Secondarily, data allows AI systems to string tasks together, surface insights and support student success at scale, which is especially important as we begin thinking about agentic AI and autonomous activity.
This is only successful if the data underneath them can be trusted. Without that, you’re not building AI capability. You’re automating existing problems. What we don’t want is to have some form of bad data that’s acted upon to give a bad result. What we want to be able to do is reduce the probability that our data would be the reason that an algorithm failed.
Another consideration is that data is inherently biased on its own. So, a good data strategy helps identify where those biases exist, how to counter them and be aware of the way that it is used. The point is, we want to trust our data. It’s about understanding and recognizing those biases and how to properly construct or use AI to improve the data.
In the face of enrollment challenges, budget constraints, supporting student success, and trying to keep pace with AI adoption and evolution, the research is sobering: According to a recent EDUCAUSE report, data-mediated insights could make the difference between surviving and closing for some institutions. Ensuring that staff have the right skills to deliver analytics services and that both staff and leadership are equipped to use data and analytics to support decisions are critical enablers.
This is why EDUCAUSE named the data-empowered institution the number one IT priority for 2025. Not AI, not cybersecurity — data.
Where to Start Your AI Readiness Assessment
My answer is always the same: start with relationships, not dashboards. Understand how data is actually used across campus before you re-architect anything. Then establish even a light data governance structure: a small, cross-functional group with clarity on who owns which data domains, who's responsible for quality and what the guiding principles are for AI use and sharing.
When you have a governance group looking over that data, you are raising the confidence level that every query you do or every AI prompt that you create is going to return to you value versus uncertainty.
Build the basic artifacts: a shared data dictionary, so that everyone means the same thing when they say “retention,” and a data classification framework that maps to your compliance requirements under FERPA, PCI and applicable state privacy laws.
LEARN MORE: How to establish a right-sized modern governance approach for your campus.
CDW’s data strategy practice is built around exactly this kind of structured starting point. Their MOAT framework — Manage, Orchestrate, Act, Transform — gives institutions a sequenced path from current state to AI-ready. The goal, as I’d describe it, is to meet institutions where they are, and not leave them there.
We have access to incredibly brilliant people who understand data at a very high level for its governance and its value, and they can provide very detailed insights on how to create these frameworks to help universities better manage and govern their data.
I hear “I have to do governance” or “I have to create something. Where do I begin?” The assessments can help us understand where to begin. CDW has the expertise to break down these very big topics into very straightforward and simple steps to help higher education institutions. I'll say it this way: we are here to meet organizations where they are and to help them advance to the next step and the step beyond that.
