No Easy Button: Horizon Report Insights on Data Literacy and Governance

Sophie White: Hi everyone. Today's Shop Talk is a companion to the 2025 EDUCAUSE Horizon Report Data and Analytics Edition. In this one we talk with researchers and panelists, Nicole Muscanell, Kim Arnold, and Todd Barber about the state of data and analytics and higher education and how to use the report to inform your work. So we talk about how higher ed is still struggling with some of the foundations around data governance and data literacy, and really how AI is the overlaying factor that affects all of our work around data and analytics, some of the opportunities that come with AI and also some of the pitfalls. And then we kind of contextualize this in the larger trends of social, technological, economic, environmental, and policy areas that we're looking at across the world and how this affects higher education and data analytics professionals. So check out the episode and then make sure to read the entire EDUCAUSE Horizon Report on data and analytics to get the full picture.

***

Sophie White: Hi everyone and welcome to EDUCAUSE Shop Talk. I'm really excited that we'll be talking about the 2025 data and analytics Horizon Report that EDUCAUSE is publishing today. So I am Sophie White. I'm a content marketing and program manager with EDUCAUSE, and I am one of the hosts for today's discussion.

Nicole Muscanell: My name's Nicole Muscanell, I'm a co-host today with Sophie, and I'm a researcher on the research and insights team at EDUCAUSE. And I am also one of the researchers and co-authors of the Horizon Report.

Sophie White: Beautiful. Thanks Nicole. So I'm just going to introduce our guests today and then we'll jump into it. So first up we have Todd Barber. Todd is the executive director of enterprise applications and data services at the University of Tennessee Health Science Center. He manages the teams responsible for data warehousing, data architecture, data integrations, data governance, and custom web development. Todd has over fifteen years of experience as an adjunct faculty member at three institutions. He is a certified data management professional and also the primary co-lead of the EDUCAUSE Data Governance Community group. Thanks Todd for being here and for all you do. I know you do a lot for our data community.

Todd Barber: Thanks for having me.

Sophie White: And we just found out Todd and his daughter actually have a podcast about music that we have a celebrity with us today in the podcast world.

Todd Barber: NACHA

Sophie White: And Kim Arnold, our second guest today, Kim is the director of the teaching and learning program at EDUCAUSE. She has twenty years of experience in higher education with deep roots in learning analytics, data governance, ethics and privacy, student learning assessment, theory and design, and student digital ecosystems. So thanks so much, Kim, for joining us.

Kim Arnold: Yeah, so excited to be here.

Sophie White: Great. So yeah, let's jump into it. I'm really excited about this data and analytics Horizon Report data has been top of mind in so many conversations at EDUCAUSE. We'll be releasing the 2026 EDUCAUSE top 10 at our annual conference in Nashville. But last year the top 10 for the 2025 had the data empowered institution as issue number one. It is an issue that a lot of institutions are thinking about right now, especially with the proliferation of AI. So I'm excited to chat about the panel's findings and the report today. So I'm curious, all of you worked on the report, are there any elements that really stood out to you as interesting findings, things that surprised you about this year's data and analytics Horizon Report?

Todd Barber: So I'll go first just being part of the data governance community group with EDUCAUSE and hearing a lot of institutions talk about data governance and the struggles. It's nice to see that the panel says yes, that is a struggle, but it's also negative in the sense that we still see that the need for data governance is still there. And as a industry, why hasn't higher education as a whole really taken hold of that more so both positive to see that it's still needed but negative in the sense of why is it still there?

Kim Arnold: Yeah, I think participation on the panel can always be a really two sides of a coin, right? Really validating on one side, Todd, like you said. And then on the other kind of like, ah, in a similar vein I was going to talk a little bit about it was one of the trends that we look at when we're looking at signals as the panel is trying to identify signals. One of the trends that we are seeing is this persistent lack of data literacy. And I think that's broadly, that's outside of higher ed, that's not a higher ed only issue. That's a societal issue that we're facing. And that also then led to one of the key techs and practices being institutions really doubling down and focusing on building capacity around data literacies. And this one really stood out to me as a longtime kind of data and analytics professional, but also in my role here on the teaching and learning side, we saw this exact same thing show up in the last two teaching and learning Horizon Reports.

So this really strong thread through a need for data literacy. So kind of in the same vein, often driven by data governance. This is persistent and it has a lot of staying power powder, did I say staying powder, staying power. And so this isn't something that's fleeting, it's not a shiny thing that we're just chasing momentarily. This is something that as an industry, we're not seeing sufficient progress across the board. And I think this is where we see this just pop up time and time again and where it does really need to be a strategic priority. So that's one of the things that was one of the initial findings that just really jumped out at me. Hey, this showed up in another Horizon Report.

Nicole Muscanell: That's a great point for both of you of data governance and I think literacy were probably the two biggest themes that cut across the different sections of the report. And so I'm wondering if you two have thoughts on why these issues are so persistent. Why is data governance just this ongoing problem that seems to never go away? Why is data literacy such a massive problem? And Kim, to your point, you're saying even beyond higher ed, why is this such a huge thing that we need to tackle?

Todd Barber: Yeah, I mean ultimately it's hard. It's difficult. It's not very easy. And I think institutions are struggling with that. I think a lot of people, this goes to the data literacy side of things. People look at it and say, that should be easy. Well, it's not. And so there needs to be some outlay of resources, whether that's money, whether that's people, whatever it may be. But I don't think there's enough resources yet that are dedicated to it across the industry. There's always going to be institutions that are, okay, let's point to them. They're doing a wonderful job and how can we model what they've done, whether it be governance, AI, literacy, whatever it is. But it's not as easy as what a lot of people think it is. And so I think that's one of the things that does hold it back is you say, okay, yeah, we need this. And then institution sees how much it may be and how much effort it may be, and so they start to pull back with whatever that is.

Kim Arnold: Yeah, I mean, Todd just hit it on the head. There's no easy button for this. I wish there was, but I think at the root, this comes down to cultural change management and that is something that spans a very wide kind of breadth of stuff. And it's interesting to think right now in terms of data literacy specifically, I don't want to continue to just talk about data literacy, but it's really interesting in that we're getting students coming into the higher ed environment now that are using AI on a regular basis. And so they are getting some level of literacy these students before they come up and graduate into our systems. And there are questions there. And I think AI is a huge, huge driver of this. I think you've seen in a lot of EDUCAUSE reports lately that the majority of students do use some type of AI in high school, even in elementary school before they come up into higher ed.

They are using AI. And I think this is an age old, I'm dating myself here, but this is the age old thing that just because students are using tools doesn't mean that they are data literate about it. So they might be using it, they're using it pervasively, but do they understand what's true and not about things that come out of AI? Do they understand privacy concerns about using any random tool that kind of pops up in their ether? And so I think we don't want to let up on that gas and even though we're saying, well, students are using it earlier, they're coming to us, they're using it in every part of their life, their personal life, their academic life, that doesn't mean that we can assume that they're fully data literate and are engaging critically with the tools that they're using. And so for me, that just means we have so many layers of literacy to consider.

And when we think about the whole slew of digital tools that supports higher education across the spectrum, that's a lot of different roles and a lot of different people to get to think about the data and how you use it, whether it's AI or an analytics capacity. And it's something we're asking people to do that we already know people are feeling overwhelmed and their plates are really full. And we're like, this is something new that you need to learn, that you need to do is just vitally important. And so I think at the heart of it, Todd summed it up, it's really hard. There's so many layers and it's really difficult to ask people to continue to do more and more, even if they know it's very important.

Todd Barber: And you mentioned the students primarily, but faculty, staff are using AI as well, but in different ways. And so literacy training, literacy concept may look even different for those sets of people. And so now if you've got somebody that's doing data literacy training for an institution, now you've just overwhelmed and put more on their plate as well because oh, I've got to give one message potentially to students. And you try to generalize it as much as possible. But oftentimes there are specifics to each of those groups that have to be aware of and portray to them.

Nicole Muscanell: And I think, oh, sorry, go ahead.

Sophie White: No, you can go, Nicole.

Nicole Muscanell: I just wanted to call out a good point that came up as part of that discussion. And I think something that is really the thing that's making it hard or difficult for institutions is we're talking about changing the culture in such a way that we're not only teaching people what is data and how do you interpret it, but a bunch of other things in terms of how to use data responsibly, ethically. And so I think maybe it's because it's a larger cultural change. It's not about just here's a training program. That's probably why it's so difficult, because it's really multifaceted.

Sophie White: And I know a lot of our AI conversations on this show recently have come down to AI literacy, and it's so hard to think about, oh, telling people that they have to do this AI literacy work, understand the ethics behind it, the biases, how to evaluate the outputs, and now also do data literacy on top of it. So I also thought it was an interesting, there's a section in the Horizon Report, it was the key technologies and practices section that told me that users don't even necessarily understand their own pitfalls, that they may be overly confident in their data skills, which I thought was interesting because you can say, oh, you need to do data literacy, but someone might say, oh, I know how to use data already. So how do you make that cultural shift of, oh, well, things are changing actually you do need to do data literacy where it's different than it used to be. It's complex, as you've said, Todd.

Todd Barber: Yeah. And I would say data literacy is bigger than just reading our graph. And so it's knowing and whether right or wrong, I consider myself data literate, but I know my limitations. And so if I'm going over into something that's heavily researched where a lot of the math and statistical side of and data science side, okay, I'm going to back out of that in the data literate conversation and let that be somebody that truly is literate in all the formulas and how to use that. And so I think that's where that overconfidence comes in is people saying, okay, I can read a graph, which is mainly what you think of or a lot of people may think of, but there's so much deeper, which again, makes it harder. And going into that. And I think the best thing, one of the best things I think that people in the data and analytics profession can do is be okay backing out of some of those data conversations and saying, okay, this isn't my area of expertise within the data world. I'm going to let Kim talk about this or Nicole talk about this and be able to say, okay, you all are the expert. Let me hear what you have to say and then be willing to learn from them.

Sophie White: Yeah, humility is so important. It's a great point. It feels tough, and let me know if you all think I'm interpreting this correctly, but some of the Horizon Report also talked about this larger landscape of how the work of data professionals in higher ed fits into our social, economic, technological, political trends and the fact that we're in a space where higher education's looking at funding challenges, also all these policy changes that may require more transparency in reporting. So data and analytics professionals are getting stretched thinner to both do this data literacy training and also fulfill all these reporting and compliance requirements. So I don't know if you all have thoughts about how do we address this challenge for making sure professionals have the bandwidth to take care of themselves and do the important work in this current landscape?

Todd Barber: Wow, Kim.

Kim Arnold: To say it's like Sophie always asks the softball questions, right? No, no. Well, and I think this is exactly it. I think data in higher education is fundamentally changing based on federal policy. So there's a lot of things for data and analytics professionals to really take a step back and think in any institution that's really leveraging data. And let's be honest, there's very few institutions you're going to find that don't have some type of analytics system in place that they're trying to use data, trying to be more data-driven, trying to show value. So I think most people are using this. And when we're looking at some of the basic foundations of educational data that have been around for decades, when the structure of that data begins to change based on policy, that creates an issue. So it's not taking pressure off Sophie, like you ask, right?

But it's adding that additional really complex layer of how do we look at things longitudinally? Anything AI, any models we're building, we're looking at data that's trained from the past, we're looking at longitudinal data sets, and when data just stops being available or the structure of it changes, there are things for us to consider in terms of how accurate our analysis can be. These reports that so many institutions are relying on things that we report up to accreditation boards, anything that to national data sets, it all has the potential to kind of shift right now. And so these are areas that will be a challenge and we'll have to come together as an industry and figure out how are we going to address these. Anybody in the data world knows that there's always blips in data. Data's never perfect. There's always things that happen. And I think this is really adding to some of the trend data that we're seeing where there is a lot of uncertainty in how to proceed in some areas.

And that's just our reality in higher ed right now. And I think it comes down to focusing on what we can control and what we can move forward. And that is often coming together with our peers and saying, can we form some kind of partnership? Can we figure out how to do some little slice of something to keep moving things forward? Because the silver lining in all of this is that we still have more data today than we did 10 years ago. And so it's figuring out how to harness that and data and analytics professionals are going to be at the center and the core of that. And so there's huge things to look forward to in terms of efficiencies and institutional processes, and there's progress. All you have to do is look at the exemplars, right? Look at the exemplars in the Horizon Report. And there are a lot of examples of people making headway even if it's a little bit at a time. So I think kind of trying to shift our narrative and our own institutions and say that little baby steps are going to ultimately kind of get you to that optimized efficient future that we all hope to get to. And these are all just steps, steps on the journey. So I think reframing that is one way to possibly remove a little bit of anxiety and stress from a lot of stuff getting piled on top of data and analytics professionals right now.

Todd Barber: And Kim, you mentioned the longitudinal side of it, of all this data that we've got and trying to do trends and things like that and realizing that all of a sudden due to policy shifts, we're missing data and we all recognize that institutions are scrambling to say, okay, I don't need to send this data here anymore. It doesn't even need to be available to these people anymore, but yet I still can't get rid of it in our system of record. And then all of a sudden politics and policies are what they are in four years, in eight years, it may shift back to what it was. And so now we have to be able to, I think, become, as data and analytics professionals become more agile in how we're gathering and distributing data because it is going to start to shift, and we have to be able to do that fairly quickly.

Kim Arnold: And I don't want to monopolize, I just One second, Nicole. Sorry. No, go ahead. One of the really, really interesting, Todd, as you were talking this spring to mind, one of our KTPs, so key texts and practices is not relying so heavily just on quantitative data. So this is something obviously reporting up for federal standards and things like that is always quantitative data. But locally, one of the KTPs that got voted up was using qualitative data, so using mixed methods approaches specifically that one was specifically around student learning data. But I think that's a really interesting way to help us reframe too, that there might be some quantitative data shifts, but we still have some type of data to work with there. But we can supplement and scaffold that with some really deep meaningful qualitative data. And I will tell you that I recently just, I'm a quantitative data scientist by training, but I recently just went back to my grad school roots and did some qualitative data analysis and it is stretching a very different muscle and brain in my space. But it's super interesting to say, okay, there's a shift and maybe the quantitative side of things, so maybe we do pull in more qualitative stuff when we approach things in a more mixed method.

We just approach it in a kind of different way. So another way to kind of reframe some of what we're thinking. Sorry, Nicole, I wanted to get that out. I thought it followed on what Todd was saying.

Todd Barber: And I'll even tack on more to that to just see how the key technologies and practices from the Horizon Report, how they're all meshed together. Because you talk about the mixed methods and the qualitative analysis, well, that's become so much easier for higher ed institutions because of AI and the proliferation of AI. So you've got AI powered assistance and decision intelligence in there. So all these things are mixing together with the governance and data literacy to have the good foundation being able to make good decisions from a qualitative analysis. And so it was nice to see as somebody that was going through it, seeing all the information that was thrown together when we first started and how these have bubbled up and how well they do mesh and meld together to push data and analytics further in the future.

Nicole Muscanell: Todd, you kind of hit the nail on the head with what I was going to add. I think in an ideal world, we institutions would just have enough resources where they could be like, we're going to bring on all these data and analytics experts to deal with these compliance requirements, and we just have all the resources to be able to do that. But obviously most institutions are not in that position. And so I think it does come back to the governance and the literacy that's probably just going to keep coming up in our conversation. I think if we can start to see a lot of movements in maturing governments and data literacy across the institution that might help alleviate some of the pressures that institutional research teams are having to face and the individuals who are really responsible for ensuring that, Hey, we have to keep up with these changing regulations. I think strengthening literacy and then with AI, as we're seeing that data and analytics tools are becoming more accessible to stakeholders across the institution, those two coupled together could really pave the way for taking some of that pressure and strain off of the individuals responsible for keeping up with those requirements.

Todd Barber: And I think one of the analogies that is popular in our community group for data governance is the foundation of a house because everybody knows what that looks like. And so as you hear people start to say, oh, well my number is different from your number, or Why is AI giving me bad answers? Or whatever the problem or issue is at the end, it goes back to something's wrong. There's a crack in your foundation, and that's a governance issue. So if your governance isn't there or just starting, give it some time or put some resources and effort to it to booster it so that it can be the foundation that all these other initiatives need.

Sophie White: What should we dive into next? I want to make sure we talk about all the exciting horizons.

Kim Arnold: Yeah, I was going to say, I'm kind of surprised I'm not, because knowing Todd and I are the guest, we like the data governance and literacy, but two of the really big KPTs that came out were AI related. So AI assistance and AI decision making support. And I think we can kind of segue into that because there's some interesting connections here and how do we equip our institutions to deal with this influx of so much data while we're thinking always about governance and literacy, that's always, as Todd said, kind of our foundation. There's a lot of really interesting things, and for me, one of the really interesting ones was looking at AI assistance and just looking at how as tools and practice, these are things that can really create some efficiencies within higher education and really help fill some gaps. And for AI assistance specifically in looking at the exemplars, I mean looking at the exemplars are always one of my most favorite thing because we talk about trends, we look forward, we look about what's coming in the future, but the exemplars really show how institutions are actualizing this right now.

It's happening. This isn't just something ethereal that we're hoping to get to. We know this is happening right now. And I think the AI assistance KTP section particularly was really interesting to me because of how people are using AI, whether it's creating agents, creating chatbots, just ways to really fill gaps that they have. So an example that really sticks out in my mind, one of the exemplars was San Jose State and Quality Matters partner, and they created an AI assistant that basically serves as a frontline instructional designer, and it focuses a ton on insights and pedagogical best practices and provides a lot of really meaningful recommendations. And that sounds cool. Yeah. Okay. It's a way to save some things because across the industry there's a critical shortage of instructional designers. So we have overworked instructional designers, especially often thinking about accessibility. People are called in a lot to do that, and we don't have enough instructional designers, so how can we use things like AI assistance to help us bridge that gap and take some of the load off the instructional designers.

So folks who are looking for very entry-level basic instructional design help can go to an agent like this, get a lot of really good recommendations based on a lot of theory and best practices, and then implement those recommendations. And then if they want some more advanced stuff, then they go to the human, but it frees up those instructional designers who are exceptionally overworked. And so that section, particularly for me was just really interesting. In light of everything we've talked about earlier this morning, things are hard. People are overwhelmed. We're having more and more added to our pile of list. So being able to leverage responsibly of course, and within data governance frameworks, AI assistance in a way that can really help grow efficiency is a way to really optimize what we're doing and in a way help with the mental wellbeing of the people who work and support our students.

Todd Barber: Yeah, I agree. And I think with chatbots and professionalized tools, I've got some technical people on staff that use AI to help with generating code, and then, okay, yeah, I can manipulate it a little bit to further fine-tune it for what I need. But even as simple as they are, and they've been around for some time, service desk chatbots are a very easy way to shift some of the burden, especially of those easy questions away from your frontline IT people and allow them to do more meaningful deeper work, maybe go scour through the knowledge base to update those knowledge base because that's what feeds the answers to the chat bot itself. So it's a way to allow maybe some people in those particular areas, whether it be instructional design or a service desk person that says, okay, I'm tired of clicking a button to reset a password.

I want to do something more meaningful. Well, let's let the AI do the menial tasks, the ones that are easily trained, that we can almost turn loose without much human interaction. Because that was one of the things that maybe one of the trends was we don't know where that line is to turn the AI loose versus when do we still need that human intervention yet? Well, there's some that we know we can turn the AI loose for the most part and let's let them go so that now we can let those people that were doing that job do something a little bit deeper.

Nicole Muscanell: Yeah, I love that you brought up human oversight. That was one of our, I think, social trends. And I'm just curious for you two, do you see AI assistance playing an increasing role in data driven decision making? So doing more of the work of here's the data, let me help you figure out how to analyze and interpret it, and then here's some decisions that you might want to follow based on this data.

Todd Barber: Yeah, I do think that's going to happen, and it's already happening. I know we are a Microsoft shop, so we've got Fabric Copilot, and you can get a graph, get a dashboard on the screen and start to ask copilot questions of, well, what should I do here if you see this and it gives you answers. And I was in a meeting with some Microsoft people and told 'em, yeah, the first time I tried this, it was awful. It did not give me a good answer. But sure enough, it is learning and the answers are getting better. So due to your point, Nicole, yeah, there has to be some of that human oversight as we get going, data literacy. Hey, just remember when you ask Copilot a question about this dashboard, take it, but then go verify what it tells you to do because it may still be getting some things wrong. But yeah, it's definitely happening and happening quickly where people will start to make decisions based off what Copilot comes back as. And again, that goes back to that data literacy training. Again, don't always trust that answer.

Sophie White: Yeah. I love the examples that you shared related to augmenting the human workforce too. I know our EDUCAUSE workforce reports that came out this year had a lot related to folks in IT roles feeling overworked and stressed. So I feel like we've talked a bit about AI and how in some cases it is replacing especially entry level type of roles. But I also love the framing of when used well, it can really augment what human staff is doing in order to create a better learning experience, a better institutional research experience, all of these things. So intentional integration of AI is kind of what I'm hearing from this conversation

Todd Barber: For sure.

Sophie White: Kim, what was the other key technology and practice that you mentioned? We had AI powered assistance.

Kim Arnold: So we had AI powered assistance, and then AI powered decision intelligence, which Nicole kind of segued us a little bit into. And then there are two additional KTPs that build on everything. And this is what we've mentioned. They're all so interconnected in this report, unlike some other reports, it is data mesh architecture and federated data governance we're the other two KTPs. And I think these are examples. Again, you've heard Todd say mesh a couple times during the call today, and it's kind of a new, if we're really going to optimize our AI agents, our analytics spaces data, good quality data has to be discoverable and accessible. And so this is what data mesh and federated data governance really allow us to do. And as we're seeing more and more data come into our ecosystems, the federated data governance model is becoming very, very important so that you can contextualize data locally and autonomously own that without moving towards one data store to rule them all like a data lake or a data warehouse.

So those were the other KTPs that came up as kind of a supporting way for everything else that we've talked about. If we're going to be optimizing what we're doing at institutions, whether that's admissions, student support, teaching and teaching and learning, even parking, there was some cool examples around using decision intelligence to monitor parking at different events or mobility aids like scooters, etc. So if we're going to really be harnessing all of that foundational layer again of data mesh and federated data governance is going to become something more and more important. And for me, this was interesting that it came out. This is something that we're looking when we're forecasting to the future, many possible futures, these are things that likely will come up in there, and these are kps that will really help us kind of move towards those signals that we're seeing.

But in higher ed, we may be less mature in these spaces than other industries. And I think that is pretty clear in looking at exemplars that we had submitted. We know this is important, we're aspiring, many people are starting working in this area, but it hasn't been fully actualized at very many higher ed institutions. But again, baby steps along the way. So lots of proofs of concepts and initial kind of pilots in this space, some moving certain segments into operations kind of firmly, but there's still a way for us to go into maturity cycle for I think both data mesh and federated data governance.

Todd Barber: Yeah, go ahead.

Nicole Muscanell: I was wondering maybe it would be helpful to provide just a little bit more context on a movement toward those types of models for people who are watching this,

Todd Barber: A movement toward the data mesh or federated.

Kim Arnold: Todd, do you want to talk about maybe the federated data governance? Because I think we're more mature there than we are in the mesh space, but do you want to talk about that shift?

Todd Barber: Yeah, sure. I think data governance goes back to data literacy really as well. What I was saying there was that you've got to realize for governance to be effective, you have to make sure that you're including the people that know the areas that they are "in charge of" your data stewards, data owners, trustees, whatever your institution calls that. And so I think if you're able to say, okay, here's a centralized set of definitions and here's a centralized set of processes and policies around data governance, you're now able to say, okay, the registrar as a data steward can go off and in many cases do their own thing being that they are adhering to these centralized set of policies and definitions and colleges can do that. And so in the past it's been everybody doing their own thing. And so that's why you get all these various different definitions and why a department inside of a college can have one number. The college can have another number, and the institution can have yet another number. And that's not including what has been sent to IPEDS. And I'll stop there, but once you have that centralized to be able to do this hybrid federated modeling, you are able to allow people to do their own thing because they're operating within the larger model of governance for the institution. And you don't necessarily have at a central level, you don't necessarily have to govern all of the data, but you do need to govern the data that is being used institutional wide. If there's a piece of data or a set of data that is only being used in the college of business, let the College of Business govern that and let them do their thing within the boundaries and sets that a centralized model is set up for 'em, and it makes it way more efficient. It lets people move faster in the areas that they need to move faster without the bogged down of a central governance model that has to get all these various things. And now you have to, again, relieving the stress of people. You're not adding to a stress of somebody saying, oh, well, I've got all these various policy shifts that are happening, but yet I need to go define something that only three people are going to use. So let them do that. And I think that's where this federated data governance is starting and can really, really help.

Kim Arnold: And I think it's emerging from the reality of the world we live in. So it used to be, if you go back twenty years, most technology and data was controlled by individual folks, right. Now, as people are able to create their own AI as they're able to choose different tools to use all of that, just the sheer proliferation of data that's available makes it really, really difficult to do this in an old or the classic kind of centralized way to manage this. And so, okay, we know if there's going to be a department or a domain that has 20 tools that are only relevant to that domain, then they are accountable for managing that and making it discoverable to folks who might need it, even though it might be a small amount. But again, it takes, this distributed approach is just exceptionally efficient when we figure out how to leverage it kind of accurately. And so that's just, I think you do see the emergence of it because there are so many more tools. There are so many data sources that are kind of coming into our environment. It just gets kind of wild, real fast.

Todd Barber: And I would say though, for institutions and if there's institutional leaders that happen to listen federated data governance doesn't mean there's no centralized data governance. If you don't have a centralized data governance program already at your institution, by all means start there because that's where you're going to get the most value for your governance program. But you may realize as that starts to go that you need to do some different things within the federated world. And that's when you can start to say, okay, we can make this part federated, but if you've got nothing, don't start with federated data governance because all you're doing there is willfully, knowingly creating silos that you're not really solving anything. You're still going to have different numbers at the end of the day.

Sophie White: Thanks for putting that in context. So if data governance is the foundation, then the federated element is kind of built on top of that as an upgrade, we could say. Is that a fair assessment?

Todd Barber: Yeah, I think it makes federated Data governance makes your centralized model more efficient because it takes away the decisions that are only in one small niche section and allows them to do that. Because I've set up the framework, I've got back to our house analogy, I've already got the framework there. So yeah, you can go build that room over there however you want to because you're building it on top of the foundation that I've already set.

Sophie White: Okay. That's really helpful for contextualizing it a bit for folks who don't work in the weeds of data quite as much.

Todd Barber: It's a lovely place to be.

Sophie White: It is. It's a lot of fun. So we're getting to the end of our time and just thinking about the entire data and analytics Horizon Report, I'm curious, do you all have maybe one actionable thing that someone who reads the report should start implementing or doing differently as a result of reading the report at their institution? What do you think a good starting point could be? And Nicole, feel free to answer this if you have ideas too.

Nicole Muscanell: Sure, sure. I mean, I can start it sort us off. One of the points I would kind of suggest or give to people is about the report itself. Just remember what the purpose of the Horizon Report is. It's built upon strategic foresight. And so we have a lot of things like, here's some trends that are either happening or we're going to come down the pipe. Sometimes people think it's like a prediction tool, but that's not really what it is. So I think a big point or a takeaway is jump in the report, get a sense of what the trends, emerging trends are, get a sense of the key technologies and practices, which are basically levers that you can pull to kind of change to what extent the trends are going to impact your institution. But it's not a prediction tool. So when you're looking at this report and you're trying to use it for strategic planning or conversations with others at your institution, think of it more about a tool that will help you anticipate multiple scenarios or multiple possible futures where any combination of these things could happen. It might simultaneously impact your institution. I think that would be my main advice for just making the most use of the report itself.

Sophie White: That's great context. Thank you.

Kim Arnold: I mean, no easy button from Todd and no crystal Ball from Nicole. Come on. These are no, my kind of two cents with anything Horizon Report related to anything really is just find a way for your context, for your role that it makes sense to do something and be intentional. But doing something small can be very impactful in the long run, and it can really snowball. So be pragmatic about what you do, but look over the KPTs, look at the exemplars, and I guarantee you, you will find inspiration to do something, but start with little steps and that will pay big dividends. That's my overall thoughts there.

Todd Barber: Beautiful.

Sophie White: Thanks, Kim.

Todd Barber: Yeah, and I would be remiss if I didn't say something about the governance being part of the governance community group, but realize that there are a lot of institutions that are already doing data governance in various levels of maturity. And so find those groups. We've got a great group. We meet all the time, we email, you can ask questions, find me at the conference. If you're going to be there, I'll be there. We got a poster, we'd love to hear from you. Have you join the group because that's where a lot of this starts. You've got to have that governance foundation. And so you can start with baby steps like Kim mentioned. It doesn't have to be overwhelming. If you take a step back and you look at this report and you're like, oh my goodness, they're talking about federated data governance and all these things and we can't do that. Yeah, you can. You just have to do it a baby step at a time. And I think the important part for me would be that I would tell anybody is take action. Start. You have to start, if you keep thinking about it paralysis by analysis, you just have to start somewhere with whatever baby step it is. Your baby steps start to snowball and before you know you're running.

Sophie White: Beautiful, thank you all. I think this is a great way to wrap it up. So call to action, read the report. It will not be your crystal ball, but it can inform you as you make decisions. Start small and do something so that you're not running into this analysis paralysis. And thanks again for mentioning the community group. Todd, we talked about community and the power of the higher ed community a lot during this conversation. So EDUCAUSE Connect has a great collection of community groups with leaders from across the world who are really involved in higher education and technology. Todd supports the data governance community group. You can join that to get more involved. And I think that's it. So thank you all again for all of your work and time to create this really fantastic report. And thanks again for talking with us today.