.png){kind=small}
AI headlines often promise transformation. But at Enroly Bites: AI in Action, one of the strongest themes that emerged was far more grounded: the biggest gains in higher education are coming from doing the boring but practical things well.
Institutions in the room were at very different stages of automation and AI adoption.
As one delegate put it:
“The most successful institutions are doing the boring stuff really well - automating workflows before adding a layer of generative AI.”
This shift (away from hype and towards practical progress) set the tone for the discussion.
Why starting with the basics matters
Across admissions, student enquiries, compliance, and operations, attendees repeatedly reflected on how foundational automation is creating capacity for teams.
One participant shared an example of automating an end-to-end admissions workflow, including:
- Inbox management
- Folder processing
- Data visualisation
Once these processes were automated, teams were then able to layer in personalised generative AI for different staff roles. The changes weren’t glamorous, but they were meaningful.
A consistent theme emerged: get the workflows right first, then layer in AI.
AI literacy: A skill every university team needs
Beyond tools, the session reinforced the importance of AI literacy - understanding how AI works, not just what it produces.
During a hands-on activity using Perplexity AI, participants compared vague prompts with more structured ones. The difference was immediately noticeable to those taking part. Weak prompts led to generic, surface-level responses, while structured prompts produced more detailed, actionable outputs with cited sources.
Two reflections resonated strongly in the room:
“Always edit the prompt - never edit the output.”
“If you give every single person access to tools and education… everyone’s individual velocity increases.”
Participants noted that better prompts didn’t just save time, they supported clearer decision-making, stronger confidence, and more consistent outcomes.
Prompting approaches that resonated with university teams
When discussing how to build confidence and consistency, attendees highlighted several prompting approaches that had worked well for them:
- Using roles and context (e.g. “Act as a strategic consultant…”)
- Setting clear constraints, such as structure, focus areas, or exclusions
- Starting with smaller datasets to avoid model limitations
- Breaking complex analysis into manageable chunks
- Sharing prompt libraries so teams could learn from each other
Several participants referenced free, sector-specific prompt libraries designed for higher education, supporting recruitment, admissions, and operational tasks.
Creating confidence through experimentation
A recurring theme was that confidence grows fastest when teams are given space to try things out, without pressure to “get it right” immediately.
Rather than large-scale programmes or complex rollouts, many institutions described informal experimentation:
- Testing tools on low-risk tasks
- Comparing outputs collaboratively
- Sharing what worked (and what didn’t) across teams
These smaller, iterative approaches helped normalise AI use, reduce anxiety, and build shared understanding, particularly among teams who were newer to AI.
Practical starting points discussed in the room
The session closed with a set of realistic, low-risk starting points shared by attendees:
- Automate a single operational task
Email triage, enquiry filtering, or workflow routing. - Introduce a structured prompt template
One shared framework across departments. - Run a short AI literacy session
No slides - just tools and hands-on practice. - Create a safe sandbox space
Somewhere staff can experiment without fear of judgement or job insecurity. - Start small with data
Clean and standardise one process rather than waiting for perfect data.
Practical AI is about people, not technology
Many attendees reflected that AI adoption doesn’t begin with enterprise platforms or large-scale transformation programmes, it begins with confidence.
What emerged from the discussion was a shared view that by focusing on foundational automation, AI literacy, effective prompting, and a culture of experimentation, universities can make meaningful progress using tools they already have.
Practical AI isn’t about replacing people.
It’s about enabling teams to work with greater clarity, confidence, and impact - starting where they are today.
