3 Questions for UVA’s Sarah Cochran

This Q&A with UVA’s Sarah Cochran was catalyzed by a LinkedIn post by her colleague Kemi Jona:

Sarah is the senior director of learning experience and digital innovation. Kemi’s LinkedIn post intrigued me so much that I had to reach out to Sarah to ask if she would unpack things.

Q: Sarah, thanks for doing this Q&A together. Before we dive into what Kemi wrote about how you are driving the cost to develop online courses, let’s learn a little bit about you and your role at UVA. I love your background in music history, literature and theory. How does someone from your academic background go to a role as a senior director of learning experience and digital innovation? What are the big things you work on at UVA?

A: Thought it would be fun to try something a little different to kick off this Q&A. In addition to a traditional response for this first question, I have created a video using AI-powered tools. It is a quick and small example of how we are experimenting with emerging technologies to make course development more flexible and accessible.

Yes, my background is in music history, literature and theory—very niche! Like a lot of folks in higher ed, I thought I was going to be a professor. But after my TA funding ran out at the University at Buffalo, I started applying for jobs across the country and landed at Southern New Hampshire University as an adviser. At the time, I had no idea what instructional design was, but when a role opened on their ID team, I took a shot.

What I quickly discovered is that instructional design blended everything I loved: creativity, learning, collaboration and problem-solving. It gave me the chance to build experiences that matter and work with smart, passionate people every day. From there, I grew quickly, as SNHU was expanding rapidly, and I got to be part of that shift from scrappy start-up to a major player in online learning.

Since then, I’ve had the chance to lead learning experience teams at Northeastern, Mary Baldwin and now UVA. At Northeastern, I worked closely with corporate partners like PwC and GE to build custom learning experiences—everything from equity-focused graduate programs to real-world applied learning in advanced manufacturing. Those collaborations helped me think differently about scale, industry relevance and what today’s learners really need.

At Mary Baldwin, I led both the learning design and advising teams, which gave me a unique opportunity to blend academic design with student support. We launched targeted anti-melt strategies and academic recovery initiatives, and I worked with our advisers to gather direct feedback from students. That feedback became the foundation for redesigning both the learning experiences and the support structures—ultimately boosting student success and cutting our summer melt nearly in half, from 25 percent to 13 percent. Who knew that my first job as an adviser at SNHU would come full circle?

Now at UVA, I serve as the senior director of learning experience and digital innovation within the Office of the Vice Provost for Online Education. My role is all about helping schools and faculty build high-quality online and hybrid offerings that are aligned with the university’s standards—and that students actually want to engage with. I lead a growing team, manage vendor partnerships, develop design standards and playbooks, and help pilot new tools and technologies, including the integration of AI. It’s a new division, which means a lot of the work is about charting the path forward, building trust and showing what’s possible.

So, from Schenkerian graphs to online learning ecosystems—it’s been a wild but deeply rewarding ride. I’ve found that having an eclectic background has actually been one of my biggest strengths in this field. Some of the best learning designers I’ve worked with have come from all kinds of disciplines. They’re creative, adaptable and quick to learn—traits that make a huge difference in this kind of work. Of course, there’s immense value in formal training, too—but I think it’s that diversity of thought and experience that makes this work so dynamic and impactful.

Q: OK, let’s get to the money. Kemi’s sharing about how UVA, under your leadership, has cut the development costs for online courses in half garnered some significant attention within our community. Can you share how UVA has been able to achieve these savings and what all of us in the online learning program development business can learn from your experience?

A: Absolutely! Cutting costs without compromising quality has been a key priority for us, and a lot of it comes down to being highly intentional in how we develop courses.

Before development begins, we meet with faculty and program leadership to clarify the course’s goals, the target audiences and competitive positioning. This early alignment helps us avoid unnecessary expenses and focus investment where it truly matters.

One of the biggest shifts has been reimagining the role of video in online learning. Traditional high-production, in-studio videos are expensive, difficult to update and often don’t add much beyond what students can already read. Instead, we have shifted to a more strategic and purposeful use of video, focusing on moments where faculty insight adds real value rather than just repackaging existing content or reading highly scripted text.

Some easy strategies that anyone can adopt right away include:

• Encouraging Zoom-style recording from faculty offices instead of polished studio-quality footage.
• Rethinking the purpose of video by using it for important context, storytelling and real-world cases and applications rather than just delivering lectures.
• Leveraging existing high-quality content from respected organizations, rather than reinventing the wheel.

We are also piloting a new generation of AI-powered video tools to explore whether they can improve efficiency and accessibility. AI avatars allow faculty to easily create and revise videos on their own, reducing postproduction costs and making content updates faster and more affordable. For example, creating a three-minute faculty video using an AI avatar can take as little as one to two hours if the script is already written. Once the initial video is generated, making changes is incredibly efficient: You simply edit the script text and reprocess the video—no need to book studio time again. That kind of edit takes about 20 minutes and costs just a few cents. And the best part for anyone who has worked in course development? Faculty or the designer can make those updates anywhere, anytime—even while traveling for a conference.

Now, contrast that with a traditional half-day studio shoot. While it can capture multiple videos, the costs add up quickly—typically $1,500 to $3,000 for the shoot itself, plus another $1,000 to $2,000 for postproduction. Even just one polished three-minute video can end up costing $1,500 to $2,000 or more, depending on complexity.

With an AI avatar, you are primarily paying for access to the platform, and most tools offer flexible subscription models—you can choose between monthly or annual plans based on your needs. If you don’t require a custom avatar or voice clone, there are also free or low-cost options available. This kind of flexibility makes it easy to experiment, scale and adapt without making a major financial commitment. Compared to the cost and logistics of traditional production cycles, the difference—especially for institutions looking to deliver high-quality content quickly and affordably—is substantial.

Regardless of the video format or production quality, video remains a passive learning medium at its core. And if there is one thing we know from strong learning design principles, it’s that active learning consistently outperforms passive approaches. With that in mind, we are encouraging folks to think beyond video. We have shifted more of our investment toward interactive content—using tools like H5P to create engaging, flexible and cost-effective learning activities. These activities are not only easier to update over time, but they also promote deeper engagement and stronger alignment with learning outcomes.

We also take a strategic approach to vendor management, negotiating effectively and ensuring that vendors align with our long-term vision and financial goals rather than upselling unnecessary features. By centralizing our vendor contracts, we are able to negotiate preferential pricing for all our projects and can also hold our vendors to the highest production standards, since they stand to lose not just one course build but the entire portfolio of work if they can’t deliver to our quality standards.

Finally, we are experimenting with AI-assisted course development, where AI helps draft initial content for faculty to refine—cutting costs and faculty workload while ensuring the final course remains accurate, customized and enriched with personal expertise.

But reducing costs in course production isn’t just about efficiency—it is about freeing up resources to invest in student-facing supports that matter just as much as course design—things like mentorship, networking opportunities, mental health resources and career coaching.

Ultimately, our approach is rooted in intentional investment: prioritizing active learning, streamlining production and reinvesting in the high-impact practices that truly enhance student learning and engagement.

Q: One of the concerns I hear from colleagues across higher education is that AI-generated content (from videos to lectures to assessments to study guides) risks crowding out educator-created educational materials. The worry is that if AI-generated content is 80 percent as good and 20 percent of the cost (or some ratio), the incentives to go with AI will be overwhelming. How do you address these concerns about quality while balancing the potential for great productivity and efficiency in online course/program creation?

A: I completely understand the concerns around AI-generated content and the fear that it might overshadow or even replace faculty expertise. As a community committed to high-quality education—regardless of modality—we should be vigilant about AI tools flooding the zone with low-value content. That’s not the future we want, and it’s not the future we’re building.

At UVA, our approach to AI isn’t about replacing educators—it’s about amplifying their impact and engineering more efficient, sustainable workflows for course creation and revision. We know how stretched faculty already are. Anything we can do to reduce production burdens while maintaining academic quality is a meaningful win.

One of the things I’m most excited about is using AI to remove the barriers that once made certain types of online learning impractical—because they were too costly, time-intensive or simply not scalable. We’re currently developing AI-driven, highly interactive and personalized experiences that allow learners to engage in ways that were previously too expensive to produce. For example:

• AI-powered negotiation simulations, where learners practice decision-making and receive real-time feedback.
• Customized branching scenarios that guide learners through decision trees or logic models—formerly too complex or costly to build at scale.
• AI-driven course development assistants that help faculty structure high-quality courses more efficiently, making the design process more accessible and less intimidating.

Coming from a smaller institution before UVA, I know firsthand how transformative these capabilities can be. Just 10 or 15 years ago, institutions like SNHU began to disrupt the higher ed landscape by rethinking operational models and introducing scalable, process-efficient approaches to course development and student support. That shift changed the conversation around innovation and access in a meaningful way.

Now, we are at a similar inflection point—but this time, it’s being driven by the rapid advancement of powerful, accessible tools like AI. What’s exciting is this technology doesn’t just improve efficiency—they open the door to entirely new kinds of learning experiences. And unlike earlier waves of innovation that favored institutions with deep resources, today’s AI tools are within reach for organizations of all sizes. That democratization of innovation could be just as disruptive—and just as impactful—as what we saw a decade ago.

Of course, innovation doesn’t happen in a vacuum. Academic leaders should care not just about cost, but about the broader impact these innovations can have across the learning ecosystem. When we create more efficient, scalable processes, we naturally reduce the cost of course development—and that’s worth paying attention to. Not because we’re trying to cheapen the experience, but because greater efficiency shortens the time to sustainability, frees up resources to reinvest in student support and opens the door to broader access and equity. When used thoughtfully, AI can help us do exactly that—streamline development, enhance quality and improve agility without compromising academic rigor.

Importantly, we never remove the faculty voice. AI might help generate a first draft or automate repetitive tasks, but it’s the faculty who shape the learning experience, infuse it with disciplinary insight and ensure it meets UVA’s academic standards. Our focus is on intentional integration—not efficiency for its own sake, but to create experiences that are more dynamic, inclusive and learner-centered than ever before. Ultimately, this is not about doing more with less. It is about doing better with what we have. We are not using AI to cut corners—we are using it to cut friction. That way, we can focus more energy and creativity on what actually moves the needle for students—building a learning ecosystem that supports faculty, excites learners and keeps us at the forefront of what’s possible in higher education. If we do this well, it benefits faculty, students and the future of higher education.