5 eDiscovery Best Practices From Industry Experts

As we enter a new decade of eDiscovery, we continue to follow the rise of new laws and tech advancements. From emerging collaboration platforms to the latest in privacy laws, the changes in the industry can feel overwhelming — leaving us sprinting to keep up.

To figure out how to best adapt to these changes, I sat down with five trusted eDiscovery experts to hear their go-to approaches. Read on for some fresh perspectives and hacks to apply to your eDiscovery process in 2020 and beyond.

1. Ask These Questions Before You’re In The Throes of eDiscovery

Robert D. Brownstone, Esq., Fenwick & West LLP.

A major eDiscovery pain point for many legal teams emanates from the preservation, extraction and review of data from communication/collaboration platforms. Robust, versatile tools such as Slack, Microsoft Teams, Atlassian Chat, HipChat, Evernote, and Wickr are all still relatively new, making their day-to-day information management still uncharted territory. Yet, the inevitable data request can strike at a moment’s notice, plunging a team into reactive mode. So, every company’s legal team should follow the motto of the Girl Scouts, the Boy Scouts and even the wacky myriad-horned goat in the animated film Hoodwinked: “Be Prepared.”

Ediscovery preparedness should be proactive before the corporate “body is on the table.” It’s crucial to have engaged in some effective advanced planning. To do so, ask the correct questions before you’re in the throes of eDiscovery. Here’s a list of some key inquiries to prepare your team. An organization that deploys one or more of these platforms should assess each tool, understand its auto-delete settings and routinize its approach to eDiscovery.

For the platform itself…

1. Is it hosted internally?

2. Is it in the cloud by an outside company and/or the purveyor of the software /platform?

3. Is it a free version as opposed to a for-fee version?

4. If it is a for-fee version, then which version is it?

For retention and capture…

1. Is content routinely logged in a given platform?

2. Is logging the default?

3. If no-logging is the default, do the settings allow individuals and/or teams to turn on logging ad hoc or in general?

For auto-deletes…

1. Are auto-delete rules in place or should they be? [e.g., age-based and/or volume-based]

2. If there is not any auto-deletion rule, is there a routine by which a platform’s administrator, a team leader and/or anyone in IT manually deletes content? 

3. If there is an automated or manual process, is there a protocol that delineates a process and, if not, then who will collaborate to write one up?

4. Whatever is or isn’t memorialized, what actually happens in the trenches — including vis-à-vis pertinent users and/or teams — as to retention, logging, deletion and the like?

2. APIs Are The Keys to Your Data

Olga V. Mack, CEO & General Counsel, Parley Pro

When assessing eDiscovery tools, there’s so much that goes into finding the right platform. From user permissions to deployment options, there’s always a number of features you’ll want to tick off before settling on a solution. One feature you may have heard about in the IT world, but maybe don’t know enough about, are APIs or Application Programming Interfaces. If you’re an eDiscovery professional, APIs will become increasingly important for you in this new age of technology. Here’s why:

When it comes to data collection, it goes without saying you’ll want to collect data in its most native, defensible format. With the rise of newer communication and collaboration tools, collecting defensible data is getting trickier and trickier. Why? The more dynamic capabilities in a platform, the more dynamic the data. Take eDiscovery for Slack for example. The more dynamic its data is, the harder it is to extract in a comprehensive way. APIs are the solution to this problem — they’re essentially what allows applications to talk to one another and transfer data from one system to the next.

For example, when you post a photo to Instagram, you’re given the option to simultaneously post it to Facebook. The reason Facebook is able to automatically populate the same photo, location, caption, etc. is because Instagram’s open API transferred all of the data that makes up that post over to Facebook seamlessly. Apply this same idea to your eDiscovery tool, and it can seamlessly extract and transfer data from any of your apps such as Slack, Gmail, or Dropbox.

Without an API, eDiscovery professionals must resort to screengrabs which won’t hold up as valid evidence. Think of the application you’re collecting from as a house, and APIs as the keys to that house — without an API connection, you can only see what’s on the outside of the house. However, with an API connection, you can see what’s going on inside the house. APIs are what turn surface-level information into valuable insights, and that is what makes the difference.

2. IT & Legal: Learn Each Other’s Languages

Nicole Thompson, Head of Client Success, Onna

As a vendor in the eDiscovery space, I work with both legal and IT teams to ensure our product is utilized to its fullest potential. As the years progress, I’m seeing a greater need for collaboration between these two teams. Each should have a shared strategy on how to tackle enterprise data / shared workflow — especially when it comes to onboarding new eDiscovery and compliance tools. The sophistication of today’s legal technologies has caused a high demand for lawyers who have IT expertise as well as IT professionals who have legal expertise. As the platforms we collect from become more complex, having transparency and alignment between these two teams is crucial.

The first step towards transparency and alignment is understanding every data source on the company tech stack. I recommend that IT and legal have monthly or quarterly check-ins to see what tools have been onboarded and how they’re being used. This ensures everyone’s on the same page, and there are no surprises when the inevitable eDiscovery request comes. When analyzing your tech stack, you may want to create a shared questionnaire to understand the limitations/considerations of existing data sources. This questionnaire can also help set your team up for success when evaluating future tools.

Potential Questionnaire:

1. What type of data is going to be stored?
2. Who is going to be using the tool? How is it going to be used?
3. Can it be connected to others via an API? Does it have an open API?
4. What is the probability it can store personal or consumer data?
5. What is the search like? What information is indexed?
6. Do they have native retention policies?
7. Where is the data stored? Who is the processor?
8. If needed, how can we get the information out? Does it have native exports?
9. Are there any other limitations you should be aware of? Are you limited by the type of version you’re using?
10. What are the user types and roles that exist?
11. Does it allow guest accounts?

4. Math Will Set You Free

Benjamin Sexton, VP of eDiscovery and Analytics, JND Legal Administration

Opposing parties often struggle to effectively negotiate data scope and reduction methodology. In one scenario, they attempt to negotiate the parameters of AI algorithms, where learning curves are steep, and technical decisions have a sweeping impact on which documents are exchanged. In another scenario, parties negotiate data scope using keyword lists, seeking transparency, usually at the expense of the quality.

As a best practice, I advise clients to avoid pigeon-holing themselves into a specific document retrieval or AI-driven review strategy. Rather than negotiate a specific document retrieval/AI workflow, I recommend negotiating the quality standards that must be met by the production.

To dive deeper into this idea, let’s take the Elusion Test in Relativity for example. The Elusion Test is a great way to statistically validate the results of a population reduction workflow. You can do this by simply demonstrating that the “discard” pile is nonresponsive to the matter. One can agree to run a statistically valid random sampling of the null set before any documents are discarded, to verify that they are unresponsive.

Often the best workflow for population reduction isn’t one-size-fits-all and can change throughout a case as issues and data types are introduced in the matter. It isn’t uncommon that we may use Active Learning, keywords, communication analysis, and clustering in one case to identify a review set and eliminate non-responsive documents. Rather than justify each button we click along the way from A to B, we use statistics to justify the results for opposing, without hamstringing our team to a single predefined methodology. Quality standards are easier to negotiate, and arguably more important, than methodology.

5. Turn Custodian Interviews into Production Interviews

Amy Sellars, Assistant General Counsel and Director, Discovery Center of Excellence, Cardinal Health

Before our tech skills kick in, our people skills are of utmost importance. I am an evangelist for early and in-person custodial interviews. Meet with your custodians in person whenever possible and give them context for the data interview. Phone is second best — an electronic interview might actually be worse than nothing. A standardized custodial interview form is a must, but the interview should be a conversation. During subsequent conversations with custodians, remember to ask if anything has changed about their data use. Discovery is iterative.

As custodians share information about the data they use and access, think about what that data will look like when you actually produce it. For example, when a custodian lists a database as one of the data sources related to a matter, ask for sample reports that you can use in your 26(f) or state equivalent to set expectations with other parties and the judge. Talk with the database administrator right away, and ask for the number of records that might be implicated — will that affect the possible format of production? Conduct your interviews with a finished production in mind.

About Onna

Onna is a data integration platform that helps legal teams collect from today’s most popular enterprise apps in a quick, and painless way. Built with a powerful API infrastructure, we help some of the biggest names in tech secure, access, and search across all of their enterprise information to find exactly what they need. Set legal holds, personalize retention policies, review and collaborate on evidence, and when you’re ready, export data into the review platform of your choice. (Yes, we’re compatible with them all!) Download our eDiscovery Guide for Emerging Applications to learn more or go ahead and contact us!