Insights Articles

Array recently sponsored a webinar in partnership with the Corporate Counsel Business Journal (CCBJ) on The Modern eDiscovery Playbook: Building More Effective Discovery Workflows.

The session sparked a wide range of thoughtful audience questions. Below, our speakers address the most common topics raised during the discussion, including several questions we weren't able to answer live. 

Session Panelists:

• Julia Helmer; Vice President, Advisory & Analytics
• Evan Levine; Director, Managed Review

Evan Levine: The objectives are different, even though many of the same documents may support both efforts. 

When responding to discovery requests, the focus is on identifying responsive documents, protecting privilege, and meeting production obligations in a consistent and defensible manner. Strategic review asks a different set of questions: What are the key facts? Which documents support or undermine our case theory? Who are the critical custodians or witnesses? What themes, risks, or gaps are emerging?

The most effective approach is to connect those workflows rather than treat them as separate exercises. As reviewers identify important documents, communications, and themes, those insights should be shared with the case team in real time. The goal is not just to complete review efficiently, but to leverage the review process to develop case strategy earlier and make more informed decisions throughout the matter. 

2. Should review objectives be defined at the outset of a matter to determine whether the workflow should prioritize discovery obligations, case strategy, or both? 

Evan Levine: Absolutely. Defining review objectives at the outset is critical because those objectives influence how the review is structured, what reviewers are asked to identify, how documents are prioritized, and what information is escalated to the case team. 

In many matters, the review needs to support both discovery obligations and case strategy. Those goals do not have to compete. A well-designed workflow can address responsiveness and privilege requirements while also identifying key facts, themes, witnesses, and issues that may shape the direction of the case. 

It's also important to recognize that review objectives are not static. As new evidence is uncovered, priorities may shift, additional custodians may become relevant, or the original case theory may evolve. Review workflows should be flexible enough to adapt to those developments and ensure the legal team continues to extract value from the document population beyond production alone.

Ultimately, the earlier a legal team aligns on review objectives, the better positioned it is to satisfy discovery obligations while also generating insights that support case strategy and decision-making throughout the matter. 

3. What level of legal experience is needed to use AI effectively within an eDiscovery workflow? Can junior lawyers or trainees use these tools effectively with appropriate supervision?  

Julia Helmer: AI-enabled discovery tools do not necessarily require years of experience to use, but effective use does require an understanding of the matter, the questions being asked, and the limitations of the technology. 

Junior lawyers and trainees can absolutely work with these tools when appropriate training, supervision, and validation are built into the workflow. In fact, natural-language interfaces can make sophisticated analysis more accessible because users can interact with the data by asking questions and having conversations with the AI system rather than relying solely on complex search syntax. 

What doesn't change is the need for legal judgment. AI can help prioritize documents, identify patterns, summarize information, and accelerate analysis, but people still need to evaluate the results and determine their legal and strategic significance. The level of oversight should reflect the task, the potential risk, and how the output will ultimately be used. 

4. How much training or hands-on experience does it typically take for legal professionals to become comfortable using AI-enabled discovery tools? 

Julia Helmer: There isn't one set timeframe because it depends on the tool, the user's experience, and the complexity of the workflow. Many newer AI-enabled tools are designed around natural-language interaction, which can make the initial learning curve more approachable. 

Becoming truly effective, however, involves more than learning how to use the interface. Users need to understand how to ask the right questions, evaluate the results, recognize potential limitations, and validate outputs appropriately. That confidence develops through hands-on use and experience with real-world workflows. 

Training should therefore focus not only on how to use the technology, but how to use it responsibly within a defensible discovery process, including reviewing conflicting outputs, prediction explanations and making use of statistical sampling to test performance over a significant population of documents. 

Missed the webinar? View the full on-demand recording to hear the full discussion and key takeaways from our experts.