Issue link: https://epubs.iltanet.org/i/25416
REINVENT THE REVIEW FOR E-DISCOVERY COST SAVINGS these into discussion threads offers an excellent opportunity to streamline the review process. Technologies for automatic e-mail threading and grouping can analyze e-mail content and metadata, piecing together e-mail messages from multiple mailboxes and arranging them as needed. Also, a review project manager can redirect e-mail threads and their attachments as a group, increasing review efficiency. • Near-Duplicates: In a collaborative workplace, electronic content often evolves as individuals modify and finalize their work. This results in ESI collections that contain many similar documents. Technologies that identify near-duplicates and similar items, as well as present the critical differences between documents, can improve review efficiency and consistency. Another useful approach is to route all near-duplicate documents and e-mail messages to a specific reviewer, so the knowledge acquired can be applied to an entire review assignment. to bear by keeping a single train of thought for a single reviewer can add up to significant savings. This also allows a review team to prioritize a review by examining concept clusters and creating review strata. Quite often, such prioritization means better utilization of review resources where reviewers with greater knowledge of the case can be assigned to review complex content and those with less knowledge take on less complex content. “By applying appropriate deduplication processes, one can reduce the number of documents to review by 30 to 80 percent.” Statistical Sampling: While judgmental sampling, referenced earlier, brings substantial savings, a related sampling method –– statistical sampling –– allows review workloads to be reduced even more. The fundamental value • Concept Clusters: Concept-searching technologies overcome limitations of keyword and Boolean searches by significantly improving the ability to unearth relevant documents. That same technology can be used to group documents with “like concepts” that are reviewed as a single entity, typically by a single reviewer. Once again, efficiencies brought in statistical sampling is that by accepting a certain margin of error, one can estimate the potential number of documents in each review bucket. Statistical sampling is also a useful technique for quality control and can be used as a second-level review of the work produced by other reviewers. The EDRM Search Guide covers practical aspects of employing statistical sampling, such as selection of sample sizes. • Collaborative Review: For large review projects, one useful application of technology www.iltanet.org Litigation and Practice Support 41