“Man vs. Food” is like “Man vs. Machine” in discovery management. In the television show, “Man vs. Food”, the common denominator is the “man” and the variable is the “food”. Who wins depends on the amount and type of food. In discovery management, the common denominator is the “legal professional” and the variable is the “machine” and the outcome can vary drastically.
A machine-only driven process assumes that the software will be able to achieve everything a client wants with limited use of the legal professional. The lure of this approach is that costs will be reduced. But is this possible? Commodity-based approaches such as this don’t take into consideration the need for legal professionals to be active participants in the process, whether they are lawyers, discovery management professionals or vendors. As I have written in the past, it’s not software alone that will get you want you want — its people, process and technology together. This doesn’t mean it has to be a high cost proposition, but there has to be a thoughtful approach to using technology and the only way this happens is through the legal professional. I may be stating the obvious, but there have been many articles and blogs that leave out the details as to how a document review or an ESI processing project can be done simply by using a black box. No project is the same and a black box approach won’t work in every situation. There is no “easy button” in discovery management as much as we all would like there to be. It requires experience, efficiency and qualitative technology to achieve the goal of providing a complete work product for the legal professional to use.
In the 1980s and early 1990s, OCR technology was thought to be the “machine” to replace traditional document coding (there was no ESI processing then). The thought was “just OCR everything and search what you get”. The belief was that this would reduce document coding and review costs. It did not work and the outcome was that searching was very difficult since there was no indexing of certain fielded data for searching. OCR was an important contribution to the times and replaced much of the labor-intensive in-text coding, but it did not replace the indexing of documents. When a client asked about OCRing and not performing any coding, I would always ask them what they wanted to achieve. The response was, for example, “I want to search all letters or memos authored by John Doe from 1980 to 1990.” The problem was you wouldn’t find a document type of letter or memo; an author field with John Doe or a date field from 1980 to 1990, since none of these existed with just OCR. Those indexed fields (metadata today) were required to achieve the outcome of the search. This is an old analogy, but it resonates today, as some would say the machine only approach will achieve what you want as long as it is what the machine does!
Discovery costs are high today and with the large amounts of data to process and review, there needs to be a constant, dynamic, thoughtful approach to addressing these issues. Along the lines of the machine approach, some look to predictive coding and fixed fee unit (per document) approaches to solve this issue. The lure is by using these two approaches, costs will be reduced and the buyer will have a budget at the outset of a project. The problem is, and will always be, that there needs to be assumptions, provided by the client at the outset, that determines the amount of data to process, the number of documents to review, host, and ultimately produce, in order to provide the fixed fee approach. Can a client do that? How will they know what these assumptions are? The issue then becomes once the assumptions change, the budget changes, and if the budget has already been approved, how is this addressed in the middle of a project? I am not advocating that there is no way to perform unit-based fixed fee predictive coding, but it has to be a thoughtful approach where the vendor and legal professionals partner together to achieve an outcome of a quality product at a lower cost. This is not a machine approach, but a collaborative approach based upon data knowledge, process, workflow and people.
My intent here is to shine light where there is limited exposure and highlight why, in discovery management, “man (the legal professional)” will always win, not the “machines”.