The Human Side of Data at Any Scale

Sophie White: Hi everyone. This Shop Talk is related to our EDUCAUSE Showcase on The Strategic Era of Data and Analytics. And in this one, Jenay and I talk with two leaders from very different institutions about the ways that their institutions differ in their data strategies, everything from data governance foundations to data literacy training, but also how they have a lot of similarities as well as they're serving the mission of higher education. So we talk to guests from Western Governors University and California Institute of the Arts about how to build data governance foundations to help inform strategic decision making about how to train leaders to use data using a problem solving approach and really how to consider the human element in all of this, especially as we think about the intersection of data and AI. Hope you enjoy.

***

Sophie White: Hello everyone and welcome to EDUCAUSE Shop Talk. I'm Sophie White. I'm a content marketing and program manager here at EDUCAUSE.

Jenay Robert: And I'm Jenay Robert. I am a senior researcher at EDUCAUSE, and I'll be co-hosting with Sophie today.

Sophie White: Great. We're really excited today to be talking about the strategic era of data and analytics. This podcast discussion is part of our EDUCAUSE Showcase series where we're diving into this issue in detail. And I'm thrilled that with us today, we have two special guests, Nadeem Sayed and Shannon Shank. So I will introduce them and then we'll jump into it. Nadeem Sayed is Chief Financial Officer at Western Governors University, where he leads financial strategy for one of the largest nonprofit, fully online universities in the United States. He focuses on aligning financial operations with WGU's competency-based technology enabled model to keep tuition affordable while expanding access and innovation for more than 230,000 graduates. Before joining WGU, he spent more than a decade at AT&T in senior finance and strategy roles, spanning analytics, pricing, media, and new digital platforms. Thanks for being with us, Nadeem.

Nadeem Syed: Thank you. Thanks for having me. Looking forward to the conversation.

Sophie White: Same here. Next up, we have Shannon Shank. Shannon is a strategic IT leader with thirty years of experience in higher education, currently serving as the AVP for enterprise applications at California Institute of the Arts. She specializes in leading large scale system initiatives and co-leads institutional data governance to drive innovation and operational excellence. Known for approaching challenges through a different set of lenses, Shannon helps find clarity and strategic alignment within evolving digital environments. Thanks, Shannon, for joining us.

Shannon Shank: Thanks for having me.

Sophie White: Great. I love that part about adding a different set of lenses to challenges too. I feel like that's a really great skill for an IT leader and anyone dealing with data to have. I'm curious, can you talk a little bit, what lenses do you like to add to the work that you're doing at the Institute of Arts?

Shannon Shank: Well, it comes a little bit from some consulting work that I've done in my career, and I try to walk into an organization, mostly a higher ed institution, with a different set of lenses than they're used to looking at. And so it helps to take a really high view and sometimes a different view than what people are used to thinking about and approaching their business processes or even technologies that they're using. So that's sort of where the different set of lenses comes from. And I've been able to do that a little bit in my work at CalArts as well.

Sophie White: Yeah, that's great. I think with this topic especially, I feel like adding those different perspectives of maybe having the industry insight as well as the higher ed expertise can be so helpful. And Nadeem, I noticed that in your bio too, you spent a lot of time at AT&T. I'm curious, how do you think that's informed your work that you're doing at Western Governors now?

 

Nadeem Syed: Yeah, I think the experience, number one, being new to higher ed, let's start there. It does force you to what I call reasoning form first principles. So you're not so beholden to how things were done before. You restick about it starting from the first principles. And the other way it's really helpful is having seen an institution and corporation, the scale of AT&T, you bring a lens on how to scale things and how do you take things and continue to scale them as the high end institution continues to grow. I would say those two are the ingredients there.

Sophie White: Absolutely. I thought the 230,000 learners is a really impressive metric. So that is definitely learning at scale.

Nadeem Syed: Yeah. I mean, I would say we are approaching 500,000 degrees awarded. We are at about 485,000 degrees awarded. And I think we are serving at this point about 350,000 unique individuals in a year. So probably some of my bio stuff was a little bit dated.

Jenay Robert: Thanks for the update.

Sophie White: Yeah, the verbal update's great. So let's dive into this topic. I'm curious to hear what you two think about how higher ed can better use data and analytics strategically. And both of you represent different sizes and types of institutions. So we know that every university doesn't necessarily have the resources to hire a large full-time data and analytics team, and we're working within the constraints that we have, but we have so much data that we're gathering in higher education now that making sure that we're using it effectively to drive decision-making, especially amidst resource constraints is really important. So I'm curious, whoever wants to chime in first, what are some ways that your institution is thinking about how to use data for strategic decision-making?

Nadeem Syed: Yeah, I can dive in. I mean, the advantage that somebody, an institution like WGU has is we are a digital native institution, we were born online. So every single aspect of the student journey is instrumented and you can really observe where the students are struggling, where they are succeeding, where they're learning. So in terms of data as a strategic asset, then the flip side of that is then you have this massive data exhaust, you have to know what you're going to do with it. So data in and of itself is not sufficient unless you can ask the right questions. And so for an institution that's 100% online, it really behooves to have good data governance culture, database decision making practices. It really starts, I would say, from the point when we are designing the program, right? You start from the start, you start thinking about what are the things you should be capturing?

And because when you can, and the phrase we use sometimes internally at WGU is if there's something that you can know, you should know. And that means you have to start from the time you're starting to design the student journey, you're trying to design curriculum and program and courseware is how do you instrument those things to learn where the students are struggling? And then that's just one aspect of it. The broader use of data, if I look at it, I put it in four categories. You're looking at student success metrics and looking at how do you improve certain outcomes, that's during the primary usage. The second is how are we using data to improve operations and how we are delivering services and co-curricular support to the students? The third is decision support. The talent economy and the talent market is rapidly evolving. So how are we using data to drive decisions around content and curriculum and make decisions faster?

And the fourth one is reporting and compliance. Higher ed is a highly regulated industry. We have accreditors, we have state regulators, and having really high quality, good data makes it a lot easier to fulfill those obligations. So Shannon, I'll let you weigh in from your perspective.

 

Shannon Shank: Yeah, I agree with all of that. And I think one of the things that we are dealing with is having the ability to have the right data in all of our systems and having those move through in a seamless way through the software applications because there's integrations everywhere. And to Nadeem's point, from the beginning of that curriculum being built, you need to know what you need to report. And so partnering with the right folks, at least from my end, with institutional research or institutional effectiveness to understand what the requirements are, then we can make sure that the data is, number one, being captured and it's being moved around the way that it needs to be to be captured at the end in a succinct way. It's difficult because there's so many systems out there. It's not just the student information system. You've got your financial systems, you've got your housing systems, you've got all of these pieces that need to come together, and we have to be extremely good stewards of those.

So if we know the end goal, the external resource that we have to report to federally or locally, or even internally at the cabinet level, knowing the end goal there is what's going to be critical to get the right outcome.

Jenay Robert: You reminded me of someone, I can't remember if it was on this podcast. Somebody recently was talking about how in higher ed, what people don't realize is that our institutions are really like small cities a lot of times because we serve our students in so many different capacities that it's like this big city that we have to figure out how to coordinate. And gosh, being able to integrate data from all those different service viewpoints must be a real challenge. And I think another thing that came to my mind as you were both talking was that just as much as data can be a strategic asset for our institutions, the other side of that coin is that it can also be a liability because we have to protect what we collect and we have to make sure, as Nadeem was talking about, the regulatory environment. And so maybe the key there is really having that end goal in mind and being strategic about what we collect and why.

Shannon Shank: It's interesting, Jenay, I've said this multiple times in different areas. Someone asked me, "How do you run IT for an institution, like a school?" And I said, "Well, I don't run it." My CIO is in charge, but it's like running a small city. It really is. And I've said this so many times, so when you said it, I began to smile because that's exactly what happens. There's all these entities and all these people moving around and data just flying around. And we need stop signs actually sometimes because it's just a lot. But when you have a good concept of that, that's the foundation. If you know that that's the case, then you can build on that by having the right people in the room and having the right people movers and the right integration tools to get to the end goal. And also really good partnerships with folks like the CFO and institutional research and the provost.

When you have that level of partnership at that level, and then also right underneath, two layers underneath, you have to have the relationships with those staff too who are actually pushing the buttons. And so then it becomes a bit of an education for some so that they understand that when they put their fingers on that, on the mouse and the keyboard, it means something a lot bigger. And so you just got to show them that it exists and build that city with you.

Nadeem Syed: Yeah. And Jenay, to your comment about data, both the opportunity and risk with data, the phrase I use is you can either swim in data or you can drown in it. And so more data is not always a good thing if you don't know what to do with it. And to your point, data, just good standards and practices around protecting information around doing roads-based access, while balancing that with completely democratizing, if I can use that phrase, access to the data. And so that's where really good combination of data governance, but also data infrastructure, then data culture and data literacy matter because if you overcorrect on one side and all the data is sitting locked up, then it's no good to anyone. And if you put data out there without context, it can have many downsides. One is just the risks to exposing productive information that requires really good standards and governance, but then also inconsistent interpretation and different versions of data.

I mean, you can have complete chaos if it's free for all without any guardrails. And we'll probably talk about AI in this conversation because I don't think we can go very far talking data without AI these days. And with the advent of AI, those things are basically on steroids at this point, both the governance and the access piece of it that needs to be managed.

Sophie White: Yeah, that's such an important point. And thank you for sharing. I think the risks of having data too locked up and also the risks of having a free for all I think is really important. I recently took our EDUCAUSE Data Literacy Institute as a learner just to work on my own foundations of data. And one of the examples that the faculty used was it's almost like in teaching and learning, we do backwards course design where you look at the goals for what the course is and then teach to those goals. Data is kind of the same thing where you want to start by making sure you understand what you'll be doing with the data before collecting it so that we're not putting data at risk by collecting everything we can and then having to restrict access to that and protect that data that maybe we never needed to begin with.

I'm curious, both of you, we can definitely touch on AI later because I agree with you that in 2026, that is a conversation we need to have. But I'm curious, both of you have mentioned data governance as a really important foundational element of doing any kind of work with data. Can you talk a little bit about your current data governance programs at your institution, maybe how you formed them, where they're going, any tips that you have for other institutions working on data governance foundations?

Shannon Shank: I'll start because I know Nadeem probably has a very full situation at Governors. At CalArts, we are in the kindergarten stage, maybe first grade. I've done data governance at my last institution. I was there for a very long time, like 25 years. And so I had a lot of time to work with the teams and gather the right folks. At CalArts, we're in a bit of an infancy stage in that we know sort of what's wrong. And at a lot of institutions ... Actually, I went to an EDUCAUSE workshop a couple years ago with our executive director of institutional effectiveness. And a lot of folks there were talking about how the administrative teams need to take the problem up to the cabinet level so that then they can say, yes, go do something about data governance. At CalArts, it's the other way around. The president knows and he's like, "What are we going to do?"

So we don't have to convince anyone that it has to happen. So what we did a couple years ago was establish a data governance committee and I'm a big proponent of having the right folks in the room. So as you can imagine, the registrar's office and advancement and folks from the accounting and Bursar's team and admissions, but we soon realized that we had the people in the room, we had the right people in the room. However, folks weren't really involved with data governance. We had to get down to the basics like, what is a data steward? Literally, why do we have data governance in higher education and how does that affect everything else that works? And so we spent a good deal of time doing some education around that. EDUCAUSE was really helpful in some of the documentation that they have. So we did that, but then we're taking steps today in lifting the hood, literally.

So I'm AVP of enterprise applications, so I'm systems. So I'm looking at our student information system, I'm looking at tables and data that's in tables, data that's not in tables. There's just files that just aren't used, but could they be used? Yes. But are we using them? Not really, but we have to take stock in that. So let's do our inventory so that our foundation is set so that we can build a good data governance team. But sometimes like with us, we have to really, really start at pre-K and determine what it is that we need to do. Is the data clean? For the most part, yes. Is it in the right place? Maybe not. Folks make decisions, or they did prior to me joining. "Oh, we think we like that spot for this piece of data." It's probably not the right spot for that, because it doesn't roll up in the way that it should to be able to affect the future report that we need to do for IPEDS or something like that.

So our journey right now is learning and educating our users and helping us know what's wrong so that we can fix it to get to the next step.

Sophie White: Thanks so much for sharing that and for the honesty too of some of us are maybe in the kindergarten or first grade level and we're working on it and that's okay. So appreciate you sharing the journey with us. Nadim, do you want to talk a bit about your data governance?

Nadeem Syed: Yeah, I was smiling when Shannon said they formed the committee and realized they were not really governing because one thing that with just even my experiences and background I've learned is committees don't ... There might be a symptom, not the solution to the problem. And because if you think about data as a strategic asset, governance becomes more of an institutional capability. And so for me, it's different things, not just a bunch of individuals and rules and Shannon sharing the experience over there. So if you think about it as an institutional capability, then it really comes down to building those, I would call ownership standards and accountability for data, but then it goes beyond that. It's really about ownership across the data pipeline, right? As you were talking, Shannon there, role-based access and making sure that individuals based on the needs and roles, they can get access to the data, publish data engineering standards, those things that become critical.

Data instrumentation into product design, that means the product design is thinking about data. I mean, we don't think of these things as governance. And then data literacy, like people knowing where to find data, how to use it, the ability to reason with data, because again, then it starts becoming a core competency and capability of the institution, and then it's not a fixed structure thing. It kind of grows as the institution grows and as the technology evolves and as the analytics approaches change. So that's my more broader perspective on what governance might mean.

Jenay Robert: Something that has come up a lot lately, and I think the conversation on AI has really pushed this struggle forward is how we balance the need to have some consistency in the way that we govern. And I think this can be rolled all the way up to just technology governance. It doesn't have to just be data governance, but that governance, there have to be some consistent principles across the institution, but we also want to leave some element of flexibility and space for things like unique needs in different disciplines or in different units of the institution that work differently. And I'm wondering how you both have tackled that balance. It seems to be one of these enduring challenges that we haven't figured out yet.

Shannon Shank: Jenay, you mean how do we meet the needs of the different kinds of needs across the institution from a data perspective?

Jenay Robert: Yeah, but also then balancing that against the need to have some consistent principles that always stay true. So just to give a concrete example in the AI space, for example, there have been a lot of new policies spinning up at institutions, new guidelines coming up to help people understand the appropriate uses of AI technologies, not just for teaching and learning purposes, but for how they do their jobs on a daily basis. And this heavily overlaps with data governance principles. And so there are many ways that we want to be able to put guardrails on how people use those technologies across the institution, but then there are also times when we have to understand that there are unique needs of specific individuals. And so I do think there's some tension between that institution-wide consistency, but then also that the flexibility that might have to happen at lower levels.

Nadeem Syed: Yeah. I mean, one risk with centralizing everything is, do you want to slow everything down and bring innovation to a screeching halt? And fortunately for me, I get to have the joy of leading not just finance, but all of the analytics and institutional research and all of the enterprise technology at WGU. So the way we approach it is there are platforms and foundational pieces that need to be centrally managed for the whole institution, for all the different entities that are part of the institution, because you do want to have some standards and consistency and controls and guardrails whether it comes to tools or AI or software development or technology. And then I can take for the data, the way we have built it is because this is one of those things where you want to hold objectivity and accountability clearly separate. So we have all of the analytics and research teams reporting into me, but then my teams are organized in a way that they're embedded within the operating units.

And so they're embedded within the schools. And so they have, each school knows who their analytics team is, even though they're reporting up to the CFO. And so because they have to know what's going on in school of health and with the nursing program and what are the challenges there? Just taking one example and for them to be removed from that, they won't be effective. So even though they report somewhere, they're deeply embedded and really our school deans think of them as part of their team. Similarly, more recently, what we have done is the federated technology model where you have the enterprise team building the foundational platforms and portals, but then you have a team for the schools that's building things around the LMS, the learning management system, the things that are more closer to the student journey. Then you have teams that are building things for the enrollment journey.

And so then you push those teams closer and embedded them within the units, but then you've given them the tooling, the foundation and the guardrail. So that's the balance where, and similarly for AI, we are building foundational pieces and data and giving people the ability to go and experiment and run different models and build different AI enabled processes, but then you're just giving them the tooling, you're letting the individuals. If somebody is responsible for instruction and evaluation, they are best positioned to know how to use AI to drive advances in that space. So you want to give them all the power. So that's the balance we have we have at WGU. Shannon, let you weigh in on that.

Shannon Shank: Yeah. We're not as advanced, of course, but we are really trying to help educate. We're doing it in teaching and learning, of course, with AI, of course, but we've been spending a lot of effort in educating the administrative teams on how to use AI in a proper way or best practice kind of way. I work at an art school, so people are free to do what they'd like to do in some respects. But when it comes to data, I mean, there are guardrails. We have them in place, of course. But to your point, Jenay, different people need different things, but we have to just take a pause and make sure that they're not doing some things that would put us at risk. There are just certain things that you wouldn't, certain data points that you wouldn't include in a Gemini, even a notebook LM or something like that, so that we can keep our data safe.

And so it kind of goes back to the education about being a good data steward and understanding that I think we were talking about permissions and who gets to do certain things with this data. That's really important. Years and years ago, I remember we would train people on our student information system in our test environment. And even though they were working full-time with us, they didn't get access to our production environment until they had been through everything. We don't do that necessarily at CalArts, but the concept of making sure that folks kind of understand what they're doing to make sure that we are keeping our data as safe as possible. And because I deal with software also, we have to hold our vendors accountable in the same kind of way and ask different kinds of questions about their technology and how they use our data and how we can ensure that it's not being compromised just for the sake of technology.

Every vendor I talk to three times a week, "Oh, we have AI." What does that mean? You need to be clear. I need to see what's back here. What's happening?

Nadeem Syed: Everybody has AI nowadays, that's for sure.

Sophie White: Yeah. Yeah. These are great points. And I just wanted to also mention, Nadeem, you mentioned the federated data. That's an element in our key technologies and practices portion of the EDUCAUSE Data and Analytics Horizon Report, which we include in the showcase. So if you want to read a little bit more about the latest thoughts and practices related to some of these data strategies, you can look there. I'm curious, thinking about about this idea of having maybe more of a hybrid model of some central oversight for the data governance while still giving individual schools or departments the agency to make decisions with data. Is there a specific example you can point to where you were governing the rules of the data, but you gave a school the longer leash to make a decision with data and what did that look like for them?

Nadeem Syed: For us, it's just the way we do work here. So you do not ever want to centralize the actual decision making. So for us, I would give you examples. Pretty straightforward is we develop insights around student journey all the way from prospect to graduate, and we have metrics. And so we have weekly and monthly business reviews around these. The actual set of metrics are agreed upon and targets are set. But then those accountability forums, the weekly and monthly business reviews are really run and owned by the school leaders. They are the ones who are having a conversation with their different partners across the institution and asking them questions as to why. I'll take an example if your retention rate is flat or not improving, why is that not improving? So just take one example. Other examples would be which programs to bring to the market, which programs to be developed, how does the curriculum need to evolve?

All of those decisions are being made by the schools using the data that is being made available to them through this federated platform and model. And they have access to dedicated analysts who are in their staff meeting, in their team meeting, in their working sessions. So they have access to the analysts, to the data, to the resources, and then they also control the forums where they're driving decision-making and accountability. So I would say that's how we do things as a normal course.

Shannon Shank: Yeah. I guess for us, it's on a smaller scale. However, we have a central location where we have institutional data that we're reporting on, and then we have the transactional, the real-time data. So not too long ago, we recently granted access to our advancement team to have a little bit more insight into some of that data to help with the work that they're doing with regard to donors and outreach and things like that, but also in reporting back on how our students are doing, our graduates and things like that. And so that was really helpful for them. The model like CalArts for a lot of time was one single point to go get data from. And so when I arrived at CalArts almost five years ago, that model is not really as efficient as it could be. And what people wanted was getting their hands on the data to be able to touch it and feel it and understand it so that they could then make their decisions, folks like the Provost and things like that.

So we implemented a tool that would allow them to see that transactional activity, but then we also made a distinction. This is where you go for institutional data, this is where you go specifically for real-time transactions. Registration open, we want to see that every single day. That high level institutional solution, you wouldn't go there to do that. So we had to help people understand that this is where you go for this and this is where you go for this. So it was clear and one single version of the truth at that higher level, so that when we walk into a budget meeting, everyone knows that this is where the institute stands, so we can make decisions on that. It may change in a week, but because registration just opened or something like that, but we've allowed ... And obviously folks like the CFO's office has some specific reporting needs that everyone else doesn't have.

Jenay Robert: Something keeps coming up for me because you both keep referencing scale at your institutions and have, I think, very different needs and environments in that regard. So I'm wondering if you could comment on how you feel institutions that are operating at very, very large scale versus institutions that are operating at smaller scales differ in their data governance needs and approaches, but also how they're similar. And I have sort of a selfish desire to prove a little personal hypothesis that we're much more similar than we are different. And as a researcher in higher ed, I've been working in higher education for over 20 years. And something that has been true for that entire time is that every individual you talk to at an institution will say, "Well, I'm not very much like my counterparts. My job is quite different because X, Y, and Z." And every institutional representative you talk to will say, "My institution is quite different than other institutions for these reasons." And yet we have so much common ground and maybe we would get more done if we focused a little bit more on those commonalities and supporting the work across.

But yeah, can you comment a little bit on those ideas around scale and how that might impact your data governance?

Shannon Shank: Yeah. I'll start. Is that okay, Nadeem? It's funny, Jenay, you're like, "We're so different." No, not really. At the end of the day, the data points are the data points, to be honest. Those aren't going to change. There's a field for this, there's a table for this. And what is helpful is, as I said before, knowing the end game, right? What is the goal? Who do you have to report to? What do you have to report? That's the common ground. If we have that foundation established, then we know what we need to get. The way that we would think of scaling is mostly internal because the external reporting, I mean, that changes because the federal government may ask for different reports and things like that, but for the most part, IPEDS is IPEDS. Everyone has to report it, so that's a good foundation. But internally, there could be all kinds of things.

And so as I sit in meetings and have conversations with folks, I'm often channeling my director of institutional research or executive director of institutional effectiveness because somewhere down the line, they're going to ask for something that nobody ever thought of. And so my scale to scale is thinking of if we don't have it, how can we get it and make sure that it is in the right place so that we can get it and report on it effectively. But also I'm challenging people to think differently about their work and how they're approaching the work that they do because someone else is going to need it. It's not just you in this bubble. So share widely, share broadly your thoughts and your thinking as you come into these things so that we're actually being good stewards of our data, but also stewards to each other as colleagues because you don't want the folks that have to do high level reporting to have to go to seven different places to get what they need.

So you want to be able to, if it's your area and you are responsible for this certain dataset, then make sure it's the best dataset, make sure you're thinking about what might be missing, how we could be doing things better for our business processes so that high level resource gets what they need. So I'm scaling in that way in terms of education and helping people think broadly and differently about how they approach data.

Jenay Robert: Love that.

Nadeem Syed: Yeah. As I think about it, maybe one simple way to put it is what is similar is probably the human element. And what I mean by that is the student journey and the student outcome, I mean, it's astonishing the more and more people you talk, we are all trying to solve the same problems. Student retention, completion rate, what are students struggling with? Is it personal life? Is it curriculum? The questions we all want to understand about what's causing our students to not complete and succeed are remarkably similar. This is very a fixed set of things in higher ed. We are trying to really get people to enroll, to succeed in their curriculum to complete and to get a good paying job after completion. And so in that human's journey, the questions are remarkably similar. The other thing that is very similar, no matter how big or small you are, is then the faculty and the leader, administrators trying to answer those questions, which means the experience, the judgment, the data literacy, you might have that at a smaller scale or larger scale.

Ironically, the challenge can get bigger when you're at a larger scale, the human side of it. So I would say those are similar things. And another thing that are similar, maybe the underlying tools that are available to everyone, a data warehouse, a dashboarding tool, those can all be similar. I think where it gets different is just the infrastructure and the scale and the ability to invest in technology. As you scale and you get access to more resources, you can instrument more parts of the journey, collect more data, have a more deeper data lake, better infrastructure to run the analysis on those things, that becomes different. And the other part that does become different ironically is driving that consistency and communication, data literacy and messaging. Because when you're trying to develop and grow 500 people versus 10,000, you might think now your scale and you have more resources, but actually the levels of communication a lot more that individuals with their own background and experiences and context, trying to interpret the data and make decisions is a lot more.

So it does become a different flavor of a similar human problem. So the way I was thinking about it, I just came to the conclusion that the human part is remarkably similar. The infrastructure part is maybe a little different.

Sophie White: Yeah. I think the human element is a really great point. And I'm curious, both of you have mentioned just data literacy quite a bit. One of the reasons we decided to tap into this issue is because that came up in our EDUCAUSE Top 10 decision maker data skills and literacy. It sounds like Shannon has an incredible leadership team who understands the value of data, but I'm curious, do you all have tips of some data literacy strategies for maybe training or managing up to make sure that decision makers you're working with are data literate and are using data in effective ways and are understanding this? What can other institutions take in terms of data literacy skills?

Shannon Shank: So, okay, I'll take this really quick. At my former institution, okay, the answer to your question is show them. They need to see it and sometimes I need to see it in action. And I didn't realize it at the time, but this is essentially what we were doing, but we did an exercise, and this was really about customer service and how we serve students, but the lesson there was, it was a teaching moment for sure. But what we did was we did a little enactment of a student coming to our institution for a visit and going through the student lifecycle. They went to admissions, they went to financial aid, they went to advising, they got registered, and they went to pay their bill, and then they went to class. And that student did that at a regular time of the student lifecycle. It was, oh, I don't know, February.

And then they did another one sort of April, May. And then we had a student come at the very last minute, right before school started. If you think about this, one had more time than the other, of course, but the staff had to do a lot of lifting and a lot of communicating to make sure that this student, to Nadeem's point, was recruited and foot in the classroom. And how we approached that student versus those other students was a customer service kind of teaching moment because then of course it got sped up. "What? Okay, go over there. You got to go here really short break and that didn't serve the student, but our data also didn't get to where it needed to go in a timely fashion. So we talked about that student experience, but during that, we talked about how data gets entered into the system and gets to different places around these offices.

And one thing that popped up, one of our folks in the housing office said," Oh, is that where that data came from? "I never knew that. She had worked there for 15 years. She had no idea that somebody in the admissions office was the one that input something, but now what they learned was, " Oh, if it's missing, then that's who I need to go talk to about it. "And so my point about showing them is to show not necessarily mistakes all the time, but it's a cause and effect. If this doesn't happen here, this will not happen here. That's just the basics. And so when you reveal that to folks in however way you want to do that, it's helpful for them. It sticks and it really helps them serve the student better, but also serve their colleagues better because they know instead of working in their own little silo, they now know where it came from and also who what you're doing is affecting going forward.

So that's how I teach them. I show them.

Jenay Robert: I love that. And I think that should be a pre-conference workshop at an EDUCAUSE conference, teaching other people through

Shannon Shank: You know what?

Jenay Robert: That process. That is brilliant.

Shannon Shank: You know what? What is it? Topics just opened, right?

Jenay Robert: Yeah, the RFP just opened.

Sophie White: Submit it

Jenay Robert: First.

Nadeem Syed: Yeah. And I would take the later data literacy topic even one step above that is it really comes down to asking good questions and what I call structured problem solving, which is how do you train people to take a problem, break it in its components, and then ask what question they need to answer to solve that component, and then you get to what data do I need, and then where do I go find that data? And if that data doesn't exist, who do I need to talk to, to get that instrumented and captured? And while I'm capturing it, how will I make assumptions about that data? So it really comes down to, again, starting with the outcome and saying, what is it that we are trying to solve for? And the reason I raised that, and for us, what we are doing, I think you asked about examples of how we are developing data literacy and Shannon, show them is exactly the way to approach it.

In fact, for us, back to my example of trying to train thousands of people is we are developing curriculum where we actually have mock video interactions where we are showing two people having a discussion about a problem and how they went about then looking at why the outcome was not where they expected. How do you then go find the root location on where this is happening and why it's happening and how do you find the data? And then you just do that demonstration so people can watch it and learn how to do it. The second thing I would say, we are not there yet, but it wouldn't be more than six months to a 12 month before we get there is ... I also think that with AI, the data literacy and what it means will dramatically change because you don't need to be able to know where to go find the data.

You actually don't need to be able to find trends and patterns in data. You don't need even to be able to chart and track the data because guess what? These models, as long as they're connected to all your data sources, they can talk to you and train language. And so then the upper end of that experience, judgment, ability to ask the right questions and break the problems into piece parts and then interrogate your systems for answers to those questions becomes an even more critical skills because those basic foundational things are just going to get ... I mean, AI can take care of it. You connect all your data systems, it can answer the questions. What it doesn't know is which questions to ask. And I think that's where it gets really important that we continue to advance that high level, what I would call structured problem solving with data kind of approach.

Because if I could add onto that, one of my least favorite phrases is data-driven decision making because I'm like, wait, if data is driving decisions, what are the humans doing? And so I prefer data-informed decision-making where you still own the accountability, you still have to bring in your judgment, your knowledge, your experience to the problem set, and then you let the data inform it. You don't let the data drive it. And that's going to just become more and more critical because of what the AI systems can do.

Shannon Shank: Yep.

Sophie White: Great point. As an English major, I loved the discussion about specific words I think matter in this context and making sure that we're really diving down into why do we need the data? What decisions are we making? So thank you for highlighting that problem solving element. Unfortunately, this is the end of our time, so we have to wrap up here, but this has been a fascinating discussion. Thank you, Nadeem. Thank you, Shannon, for sharing all of your insights and for your journeys, where you currently are and where you're going as it relates to data-informed decision-making and working with data for strategic data and analytics. So thank you all. We appreciate your time.