Follow ACC Docket Online:  

10 Steps for Early Awareness and Relevance/Non-Relevance Selection

This is the second part in a two-part series on developments in litigation technology. In this article, the authors discuss the different methods of e-discovery. Learn how legal departments can save time and money with this new technology in the first article.

T here is no single way to gain awareness and select responsive documents. Here are steps often recommended by Jeff Johnson, a Kansas City e-discovery consultant, depending on the circumstances of each case. These steps are typically applied prior to loading content on the final review platform.

1. Spam and non-relevant standard email reports and messages

Review emails sent to large numbers of recipients and remove non-relevant messages.

2. Concept clustering

Cluster documents based on textual content - it may be possible to include or exclude large clusters of documents based on examining just a few of them.

3. No-response emails

Significant emails tend to occur in email threads with replies and forwards. Emails without responses can be clustered so non-relevant emails can be removed.

4. Inclusionary search term testing

Based on what has been learned so far, test initial key terms and logic for responsive documents across key custodians, noting related terms.

5. Exclusionary search terms for non-responsive items

Perform iterative non-relevant key terms analysis and remove non-relevant documents.

6. CAL

Use continuous active learning TAR as way to pinpoint key documents prior to sending files to review platform provider.

7. Specify embedded object treatment

Evaluate whether to create separate "documents" for each embedded object (e.g., a spreadsheet graph embedded in a Word document). "Exploding" embedded objects to create additional documents can clutter the review database and inflate data ingestion and storage costs. The review collection should be audited to ensure the specifications were followed.

8. Largest size files

Examine the largest sized files to avoid paying excessive data loading and storage on final review platforms (e.g., identify large PDFs, graphics, and video).

9. Sentiment analysis

Review documents that contain highly charged emotional content (e.g., profanity). Labor or employment cases and contract disputes often involve angry or accusatory emails that help prioritize and scope review tremendously.

10. Visual classification and glyph search

Visual classification technology can cluster visually-similar documents without using or requiring textual analysis. It is ideal for some collections and serves as a cross-check on text-based tools for any collection. The same technology enables searching for key graphical elements, such as logos, stamps, and graphics. 


Organizations processing e-discovery data


Lit tech-organizations-chartOrganizations processing e-discovery data

About the Authors

Seth EichenholtzSeth Eichenholtz is an in-house lawyer and head of electronic discovery at Mastercard. Prior to Mastercard, he worked as an e-discovery consultant and managed e-discovery at Swiss Re. [email protected]

Tim DonovanTim Donovan is an A-rated intellectual property attorney who has held a number of positions as general counsel in large and small corporations in the technology and healthcare sectors. He is currently a legal advisor to a healthcare software vendor in Kansas City and an advisory board member for Advance Law. [email protected]

Anne KershawAnne Kershaw is a lawyer and consultant who has been immersed in e-discovery for many years. She co-authored Judges’ Guide to Cost-Effective E-Discovery, teaches at Columbia University, and has written earlier articles for the Docket on discovery topics. [email protected]


The information in any resource collected in this virtual library should not be construed as legal advice or legal opinion on specific facts and should not be considered representative of the views of its authors, its sponsors, and/or ACC. These resources are not intended as a definitive statement on the subject addressed. Rather, they are intended to serve as a tool providing practical advice and references for the busy in-house practitioner and other readers.