The Artificial Intelligence  Success Story

Implementing Artificial Intelligence for legal research saves $1M for a LPO



Our Expertise

Big Data, Cloud, Platform Reengineering

Key Benefits

  • Enhanced process accuracy of 98%
  • 90% saving in effort of document classification
  • $1 million saving in process cost over 2 years


    Technologies Used


    Our client is a  leading legal services firm headquartered in Chicago, with a global presence. The company specializes in contracts, litigation & investigations legal analytics. It is one of the top ranked LPOs as per Frost and Sullivan.


    The company provides Income Tax related legal updates to law firms in the US. There are about 2600 federal, state and local websites that regularly publish relevant content for income tax analysts. The process of extracting relevant content from various websites was time consuming and involved a team of 15 legal experts monitoring websites, analyzing the changes and then summarizing the document for consumption by the clients.


    Challenges with this process involved the following:

    • Repetitive process susceptible to manual errors,
    • Labor and time-intensive process often resulting in delays in the availability of final output,
    • Poor job satisfaction to highly skilled legal experts due to non-core work, resulting in attrition issues.

    The Solution

    Ellicium worked with the legal experts of the client to increase the efficiency of the process using Artificial Intelligence. Python based Data Robots from Gadfly, the text analytics platform, were used to automate the process of monitoring the websites and downloading the updated content. Furthermore the classification algorithms developed in Gadfly automated the task of classifying the documents and websites. Support Vector Machines, Naïve Bayes, Rocchio were some of the algorithms used for the purpose of document classification. 

    The revised process involved the following steps:

    • Monitor of all the 2600 websites using automated monitoring scripts
    • Download relevant content using automated robots using Python scripts
    • Classify downloaded content using artificial intelligence algorithms.

    The Results

    • Average execution time of the entire process was reduced from 12 hours a day to 3 hours a day.
    • Manpower requirement for executing the process came down from 15 resources to 2 resources.
    • New relevant legal insights were discovered that were skipped due to manual processing.
    • Overall accuracy of the automated process was found to be 98%.

    Overall, adding ‘Artificial Intelligence’ to the Legal research resulted in a cost saving of $ 1 million over a 2 year period!!!


      Let's Talk!