Discovery Tips for Government Investigations: Develop a cleaner process with these preemptive guides
By Daniel Meyler, Ryan Law Partners LLP and Anand R. DaHarry, Precision Discovery
This article is for attorneys and in-house counsel involved in the e-discovery portion of an internal investigation or government subpoena. The commentary is meant to be general — it should apply to investigations involving the Office of Inspector General, the Department of Justice, the SEC, state attorneys general, the FTC or even civil litigation — and provides advice from both the attorney and e-discovery consultant points of view. With that in mind, here are five tips to help make e-discovery less of a headache:
Understand the Data Landscape
There are many potential sources of data. Company employees communicate around the clock, at home and in the office, and using a wide range of devices and applications. This means that when beginning a review, you need to get a handle on your client’s data sources. Are emails automatically archived in a vault system? Do employees use the Good app on their personal phones for work email? What about cloud servers? And are employees having casual conversations in instant messaging platforms like Google Hangouts or iMessage? In order to create a comprehensive plan to respond to a government inquiry, in-house counsel, IT and outside counsel must work in unison with their e-discovery and forensics consultants to understand the client’s data landscape from the onset. In certain cases, it may help to engage with a digital forensics expert (like Cyber Diligence Inc., a Precision Discovery partner) equipped with state-of-the-art digital forensics tools, subject matter expertise, remote collection capabilities and experience testifying.
Most importantly, digital forensics experts will have the ability to perform a “deep dive” forensic analysis, if needed. They can recover deleted content, analyze registry entries, reconstruct web pages, carve out deleted documents from portions of the hard drive and much more. Attorneys need to understand their data sources to conduct a smart investigation. If something is missing in the email, it may be because an important period of email was not collected. And you never know when, months or even years later, you will be called upon to explain the sources of the data in an investigation. “Scope” is often the first question asked, and you can’t explain the scope of the review without knowing the data sources. To ensure completeness and correctness of your document production process, your e-discovery provider should have tools that will provide visibility and transparency into the data from the onset of the collection all the way through processing, review and production to the government.
Train Reviewers Early and Thoroughly
In a large review with many reviewers, take the time to train everyone thoroughly. Dedicate one to two hours to present the topic of the review both substantively (establishing what the case is about) and technically (introducing the review platform and phases of review). Regarding substance, you should present the allegations, the criteria for responsiveness and any particular issues that are “hot.” Presenting the topic “technically” means showing reviewers images of what the tagging screen will look like, discussing which phase of review you are in, and, if possible, discussing what’s coming in the next phase of the project. And always include a short presentation on the basics of attorney-client privilege and attorney work product doctrine. Assume zero knowledge. Everyone approaches privilege differently; you will want everyone to be on the same page regarding what you consider to be privileged (see more below on privilege). Most important, once you have trained reviewers and they have started the review, you should (1) make yourself available for questions in real time and (2) immediately begin spot-checking the work. We cannot emphasize immediate spot-checking enough: Do not wait two or three weeks before second review. By then you may have thousands or tens of thousands of documents already marked incorrectly. Find examples of documents that were coded correctly and incorrectly. Review teams will be happy for this guidance and it will make your life easier at the end of the review.
Be Careful With Privilege
This is obvious but bears repeating. Accidental disclosure of privileged documents is at best embarrassing and, at worst, can cause subject matter waiver. Always establish early and clearly a list of privilege terms: names of key attorneys, phrases like “privileged and confidential,” “attorney work product,” and “ask legal.” Isolate all of these documents in the beginning and have a special team of reviewers focus on documents that are both relevant and “potentially privileged” with a special eye toward privilege as well as subject matter. But in addition to these basics, you should also set up a “back end” privilege search of your final production sets with some obvious key names — e.g., if there is an in-house lawyer named Gregory on many of your emails, create a spot-check search of the production for phrases like “Greg said” or a Boolean search like “ask w/3 Gregory.” The general principle here is redundancy — have multiple levels of check to prevent the wrong documents from inadvertently slipping through.
“Bad Words” List
Key case facts often come from overly casual conversation. Employees sometimes use email less as “mail” and more as “chat” and make offhand (and sometimes off-color) comments that will raise flags to an opposing party or government investigator. You will want to know about these emails ahead of time so that you can fully understand (and be prepared to explain) context. In our investigations, we often use a “bad words” list — containing classic colorful language — to direct our attention to documents that contain commentary that very likely has no productive work purpose but may have investigative import. It is a quick and — no pun intended — dirty way to find documents that will raise the eyebrow of any government investigator.
Use Predictive Coding
Do not be afraid of predictive coding, also called technology assisted review, or TAR. While it’s not for every review, it can help you achieve powerful, positive results. Many government attorneys are enthusiastic about defense counsel’s use of TAR. Get government buy-in early in your process and try to meet or exceed the standards set out by the Department of Justice in antitrust cases.1 The best TAR platforms allow you to meet if not exceed the “precision” and “recall” rates of human review; one recent study showed traditional review achieving an average of 69 percent recall rate. In a recent case review, we were able to achieve 76.3 percent recall rates using Precision Discovery’s TAR work flow. And if you work for a small firm or a client that cannot afford to hire 60 attorney reviewers on a temporary basis, TAR may be the best if not the only way to cover the ground you need to cover. The process is simple. You review a “seed set,” consisting of a few thousand documents. You mark documents that contain responsive content. Based on your tagging of the seed set, the TAR system begins to “learn” what content is responsive. While the system learns, reports are generated to provide you with an understanding of how well the system is learning. An experienced TAR consultant will then collaborate with the attorneys to determine when the system has learned enough. At that point, the system will provide all documents it has categorized with responsive content.
1 See Technology-Assisted Review and Other Discovery Initiatives at the Antitrust Division, Tracy Greer https://www.justice.gov/sites/default/files/atr/legacy/2014/03/27/304722.pdf (last accessed Mar. 18, 2016).
Dan Meyler, an associate attorney at Ryan Law Partners, focuses his practice on government investigations, securities litigation, complex civil litigation and commercial disputes. He has represented corporations and individuals in criminal, civil and regulatory inquiries conducted by the U.S. Department of Justice, the SEC, and various state and local agencies, and he has helped lead clients to victory in high-profile disputes in federal and state courts throughout the United States. He can be reached at firstname.lastname@example.org.
Anand R. DaHarry, director of e-discovery and forensics at Precision Discovery, is responsible for business development and consulting solutions. He works deep in the trenches with his clients to formulate the best business practice approach and discovery solutions to their complex litigation cases and government investigations, in a consultative manner. He can be reached at email@example.com.