The client, a multinational pharmaceutical company that focuses on the discovery, development and commercialization of prescription medicine, engages with vendors across the globe for both direct and indirect procurement thereby leading to extensive paperwork carried out manually. Due to increasing engagements, the contract management process was time-consuming, labour intensive and lacked process standardization. As the operational cost was increasing exponentially, the client wanted to build automated solutions to map modules to help find associated project documents and a contract discovery and analytical tool that would strengthen visibility in key contract obligations.
LatentView Analytics built a custom contract analytics solution for the client using cutting-edge Natural Language Processing (NLP) and Machine Learning (ML) technologies to extract contractual terms based on user specifications (e.g., term, renewal, pricing, objectives etc.). The solution had three main modules – Optical Character Recognition (OCR) data extraction, Pre-processing and entity detection and Exception module. The output of the models were made available to the end users as an excel file. As the end users of excel had limited exposure to modelling/programming an automated mechanism was built to continuously train the model on new data.
The solution resulted in overall improvement of the contract management process by providing a holistic view of the data in a single excel file without duplication. The model was able to drastically cut down processing time for a single file from 10-15 mins to 30-60 seconds since the files included meta tags that enabled easy search.