chevron-down Created with Sketch Beta.
July 24, 2013 Articles

Firsthand E-Discovery Surprises with Practical Tips

Avoid common e-discovery pitfalls by learning from someone who's been on the front lines for years.

By Claudia T. Morgan

E-discovery is constantly evolving. I have been an attorney for almost 15 years and have worked in the field of e-discovery since before the term “e-discovery” was coined. Throughout the years, I have faced a wide variety of challenging situations, and at this point, one might think that nothing would surprise me. But what we do and how we do it keeps changing. Here are some surprises I have faced in the past few years along with practical tips on how to be prepared for surprises you will undoubtedly face in your e-discovery efforts

Two common themes thread through these practical tips: (1) the importance of effective communication and (2) diligent documentation for all steps in the e-discovery process.


Knowing What You’re Searching: From, To, CC, and BCC
A few years ago I was working on a large labor-and-employment class-action matter. Shortly before this litigation began, the client’s information technology department chose a vendor to take over and host the company’s historical email archive. There were seven years of emails in the archive. Due to this existing relationship and the vendor’s custody of the data, this same vendor was the client’s choice for us to use in our e-discovery efforts.

Two weeks before we thought we were done with discovery, we were reviewing our discovery documentation and following up on questions raised in the searches we had run. And that is when the vendor told us (after we raised questions about a strange result) that in running the searches on the archive, the vendor’s staff had only searched for custodians in the FROM address field. They had not searched the TO, CC, or BCC fields. We were now faced with an additional 40 gigabytes of data responsive to our search terms.


Practical tip: Understand the systems you are working with. Many email archives are not designed for e-discovery and may not contain custodian data. Most often when email is archived, only one instance of each email is archived, regardless of how many instances may have existed on the servers or in individual employees’ inboxes. For e-discovery, this means that when collecting from an archive, one has to search for the custodians in the address fields of the stored emails. Understanding the limitations of the systems from which you are collecting is crucial to minimizing unpleasant surprises.

Practical tip: Document, document, document. Had we not followed up on search results that were clearly wrong (one of the key custodians was coming up as having no hits), we would not have discovered the oversight in the search process until perhaps it was too late. Maintaining, and insisting that your vendors maintain, comprehensive documentation that is updated during the process is crucial to avoiding surprises. This also gives you enough time to address any difficulties you may encounter in the process.


Redaction and Metadata
A few months ago, I was performing quality checks on a production we received from our e-discovery vendor before sending the production to opposing counsel. We had redacted some documents in their entirety as they were privileged attachments to partially privileged emails. The images were properly redacted, and the FULLTEXT field in the database was null. All seemed well and we were about to clear the production to go out the door. But then I noticed that the SUBJECT field for these attachments was populated.


This was odd for a number of reasons. First, the SUBJECT field is generally populated only for emails, and these were not emails. Second, when I have seen the SUBJECT field populated for email attachments, it is populated based on the subject line of the parent email. Neither of these was the case here. After further review, I noticed that the SUBJECT field was populated with the first 100 or so characters of the text of the document. This was precisely the information that we had redacted, and that was a problem.


Practical tip: Understand your vendor’s processing. Particularly when working with a vendor with whom you have not worked before, you must understand how the vendor’s processing software works. I had never encountered processing software that populated the SUBJECT field in this way. But now this is one of the topics that I go over, field by field, with new vendors. Knowing what the processing software does (and why) at the beginning of a case will save you time and headaches further down the e-discovery road.


Hidden Search Hits and Checking Your Work
Last year we were approaching the end of discovery in one of my cases. We had reviewed and produced the bulk of the documents called for by opposing counsel and were running additional searches based on documents of particular interest to us. We wanted to be sure our search terms and review had captured everything potentially relevant. In running those searches across the whole review database, we got results and had the vendor put those documents into batches for review. When checking the batches, we noticed that the batches seemed to be missing documents. It turns out that searching across the database had searched in folders we did not have access to and did not know existed. File slack files had made it past the filtering process and into the review database; this portion of the database was searchable from the vendor’s end but not viewable from our review end.


Practical tip: Know your database. When setting up your ECA or review database you must understand not only how it is set up but also what is being put into the database and how the permissions will affect searches, review, and production.


Documenting New Approaches and Technologies
I recently consulted with one of our case teams on a case for which they were about to start discovery. The challenge they were facing was that the issues in the case were of a purely legal nature; the heart of the case was the issue of agency. The e-discovery challenge is that business people do not think or speak in legal terms. So search terms were going to be practically useless in narrowing the universe of collected documents to a manageable size. Computer-assisted review seemed to be a good fit for this case. Slightly over 500,000 documents were collected, and after speaking with and reviewing proposals from a number of vendors, the case team chose a vendor with whom we had worked before and chose to use the analytics function in a popular document review platform.


The mid-level associate on the case reviewed a seed set of 3,500 documents to train the analytics. Contract attorneys then reviewed sample sets of 1,000 documents, and the associate reviewed any document on which the software and reviewers did not agree as to responsiveness. After three rounds of sample set review and analytics being re-run, the reviewers had looked at 12,000 documents, and the software had categorized all but 10,000 of the documents. Those documents that were categorized as responsive or not categorized by the software were then reviewed by contract attorneys prior to production. In total, the review team looked at 77,000 of the over 500,000 documents—15 percent of the total collection.


And then the case settled.


Leading our wrap-up efforts, I asked the vendor to give us the latest workflow chart and Excel workbook documenting the process the case had undertaken and the results at each step of the process. The vendor did not have it. We worked with the vendor for a month after that to have the both documents created so that we had proper documentation of the work we had performed.


Practical tip: Treat a new process as if you are working with a new vendor. Particularly as we move away from the traditional linear review to newer types of review, contemporaneous documentation of what we are doing and how we are doing it becomes crucial. Recent cases addressing predictive coding and computer-assisted review make clear that we may have to demonstrate to a court or opposing counsel exactly what we have done and defend our process. If we do not have documentation, we cannot defend our process. Thus, creating such documentation after the fact and under pressure is neither wise nor advisable.

Even if you are working with a vendor with whom you have worked in the past, if you are undertaking a new process treat the case as if you are working with a new vendor. Question and understand every step of the process, and require documentation at every step.


Good communication in e-discovery, as in many other facets of our lives, makes the process go more smoothly and helps minimize surprises. If you understand how your client keeps its data and how your vendor processes and sets up the data for you, then you will not face surprises like some of those I have faced. A challenge we as attorneys face in this aspect is bridging the gap between the legal and the technical. Too often project managers at the vendors we work with do not have a law firm or litigation background. So they understand the what and how of e-discovery but not the why. In my experience, working with project managers who understand the why of what we do makes the process significantly more collaborative and minimizes the surprises we face.


Documentation is crucial because that is how you prove what you have done and defend your process. Documentation also allows you to step back from the details and think about what is happening in your e-discovery efforts. Thoughtful analysis will help you be more efficient and do a better job for your clients. In addition, documentation protects you and your client from a loss of knowledge of the case if and when there is turnover on the legal or vendor teams. As we all know, discovery in some cases can last for years. Having documentation that is independent of any particular team member is a good practice and will save you time, will save your client money, and will lessen the risks and concerns of e-discovery.


Understanding that not all members of a team involved in a particular litigation will be “speaking the same language” is critical to e-discovery success. Most often litigation teams involve in-house counsel, non-attorney client employees, outside counsel, and vendor staff. The business people and vendors may not understand the legal terms or the legal implications of certain technical decisions. Inside and outside counsel may not fully, or even partially, understand the technology or technological issues that are discussed during the course of e-discovery. And counsel and vendors may not understand the business terms at the heart of the litigation. Being able to identify and address these gaps is a sign of a true e-discovery expert.


Keywords: litigation, technology for the litigator, document review, documentation, email, archive, vendors, litigation

Claudia T. Morgan – July 24, 2013

Copyright © 2013, American Bar Association. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or downloaded or stored in an electronic database or retrieval system without the express written consent of the American Bar Association. The views expressed in this article are those of the author(s) and do not necessarily reflect the positions or policies of the American Bar Association, the Section of Litigation, this committee, or the employer(s) of the author(s).