Further Streamlining the eDiscovery Review Process

Posted by

This post has been republished via RSS; it originally appeared at: New blog articles in Microsoft Tech Community.

In our new world of hybrid work, organizations continue to empower people to work effectively by being flexible in where and how work gets done. This flexibility has further accelerated digital transformation, resulting in an explosion of new types of data. These new data types are more dynamic than email and more complex to discover.


Organizations worldwide continue to face significant challenges in responding with the relevant data to meet not only their legal and regulatory obligations but also increasingly for internal investigations. A recent study conducted by IDC[1] found that legal departments’ responsibilities have drastically expanded as data privacy and compliance fall within their purview. There’s real pressure to do more with less, as organizations aren’t adding new headcount to manage these new responsibilities. In fact, 42 percent of respondents stated that improved efficiency is the primary reason they adopted cloud-based legal technology.


Given the explosion and variety of data being generated in an organization today, relying on traditional search methods to find relevant data is not only inefficient but may also result in some data not being discovered. Unfortunately, getting started with traditional solutions and processes can be challenging: organizations have to export large volumes of data, analyze this data in a separate system, then train or find subject matter experts to use the complex analytics tools. This process increases time, cost, complexity, and risk.

In March we announced several new capabilities to help legal teams find responsive content more efficiently with Advanced eDiscovery, including:


  1. Discovery of Microsoft Information Protection-encrypted content such as cloud attachments. 
  1. New iterative collections experience.  
  1. Enhanced support for Chinese, Japanese, and Korean languages. 
  1. New predictive coding capabilities.


Today we are excited to announce the public preview of several additional capabilities in our Advanced eDiscovery solution that will help legal teams efficiently streamline their eDiscovery review process by leveraging the power of machine learning to help identify, cull and reduce data without moving data outside the Microsoft 365 security and compliance boundary.


Support for large and complex cases

The explosion of data can be challenging for organizations with a large litigation and regulatory profile due to the volume and nature of the requests and current expectations of global regulators that govern data production. To help organizations efficiently manage these large cases we are expanding the total amount of content that can be managed within one advanced eDiscovery case. Legal teams can now:


  • Create up to 1TB per collection and add up to 40 million items to a single case.
  • Export Teams content as HTML transcript as opposed to individual items.
  • Collect time-based content for before and after the responsive item.
  • Export large sets of content within a single job, 5 million documents or 500 GB, whichever is smaller. 


Collection in a large case will also include Cloud Attachments and contextual Teams and Yammer content to help collect the full picture of digital communications.


With this new feature, organizations can now effectively accommodate an increase in case size in response to time-sensitive high-volume regulatory requests, investigations, and litigations.



Advanced eDiscovery allows creation of larger cases, up to 40 million items for a single case.


Leverage the power of machine learning to quickly identify relevant data

Unlike traditional predictive coding solutions that require users to label over 400 documents before seeing results, users can get started with training as little as 50 documents in a single training round. Our train-as-you-go approach lowers the friction for customers to get started and enables them to derive value from the get-go without incurring the traditionally heavy up-front costs of training a model.


Streamlined workflow to train your model

While many solutions have incorporated machine learning, they often require users to be briefed beforehand about concepts such as training sets, control sets, and confidence levels. Our machine learning solution includes these concepts, however, we are now introducing a 5-step workflow that is designed to reduce complexity without sacrificing model efficacy, transparency, or defensibility.



The new predictive coding workflow is designed to reduce complexity.


You can learn more about these innovative machine learning techniques here.


Further streamlining the review process

Legal teams are often under time pressure to deliver relevant data for a request or investigation to meet their obligations. To help organizations quickly review the data to determine what’s relevant, we have improved our review set layout. Our new layout provides customers with a clean workflow to efficiently review data in a set.



The updated review set layout makes it easier to find the relevant data.


Finding data in a review set has never been easier with our simple, yet powerful filter and query control. Users can select from a variety of filters to quickly narrow down items and find important documents. For advanced users looking to craft complex queries in order to meet requests for data, there is a dynamic KQL builder that supports multi-nested groupings of condition cards.



New filters and the Advanced Query Builder make it easier and quicker to find the right data.


You can learn more about these filter and query capabilities here.


Finally, to help organizations prioritize the collection of data, we are providing not only visibility into your queue of pending collections, but also the ability for you to prioritize specific collections and cancel those which are no longer needed. This enables organizations with multiple divisions or high case volumes to stage and sequence work to improve their eDiscovery workflow efficiency.



Visibility into queued pending collections allows users to prioritize specific collections as needed.


Get started

We are committed to helping organizations with their eDiscovery needs by continuing to deliver capabilities that make the end-to-end workflow more efficient.


You can sign up for a trial of Microsoft 365 E5 or navigate to the Microsoft 365 Compliance Center to get started today. Visit our documentation library for details on all our eDiscovery solution capabilities.


We would love your feedback, so let us know what you think of the features in preview.


[1] IDC, 2020 U.S. Legal Technology Buyer Survey, Doc # US45856020, July 2020

This articles are republished, there may be more discussion at the original link. But if you found this helpful, you're more than welcome to let us know!

This site uses Akismet to reduce spam. Learn how your comment data is processed.