back on it, years ago, I can always remember a game between the Pats and I think Ping was still in Indianapolis at the time where Belichick went for it on somewhere around his 35-yard-line in the 4th quarter. It was fourth and not one. It was more like fourth and six or seven.

Everybody was just appalled. Why did he go for it? Of course, the reason he went for it is he knew if he gave the ball back to Peyton, the probability of Peyton walking down the field and scoring was very, very high.

You could see that in the early days. Now it's systematically embedded in all teams. We'll see more and more analytics and technology being leveraged to continue to improve the quality of the game, the integrity of the game, the effectiveness of officiating.

My team is doing that for the league on a day-to-day basis. We've got lots of things to do every single day. We're asked to do a lot on that front, and it's actually one of the high points of my week is just all of the things we're working on, on that front, because it's really exciting to see how technology and data and analytics will continue to drive this great game and the excitement of the game.

To the earlier question, that's how you prepare for an uncertain future. You're excellent on the things you can control, and you create optionality in your strategy. That's what it's all about.

Generative AI and machine learning in the NFL

QuHarrison Terry: I do want to get into a few things. The first thing is generative AI. You've got a ton of data that you're collecting at scale. We're now here where consumers can participate in the experience of generative AI. How is the NFL thinking about it, especially in regard to the data that you have?

Paul Ballew: It's early days for us, but whether it's generative AI or what we're doing with ML-type applications, part of it always comes down to what you see generative AI being used for, and that is either natural language processing type solutions (insert call centers, fan engagement, interactions, or so on) or some derivative therein. I would say a large part of our focus, a large part of my team sits there because that's the journey of driving hyper and accurate responsiveness to a consumer.

Having said that, when you think about all the things we're doing in, let's call it, the advanced analytics field, not just generative AI, but you start looking at things like computer visioning and related technologies, that work we've been doing goes all the way from player health and safety support, which we provide support to them and our external partners on that front, to sponsorship effectiveness. We've been doing all sorts of fascinating things to determine the value of impressions and exposures and other things that go along with it.

Where all that takes us is an interesting question because, again, it always anchors back into the things you're trying to resolve. Can I engage with you more meaningfully? And can I provide support to our partner and/or ourselves so what they're doing operationally is having the right outcome?

It's high on our list, like everybody else. The opportunities for more advanced analytics, the opportunities for applying the science now at scale and cost-effectively is in front of us.

It's going to be an interesting couple of years. We've got a whole R&D skunkworks team in my shop that's kicking around applications right now.

How the NFL organizes data at scale

QuHarrison Terry: Lisbeth Shaw from Twitter is wondering, "What are the few ways that you organize that data and think about it at scale?"

Paul Ballew: When we get up in the morning, our organizing construct comes down to think of it as three rivers.

• There is data around the individual (insert fans) but could be businesses as well, so we're treating businesses as an individual entity.
• There's data that's specifically tied to player health and safety, which is a very important area. A very sensitive area, by the way, because you've got to govern it, not only HIPAA, but we have the relationship with the PA that we work very closely on.
• Then you have all the football data, which include things like NGS.

Our governing construct, we think of it as those three big rivers. Then we've got support for the rest of the organization, whether it's HR, finance, or so on. But those are the three big rivers.

We take a similar approach with all three of them in terms of we think of data management as an end-to-end process. It goes from what's your source, what's your ability to digest and curate at scale, what's your ability to integrated and make it available, leverage it, and then what's the continuous process of enhancements.

Then all the enabling things underneath it, whether it's data quality processes or data standardization. Or in the case of fan data, individual data, identity resolution and consent. That's the way we think of it.

I encourage anybody, when you think about data and data management, always anchor back to the business use cases and needs and the logical construct that goes along with it because, usually, what you're faced with is you're trying to bring data and transactional applications where the data was not designed to support your broader use cases. It was designed for a specific transaction.

If you think about a CRM application, that data is specifically there to generate an outbound communication. Now we're saying we're going to use all that data to integrate in to get to this complete view of an individual. That's the right way to do it.

Key data lessons: Repeatability, scalability, and governance

Michael Krigsman: When you were talking earlier about your ecosystem, it seemed that that was also very much implied that everything you think about, you do, you plan is about that scale.

Paul Ballew: Ultimately, for a data and analytics organization, you ask yourself three questions (in my mind): Is it repeatable? Is it scalable? And is it governed?

Then you get into, obviously, the related things of does it add value to the organization and those sorts of things. But if you're going to do this with maximum impact, repeatability matters, and then going right along with repeatability is can I scale it. Then the third piece of this is, because of the scale and the impact we're now having, you have to properly govern it.

Oh, by the way, you go back three-plus decades ago when I started. We didn't have any of those concerns because our data sources were all preordained or established in very structured situations. They were government statistics or the U.S. Census, or they were marketing research surveys.

The questions around scalability or governance were not really questions for us because we didn't have those issues. We were dealing with things that were so structured to begin with.

The U.S. Census has been around since 1790, and there's data from the U.S. Census from 1970. Most economist statistics on the U.S. are out there from 1949 on. And the construct of the data has been preordained for all of us.

Oh, by the way, the governance of it is, "What are you worried about? The data from Bureau of Labor and Statistics has no restrictions in terms of usage." Now, you can misinterpret it, which is a separate issue.

But then we get into the '90s. Suddenly, the world changes on its head because, for the first time, we're able to capture transactional data at scale (or observational data at scale). Then the whole world opens up to us around can we manage that data, can we govern it? Not surprisingly, it was only in that period of time that governance really started to take off both in terms of standardization and quality as well as access controls.

That's a little bit of a history lesson for everybody. But when you think about it, go down that path, and then you go out the next five to ten years, and you say, "What's next?" Well, what's next is the next generation of that is an even faster leap or bound because the ability to make sense of it with limited latency is the next big horizon for all of us.

Michael Krigsman: We have a couple of questions from Twitter, and I'm going to ask you to answer these quite quickly because we're simply going to run out of time. The first one, again from Arsalan Khan, is an interesting one. "Do all teams have the same resources to collect, manage, and share their data?"

Paul Ballew: No. We see variability by all 32, but all 32 are making substantial investments in this area.

Michael Krigsman: Then related to that, what is your actual relationship with the teams? You work for, I was going to say, central HQ, but that's not right. [Laughter]

Paul Ballew: [Laughter]

Michael Krigsman: The teams might have an issue with that choice of—

Paul Ballew: Think of it as a trade organization, so that's probably the right way. But we have a close relationship with all 32 clubs. They are, in essence, our board of directors.

From a data and analytics standpoint, it's an incredibly close working relationship. They leverage us for a set of capabilities and services. They leverage us to help drive consistency. Then within the fan environment, we are the organization that's brought all the data and the analytic capabilities together that all of them consume.

Michael Krigsman: Then another one from Twitter, from Chris Peterson who says, "How do the multiuse stadiums and their data systems (including wi-fi and cellular access points) steer data to the NFL during games?" I think what he's asking is, as the data is flowing around the stadium, how do you guys get it?

Paul Ballew: Yeah, actually, it's the club's responsibility at that point in time. Then there are specific data elements that the club share with us (all visible and transparent to the fan and all properly governed).

Michael Krigsman: Again, that governance is a crucial aspect.

Paul Ballew: Governance. Governance. Governance. If anybody is trying to set up a data organization from scratch, I encourage you start with data governance.

Collecting data from the Super Bowl halftime show

QuHarrison Terry: Love it. Paul, I have to do this. You probably are going to expect it. You thought you were in the green and you were going to get away unscathed, but we're going to ask you to tell a data story. And that data story is going to be about one big moment that happened this year, and that was the fact that you had Rhianna as the halftime performer in the Super Bowl. The floor is yours. I'm just looking for a data story on that one.

Michael Krigsman: [Laughter]

Paul Ballew: Well, it's interesting because halftime shows historically have given us pop in terms of viewership. We track that very closely, and we track how it's giving us pop.

We haven't gotten all the data in yet, but she did a great job, and we got substantial lift. You saw the 200 million viewers and the tracking that we've released recently. We'll soon see all of that.

As we saw last year, we had about 26 million additional viewers, we estimate, that came in for the halftime show, on top of 208 million viewers overall, so we look at all of that. We share it with the companies that are working with us and talk about and see what insights we have.

We spent a lot of time looking at what else we can do to drive more viewership and more energy and more positivity. It's all getting examined, reviewed top to bottom. But this year's Super Bowl was an A+ and the data is confirming that it was an A+, just like last year was an A+. We had an exciting game and it was just fantastic.

QuHarrison Terry: You had weakened the previous year, and then you had Rhianna the following year. The question I have there is, there was also COVID. Did COVID elevate your data collection abilities? It seems like, post-COVID, you've just nailed it.

Paul Ballew: I would say post-COVID and everything going on with the league prompted us to make some additional investments, and data and analytics is one of them. If you look at the last couple of years (coming out of COVID), we did a lot of data collection in COVID because we had to, to maintain health as well as keep the game going.

That was important for us, but COVID (as well as other things) are probably a forcing factor for us to put more emphasis in the area that I'm fortunate enough to be responsible for. But there are other things going on as well.

The fragmentation of media and how content is being consumed, those are really, really big forcing functions for us as well. To the credit of the commissioner and the senior officers of the league, they are being forward-leaning, and there's a heightened state of "let's be prepared for the future and not just rest on our current success."

Future of data, analytics, and converged content in the NFL

Michael Krigsman: As we finish up, can you talk to us a little bit about where this future is going? I'm particularly interested in this (to use an old term) multimedia environment that we live in. Maybe talk to us about any aspects of that.

Paul Ballew: It's exciting. As you and I have talked about before, my day-to-day job is incredibly eclectic. Every day is interesting because there's something new and exciting on player health. There's a lot going on in officiating. And of course, all the fan work.

But an area that really piques our interest is the current and future state of how content is being consumed. Why that's important for us is we want to understand this better. We want to understand this better so we can share it with our partners and help them activate their relationship with us better.

All the partners and sponsors, we want to share it to make sure that the things we're investing in and putting emphasis on are driving. Think about fantasy football, for instance, or all the things we're doing on social media.

Then it also gives us the opportunity to be out in front of where things may pivot and try to get as much of an early warning on that as possible. Early days, but very exciting for us.

We feel very fortunate that we've been able to crack that code a bit. And in cracking that code, we're spending a lot of time with ourselves and the clubs figuring out what we can do with it.

Again, the big question is, can you bring it to life? It's great to have the science. The science, when you're in a business setting, has to add value.

QuHarrison Terry: Paul, one thing before we go. I want to thank you for leveraging all of the best technology from across the world, really, and bringing that to create a cohesive experience.

One example that I've seen, just in my few years as a fan, is the NFL is one of the few places where you can get the best out of Amazon, Google, Apple, and many more (way too many for me to name), and that's not true anywhere else. I do like that about the NFL (as a techie).

Paul Ballew: It is quite a blessing. Our partners like working with us. And as we like to say, we like working with them.

It's been an enjoyable part of my role coming in because of the technology side of my job. The external technology partners are great to work with.

Michael Krigsman: A quick question, under the wire before we go, from Twitter. "Do some people consider governance as just another form of red tape?"

Paul Ballew: One of the favorite stories I like to say about governance is my first leader of data governance, which is almost 30 years ago. When I put her in the role, she used to tell me these stories.

"I probably spend 100% of my time explaining to people that I'm not the police." Really what she was about was to make sure we had standards and consistency and we're using data appropriately, which actually opens doors for people to leverage data.

It's the exact opposite so, somewhere down the road, maybe we'll change the language from governance because it does sound like it's some type of a police action versus it's an enablement action. If you do it right, it's all about enablement.

Michael Krigsman: On that note, unfortunately, we're out of time. I want to say a huge thank you to Paul Ballew from the National Football League. Thank you so much, Paul, for taking your time to be with us today.

Paul Ballew: You're welcome, Michael. Always great to see you.

Michael Krigsman: And I also want to say a huge thank you to QuHarrison Terry for being the co-host today. Qu, it was a really interesting discussion.

QuHarrison Terry: It was, and I'm always excited to be here. Thank you for having me. Paul, it's been a pleasure.

Michael Krigsman: Everybody, thank you for watching, especially those folks who ask such excellent questions. You guys are such a great audience. You guys are very smart, intelligent, sophisticated, so always ask those questions.

Now before you go – this is important – be sure to subscribe to our YouTube channel and hit the subscribe button at the top of our website so we can send you our newsletter and notify you of upcoming live shows. Thank you so much, everybody. I hope you have a great day, and we'll see you soon.