Digital White Papers

I L T A W H I T E P A P E R & S U R V E Y R E S U L T S | L I T I G A T I O N A N D P R A C T I C E S U P P O R T & C O R P O R A T E L E G A L D E P A R T M E N T S 31 in TAR 2.0 is considered an advantage over the TAR 1.0 process, especially when this decreases the burden and opportunity cost of having SMEs code documents as part of initial training. Like TAR 1.0, TAR 2.0 improves upon linear review by presenting a high proportion of responsive documents to reviewers early on. Unlike TAR 1.0, TAR 2.0 can accommodate rolling data loads. And while TAR 2.0 is still subject to challenges presented by low-richness scenarios, it can be preferable to TAR 1.0 as no time is lost waiting to determine whether a predictive model is possible. The main criticisms of TAR 2.0 solutions include that the continuous evolution of the model makes it harder to determine the overall volume of the review. Resource planning in the absence of clear estimates of review size can be inefficient and costly. A "pure" TAR 2.0 approach needs to be supplemented with a statistically valid random sampling to help manage this. Even with a yield estimate, the precise timing of how many documents will need to be reviewed ahead of review cutoff is unpredictable. Finally, the continuous promotion of high-ranked documents introduces the risk of showing reviewers "more of the same" at the expense of promoting diverse responsive content. This introduces the risk of surprises occurring later in the review when unexpected responsive content is ranked lower. TAR 3.0 Now that TAR 2.0 is well-established, innovation toward the next generation of TAR is underway. Some believe the best evolution of TAR may be to put advanced tools in the hands of everyone – essentially an improved user experience. An even better approach may be found in advancing the underlying technology to further accelerate review speed and accuracy. In fact, these different visions for the future of TAR are not mutually exclusive; it is possible for them to coexist. In terms of forward progress, any solution labeled as TAR 3.0 should include techniques for minimizing two important risks: surprises in content and surprises in cost. To minimize surprises in content, TAR 3.0 solutions should be designed to give the system access to a diverse population of documents early in the process. Minimizing surprises requires robust knowledge of the document population, which can be achieved through rigorous sampling and validation. To minimize surprises in cost, TAR 3.0 solutions should incorporate methods for determining overall richness and support principled review cutoff. This will allow teams to predict overall volume and enable staffing efficiencies. The hallmark of TAR 3.0 solutions should include the enrichment of CAL with sound methods for providing early access to the full range of documents, including examples of responsive documents. Advantages of TAR 3.0 include: • Minimal upfront investment and the immediate commencement of review. • Early exposure to a diverse set of documents to minimize the risk of surprises. • The incorporation of yield estimation supporting efficiencies in both workflow planning and identifying a clearly defined review boundary. • The ability to front-load examples of responsive documents in a measurable way to enable a system to identify responsive

• Cover