Back to case studies
QBox tuned models

Prior to optimising our chatbot
with QBox, it was returning a 72%
correctness score. Now, it’s answering
questions accurately 98% of the time!

— Digital Content Writer, Australian University


One university in Australia employs nearly 9,000 people, enrols over 80,000 students every year and welcomes hundreds of visitors daily.

Like any other organisation that deals with a high level of enquiries, this university is no stranger to chatbots, having already deployed its own solutions across different verticals. One of which is its chatbot, which provides answers to students’, staff’s and visitors’ general queries on things like:

  • Taking new courses
  • Visiting the site
  • Submitting forms
  • Enrolling students
  • Accessing different locations and services
  • Verifying qualifications
 chatbot's challenges

What were the chatbot’s challenges?

  • More than 25% of questions were answered incorrectly
  • Manual, time-consuming processes to monitor and optimise performance
  • Limited resources to understand how intents were affected

How did QBox help?

After adopting QBox, the university was
able to:

  • Increase success rates from 72% to 98%
  • Manage and train multiple chatbot models faster
  • Track conflicts when new intents were added to the chatbot framework
  • Empower subject-matter experts to become chatbot trainers
  • Easily integrate QBox into its daily workflow as a cloud-native application and upskill employees

Why use QBox for your chatbot?

Our Gartner-recognised AI middleware application is the only tool that lets you test and predict the success rates of your chatbot before it’s deployed.

Our patent-pending technology allows non-technical subject-matter experts to own the chatbot data-model development process and see the status of their chatbot at intent level. This means you can develop your model quickly and easily. Plus, as QBox has easy-to-interpret test results, you’ll save time and resources, too.

In a nutshell, QBox gives you confidence that your chatbot can answer users’ questions accurately – leaving them feeling satisfied with the experience they’ve had with your brand.

See QBox in action

If you work with natural-language data models and you’re looking to quickly and easily understand, analyse, and improve the performance and results of chatbots and conversational AI platforms, QBox is the tool for you and we will show you why.

See a demo