In a first for the use of technology-assisted review as an e-discovery method, the Delaware Court of Chancery issued a sua sponte order mandating that parties involved in a contract dispute use “predictive coding” or file a motion “show[ing] cause” why it would not be suitable for this particular case, EORHB, Inc. v. HOA Holdings, LLC. Neither party had raised the issue; instead, Vice Chancellor Laster suggested that this case – involving the sale of Hooters, an international restaurant chain best known for its “original” chicken wings, “good times on tap” and the Hooter Girl, among other things – seemed “to be an ideal non-expedited case in which the parties would benefit from using predictive coding” because “indemnification claims can generate a huge amount of documents.”
Vice Chancellor Laster’s decision is the latest addressing the use of technology-assisted review, coming on the heels of Da Silva Moore v. Publicis Groupe, Kleen Products, LLC v. Packaging Corporation of America, Global Aerospace, Inc. v. Landow Aviation and In Re: Actos (Pioglitazone) Products Liability Litigation. Despite this growing tide of acceptance, some skeptics continue to shudder at the reliability of “black box” technology because the process of arriving at coding decisions is not readily transparent – and therefore may not be defensible when called upon by the courts to explain what decisions were made and why.
Parties can take a number of steps to improve the transparency and ultimate defensibility of their technology-assisted review processes.
How to Achieve a Defensible Technology-Assisted Review Process
1. Meet with opposing counsel early in the case to agree on the technological tools to use during the review. If both parties do not agree to use technology-assisted review, seek the judge’s opinion on the matter before proceeding with discovery.
2. Leverage the appropriate expertise. Attorneys most knowledgeable about the subject matter should focus primarily on case strategy (and on the review of a sample of documents from the overall collection for training purposes), and technical, search and statistical experts should drive the use of technology. As necessary, engage the services of e-discovery experts who are able to explain complex e-discovery concepts to the court and opposing counsel.
3. Utilize an adaptive approach that includes a model process with analysis and course-correction built into each iterative cycle.
4. Create an audit trail that includes standardized reporting and documentation to ensure detailed records of all key inputs, decisions and results.
5. Measure the quality of results at every stage of the process, for ongoing QC and final QA, using statistically valid measures of precision and recall.
6. Maximize results by combining technology-assisted review with other methodologies, such as keyword search.
These efforts will result in greater transparency into the “black box,” and give those reticent to assent to technology-assisted review more confidence in the results it yields.