One of the most time-consuming tasks in vendor management is reading contracts to extract terms and summarize the agreement. Many contracts are dozens of pages long, and there are often 15+ different aspects that a financial institution’s board of directors.

Research
Ncontracts has an entire team of subject matter experts focused on contract management. Conducting a time-management study with this team, we learned that one human searches for and extracts the desired terms and cites what section it was sourced from, and a second human checks the work of the first.

The team also gives an overall grade to the contract, sets up a series of email alerts for the vendor manager, again all manual processes.

Solution
Employees of any financial institution are somewhat wary of new technology, especially those working specifically in risk management. Any solution would need to slot into their pre-existing workflow that they are confident in.

Leveraging a Large Language Model (LLM) to do the heavy lifting, the team was able to automate the process of locating key dates and clauses, as well as give the contract a score looking at 17 different factors. This shifts the the job of the user from hunting for the information they want to just verifying the information that was extracted is indeed the information they wanted.

Results
The subtle shift to the workflow has resulted (in initial tests) in a 40% reduction in time on task. When the feature was launched at our yearly user conference, 49 financial institutions signed up that day.