With Agatha, Desire2Learn increased cases closed per hour by over 30%

D2L support agents use Agatha Assist to pull relevant past tickets, articles, and templates (macros) in the sidedock app to quickly answer incoming tickets, leveraging the collective knowledge of the team use.





Agatha Assist saves us time and increases our efficiency while improving our ability to provide fast and accurate answers to our clients.
James Millard, Senior Director, Global Support & Community, D2L

What does D2L do? 

Desire2Learn (D2L) is a cloud-based online and blended classroom learning software used by schools, higher education, and businesses across the globe.


D2L wanted to help their support agents be more efficient while still delighting customers. Specifically, they wanted to reduce ticket handle time and increase the average number of cases their agents closed in an hour. However, they did not want to risk lowering their high customer satisfaction (CSAT) score. 

D2L’s support process mirrored that of many support teams: agents needed to toggle between many windows and tabs to find help articles, past tickets, or other resources to help them answer customer support cases.

With unstructured data sources and different mappings for each of their data sources, D2L struggled to find a partner who could work with their current data. Plus, the tools they relied on to search through their knowledge sources relied heavily on keyword search, which oftentimes returned a long list of irrelevant search results. Ultimately, agents were spending more time looking for information than answering customer tickets.


With Agatha Assist, D2L didn’t have to choose between reducing ticket time and CSAT. D2L support agents use Agatha Assist, which sits in a sidebar app right in their helpdesk. Implementing Agatha did not require the team to do any of the index and updating of the team’s data. When an agent opens a new case in their helpdesk, Agatha automatically pulls relevant past cases, knowledge articles, and macros relevant to the new case for the agent to use as reference, leveraging the collective knowledge of the entire support team. 

Add a suggested answer easily

As it turned out for D2L, oftentimes, the answer on a past case is exactly the right response for new cases too. Using Agatha’s “Add to Reply” button, agents add a past response to a new response with a single click. Furthermore, agents are often encouraged to reference help articles in their responses to customers. Since Agatha pulls up relevant help articles to a case immediately when the agent opens the case, the agent can easily add the help article into the response as well.

Maintain accuracy

With over 90% accuracy, Agatha significantly outperforms other AI. Agatha uses Natural Language Understanding (NLU), which means that she understands language way better than any bot ever could. Forethought’s Agatha helped D2L index their data and learned D2L’s unique data, pulling relevant information for agents while they answer cases. 

Agents who are in the habit of leveraging Agatha on every one of their interactions are 3.5 times more likely to meet their weekly efficiency goals.

Agents who are in the habit of leveraging Agatha on every one of their interactions are 3.5 times more likely to meet their weekly efficiency goals.
- Sasha Antonenko
Customer Support Manager, D2L


With Agatha at agents’ fingertips, D2L’s end user support team reduced their time to close tickets by 13.7%, increased the number of cases their agents answered per hour by 32%, and decreased their time to first response by 56%. Furthermore, D2L found that those support agents who used Agatha were significantly more likely to hit their ticket quotas compared to the support agents who did not use Agatha.

Not only did Agatha exceed our expectations in these areas, but Forethought has also become an important partner of ours as we continue to look for new ways to innovate the highest quality support for our clients.
- James Millard
Senior Director, Global Support and Community at D2L

What’s Next

Using the Agatha platform’s cutting edge machine learning, D2L to identifying gaps in their knowledge center, so that the company can build the right resources to help agents tackle the most common support issues with less time. Next, D2L wants to leverage Forethought’s spam detection model to help them close spam cases, allowing agents to spend more time helping customers in need.

Questions about AI for customer support?

Let's talk