20,000+ Filled--arrow-up-10

interactions in one day

12 Filled--arrow-up-10

user journeys

Education institution suffers from short-term influx

A UK-based university came to us with a problem. At a particular time of the year, thousands of prospective students contact them at once to see if they meet the baseline of the course they want to study. Called Clearing, this 2-3 day period sees thousands of youngsters contacting universities. Universities recruit hundreds of temporary staff members to answer phones and deal with the influx of enquiries, often following scripts and standardised conversations.

Expensive humans doing repetitive, mundane work? Ohhhh, sounds perfect for the ubisend geeks.

Too risky for the big day

Clearing is a critical time for universities. For some, it can be ~30% of the total annual admissions. Implementing new, front-line, AI-based technology with little time for testing at such a critical period is a tough internal decision - someone's bum was definitely on the line!

That being said, to reduce the burden on university staff, improve the process and impress students with the latest technology, the client thought of using technology to capitalise on the period up to Clearing day, but not Clearing day itself.

ubisend triage chatbot in 2 weeks case study

We built a pre-clearing conversational software solution (chatbot) that spoke to users and guided them through the process of finding out their predicted or actual grades, what course they were interested in and pre-qualifying them as appropriate.

Intelligent course baseline algorithm

As well as the front-end conversational aspect, the chatbot had a thorough understanding of the courses available at the university, the baselines for each course and whether it should be making initial offers to the students. Its bespoke algorithm was editable by university staff ensured the chatbot was making the right offers to the right people.

Living on-page on the Clearing section of the website, the chatbot was unobtrusive, surrounded by useful content and easy to use. It was purely an option for visitors, rather than an in-your-face, 'talk to me or go away' chatbot.

Over a two week period, the chatbot spoke to 917 people, dealing with 240 enquiries, pre-qualifying 114 of them. Not bad for an opt-in, 'there if you need me' solution.

The data for the chatbot lived on the university's servers, no third-parties involved and no extra GDPR considerations required. A bespoke solution delivered on budget and under tight time and technical constraints.

Nailed it.


Due to the confidential data we help our clients manage, we often work under hefty NDAs. If you’re interested to know more about this project, get in contact, as we may be able to introduce you to the client for a chat.