P2P

9 I L T A N E T . O R G My specialty, discovery management, has made strides to incorporate analytics and the data sciences. In my opinion, much of that relates to the technology life cycle in discovery management rather than the technology itself. As the deployment of sentiment analytics has become more mainstream, structured analytics has matured and is perceived now as less risky with mass adoption. Without disparaging the technology, the largest motivator to deploying analytics in my work is the cost reductions that reduce risk of new workf low and results. In my discovery work I have devised some general rules in addressing new technologies - analytics included. Don't underestimate. It is true that the effectiveness of analytics is proportional to the quantity of indexed data. I once spent more time than I should have telling a technophobe attorney that we were piloting the analytics and that I had some doubts I could find the one email he needed to locate based on content and not specific words. Of course, and to my delight and chagrin, it was immediately located. It is important to understand that it is not a perfect science and it does rely upon many variables to be most effective - but don't underestimate the technology. Pilot / Proof of Concept. This relies upon both the provider and firm to fully commit to testing the software. I often tell clients that computer systems are as individual as the Info Technologist who designed it. In discovery, when it comes to working with client data in discovery it can be more of an art than science to determine how it is organized and the best method to get into the data. Likewise with due diligence, contract review and other core competencies for the data sciences – manage expectations and work through the pilot to the end to get the best feedback. In my discovery experiences, the piloting of new technology is invaluable to understand the processes – end results are often secondary as we run the pilot in parallel with traditional methods. Spend some time with the data. In addition to learning about the new technology, likewise learn about the data. For me, that is spending some time in client files and bringing in the case team to develop search term reports, provide analysis and feedback – and then seeing search results. I cannot hope to be successful by applying broad sweeping inferences into data when I can easily make educated guesses and my own inferences based on sampling and spending a bit of time with the files. Cindy MacBean is Litigation Support Manager at Honigman LLP. Cindy is a highly-experienced Litigation Support professional who has recommended technology solutions to support discovery workflow to provide innovative, appropriate and cost-effective solutions in law firms, their clients and corporate legal staff. In addition to her MBA in Technology Management, she maintains certifications in eDiscovery (CEDS), Information Governance (IGP) and US/Europe Privacy (CIPP/US, CIPP/E). Clean data. Likewise where good processes are in place - the personality of the data is a key driver for success. The ILTA>ON session Why Your AI Project Will Fail Without Clean Data and What to Do About It in the Technology Adoption track discussed the value of clean data. Their coverage of the topic was comprehensive and they even used the adage Garbage In. Garbage Out in the discussion. It is always worthwhile to mention that processes that rely upon accurate data will achieve varying degrees of success when including historical data. No technology can recover from onionskin paper from 1983. The analytic tools that data science and data analytics bring to the legal technologies have a myriad of uses to deploy cost effective tools that can add value to firms in their client services as well as the business of law. I look forward to learning more tips and tools in this issue relating to the deployment, maintaining and supporting the tools - as well as growing the technologies in the front and back offices of law firms. ILTA

• Cover