The proliferation of nontraditional data in the working world is making the eDiscovery process more challenging than ever. From audio and chatrooms, to mobile applications, to photos, text messaging and emoji, the list seems to grow longer and more complex with each passing week. The bottom line for any company, legal firm or financial firm is to find a solution that allows it to render structured and unstructured data from a variety of traditional and nontraditional sources so that it is easily reviewable.

In a recent webinar, Drew Macaulay and Ben Rusch of Consilio and Dera Nevin of Proskauer Rose explored unique data types, including Bloomberg Chat, created specifically by financial services institutions. They also discussed why certain data types are less compatible with standard eDiscovery technology and shared tips about how incremental changes in capture and extraction can accommodate nontraditional data types, improve the review process and save both time and money.

According to the panelists, the data types that must be captured for review for use in litigation and investigations are growing in number and diversity and include the following:

• email,
• electronic files,
• Bloomberg or other IM/chat,
• audio,
• data on mobile devices and
• structured data.

Nevin said that new file types are emerging constantly, making eDiscovery even more challenging. For example, she cited the increasing prevalence of audio files, even in email. Other challenges include the variety of email types and the BYOD practices of many companies that make it harder to partition and extract personal messages from corporate data.

During their discussion, Nevin, Macaulay, Rusch and the call’s mediators outlined the challenges that come with reviewing audio files, particularly when you compare a trading floor call to a call center interaction. A 20-minute call to a trader may include only six or seven minutes of relevant discussion. New tools may be able to eliminate the silence and allow reviewers to concentrate only on the pertinent part of the call. In addition, using a tool that allows the reviewers to listen, rather than read, can be more effective because they can hear the speaker’s inflection and tone.

When it comes to audio, the eDiscovery “eyes on” metaphor doesn’t work: there is no reliable way for reviewers to get through 7,000 hours of telephone calls in a proportionate way. Taking an “ears on” approach is more effective, because many transcribing solutions cannot understand slang, proper nouns, dialect and acronyms. Consilio participants suggested that a phonetic indexing approach resolves these issues: for example, the solution parses audio files by sounds rather than entire words, so it can more readily identify, and thus allow reviewers to search for, acronyms.

All participants also acknowledged that they are increasingly dealing with requests for text message eDiscovery. An example included a contract that was entirely negotiated by text, requiring the retrieval of the entire series of texts.

Keeping up with today’s nontraditional data types can be daunting, but Consilio and Nevin shared examples of how incremental changes to eDiscovery processes can make extraction, review and reporting faster and more accurate. Listen to the webinar here.

Get Updates to Your Inbox

Subscribe to The Chronicle