The evolution of a school, college or university chatbot service

Introduction

When you introduce a central chatbot for students and staff at your school, college or university, it will likely evolve through three distinct stages. Each stage builds on the last, expanding what your chatbot can do and how it supports your campus community.

To illustrate this journey, we'll follow two people: Yasmin, a student, and Thomas, a teacher. Their experiences show how a campus chatbot grows from a helpful information tool into an indispensable digital assistant.

Stage 1: Contextual Chatbots

In the first stage, your chatbot connects to the major systems on campus: student information systems, assessment records, library catalogues, learning management systems and more. This connection gives the chatbot context. It knows who is asking the question, so it can give personalised answers.

Yasmin's Experience

Yasmin opens the chatbot and types:

Yasmin: When is my next lesson?

Because the chatbot knows who Yasmin is and has access to her timetable, it responds:

Assistant: Your next lesson is Business Studies in Room 204 at 2pm today.

She can ask other questions too:

Yasmin: What's my average grade this term?

Assistant: Your current average across all subjects is 72%.

Yasmin: When is my next exam?

Assistant: Your Business Studies exam is on 15th December at 9am in the Main
Hall.

Yasmin: What books do I have on loan?

Assistant: You have two books on loan- "Management Theory" (due 20th Nov) and
"Business Case Studies" (due 3rd Dec).

Every answer is personal to Yasmin. She no longer needs to log into multiple systems or hunt through emails.

Thomas's Experience

Thomas uses the same chatbot, but his questions reflect his role as a teacher.

Thomas: Does Yasmin have an education health care plan?

Assistant: Yes. Yasmin has an EHCP on file. She is entitled to 25% extra time
in examinations and access to a quiet room if needed.

He can ask other questions too:

Thomas: How many annual leave days do I have left?

Assistant: You have 8 days remaining from your annual entitlement of 25 days.

Thomas: Which students are at risk of falling behind in my class?

Assistant: Based on attendance and recent grades, 3 students may need support-
Alan (62% attendance), Marcus (grades dropped 15%), and Priya (2 missed
assignments).

Thomas: What room am I teaching in next period?

Assistant: Your next class is in Room 304 at 11am - Business Studies Year 12.

Thomas gets instant access to information that previously required navigating several different portals.

What Stage 1 Delivers

At scale, a contextual chatbot handles thousands of routine questions. During the start of term, it answers countless queries about lesson times and room numbers. During assessment periods, it fields questions about submission dates. On results day, it responds to students checking their grades. Each interaction is personalised, reducing pressure on support teams and giving students and staff immediate answers.

Stage 2: Hyper-Context and Transactions

Stage 2 moves beyond answering questions. Now the chatbot can take action. It handles day-to-day transactions, initiates workflows and draws connections across multiple data sources. We call this hyper-context.

Yasmin's Experience

Yasmin asks about her next assignment:

Yasmin: When is my next assignment due?

In Stage 1, the chatbot would simply give her the date. In Stage 2, it goes further:

Assistant: Your assignment for Unit 2- Managing Change is due next Friday at
5pm. A score of 70 or above will help you maintain the grade average you need
for your first-choice university. Would you like me to book a session with the
academic support team? You're both free at 2pm tomorrow.

Yasmin: Yes please.

Assistant: Done. I have booked you a session with the academic support team at
2pm tomorrow. You will receive a calendar invite shortly.

What once required emails, phone calls and diary checks now happens in a single conversation.

Yasmin can now use the chatbot to:

• Book appointments with tutors or support staff
• Submit assignments directly through the conversation
• Search and apply for courses
• Reset her network password
• Update her contact details and profile

The chatbot becomes her single point of access for campus services.

Thomas's Experience

Thomas prepares for a busy week. He opens the chatbot:

Thomas: Book Room 12 for a revision session on Thursday afternoon.

Assistant: Room 12 is booked for you on Thursday from 2pm to 4pm. I've added
it to your calendar.

Thomas: I'd like to take Friday off next week.

Assistant: I've submitted your leave request for Friday. Your line manager
will be notified for approval. I'll update you when it's confirmed.

Thomas can now use the chatbot to:

• Make room bookings
• Request annual leave
• Search and enrol in professional development courses
• Transfer or withdraw students from courses
• Process budget transfers between cost centres
• Add support interventions for students

Complex administrative tasks that once required navigating multiple systems now happen through conversation.

What Stage 2 Delivers

Hyper-context reduces friction. Students and staff accomplish tasks without learning the quirks of each campus system. The chatbot becomes an interface layer that simplifies everyday workflows.

However, this stage also introduces new responsibilities. Consider what happens if Yasmin tells the chatbot she is being bullied. Before deploying a Stage 2 chatbot, your institution must decide: What information does the chatbot share with Yasmin in response? How is the safeguarding risk categorised? Who receives notification? How is the conversation recorded and stored within statutory requirements? These questions must be answered through careful policy design before launch.

Stage 3: Agentic Digital Assistants

In Stage 3, the chatbot evolves into a digital assistant powered by multiple specialised agents. These agents work together, each focused on a specific objective: supporting library research, guiding career decisions, helping with university applications or maximising academic achievement. When they cooperate, they become a powerful support system.

Yasmin's Experience

Yasmin is preparing for her final exams while also applying to university. Her workload is heavy and she's feeling stretched. She opens the chatbot:

Yasmin: Help me prepare for my Business Studies exam.

The digital assistant responds with a coordinated plan:

Assistant: I've found three revision guides in the library that match your
syllabus. I've also identified two areas where your recent assessment scores
were below your average- financial planning and stakeholder analysis.

Here's a focused revision schedule for the next two weeks. Your predicted
grade for this exam is a B. If you follow this plan and score above 75,
you'll strengthen your university application.

Would you like me to notify your teacher that you'd like extra support on
financial planning?

Yasmin: Yes please.

Assistant: Done. I've notified Thomas, booked a revision slot for Tuesday at
3pm, and added the materials to your reading list.

Later, she asks:

Yasmin: How is my university application going?

Assistant: Your personal statement is complete. Your predicted grades meet the
entry requirements for your first-choice university. Your referee has
submitted their reference. The deadline is in 10 days.
Is there anything you'd like to review before you submit?

Multiple agents have worked together: the grading agent tracked her academic progress, the university agent monitored her application, and the library agent gathered relevant resources. Yasmin experiences this as a single, coherent conversation.

Thomas's Experience

Thomas joins his weekly team meeting. The topic is student progress. He opens the digital assistant:

Thomas: Which students need additional support this term?
Assistant: Based on attendance, assessment trends and engagement data, five
students may benefit from intervention- 

- Yasmin: Strong grades but attendance dropped in the last fortnight
- Alan: Struggling with written assignments, may benefit from adjusted
  learning materials
- Marcus: Grades dropped 15% since half-term
- Priya: Two missed assignments
- Jordan: Consistent attendance issues on Mondays

Would you like details on any of these students?

Thomas: What support has already been offered to Alan?

Assistant: Alan attended two sessions with the learning support team in
October. His tutor added a note recommending shorter written tasks with
clearer scaffolding. Would you like me to adapt his next assignment brief?

Thomas: Yes.

Assistant: Done. I've created a differentiated version of the assignment with
simplified language and step-by-step guidance. You can review it in your
documents folder.

Later, Thomas creates a new subject chatbot for his class. He instructs it:

Thomas: When Yasmin uses this chatbot, challenge her with questions that
connect ideas across topics. When Alan uses it, support him with short
sentences, simple vocabulary and structured guidance.

Assistant: Understood. I'll adapt my interactions with each student
accordingly.

The assistant adapts its behaviour for each student based on Thomas's instructions.

What Stage 3 Delivers

At this stage, the digital assistant anticipates needs, coordinates across systems and takes meaningful action. It supports students through complex processes like exam preparation and university applications. It helps teachers identify and respond to student needs quickly.

However, this power requires careful governance. Teachers, course teams and support staff must remain the primary decision-makers for high-stakes outcomes. The digital assistant supports their judgement; it does not replace it. Its recommendations must be explainable. Its actions must align with the values and policies of your institution. When these principles are followed, the digital assistant becomes an indispensable tool that supports the countless decisions made on campus each day.

Where appropriate, outputs can be deterministic rather than probabilistic. When responses are crafted by staff based on policy rather than generated by the model, everyone can have higher confidence in the service.

Summary

The evolution of a campus chatbot follows a clear path:

• Stage 1 answers contextual questions. Yasmin checks her timetable and grades. Thomas accesses student information and his leave balance. The chatbot knows who is asking and responds personally.

• Stage 2 handles transactions. Yasmin books appointments, submits work and updates her details. Thomas makes room bookings, requests leave and manages student interventions. The chatbot takes action on their behalf.

• Stage 3 provides proactive, agentic support. Yasmin receives coordinated help with revision and university applications. Thomas gets actionable insights about his students and tools to differentiate his teaching. Multiple agents work together to support the whole campus community.

Each stage builds on the last. The benefits grow, but so does the need for thoughtful design, strong data governance and clear policies. When these foundations are in place, a campus digital assistant transforms how students learn, how teachers teach and how support teams operate.

If you would like to explore creating chatbots that respond contextually to student and staff questions, please get in touch with us.

References

1. Chan, C.K.Y. & Hu, W. (2023). Students' voices on generative AI: perceptions, benefits, and challenges in higher education. International Journal of Educational Technology in Higher Education, 20(1), 43-18.
2. Hussain, A. (2024). Ada and FirstPass: Bolton College's digital assistants for students, teachers and campus support teams. University of Greater Manchester.
3. Milana, M., Brandi, U., Hodge, S. & Hoggan-Kloubert, T. (2024). Artificial intelligence (AI), conversational agents, and generative AI: implications for adult education practice and research. International Journal of Lifelong Education, 43(1), 1-7.
4. Whittlestone, J. (2019). AI and improving human decision-making. https://jesswhittlestone.com/blog/2019/5/21/ai-and-improving-human-decision-making