Collection was its own hurdle
As is often the case with construction litigation, there is a huge web of communication to sort through. Complicating matters in this case was the fact that many of the primary custodians had moved on and many of the managers were now focused on other tasks. Lucent had to essentially come in and function as the management team for this matter, working closely with and advising the CEO, the head of IT, and other members of senior leadership. We were ultimately successful in finding, collecting and consolidating all data related to 255 different custodians in the case, which came to a whopping 9.5TB in total.
Untangling the maze
As we began analyzing the data, we realized that there were only 3 custodians who overlapped in the 4 different matters. Everything else was unique. Further we had data in multiple sources that needed to be collected and aggregated: Salesforce, scheduling software, project management applications, Google drive, Box, Gmail and phone communications were all involved. This essential information wasn’t available through interviews, we had to piece it together from the data itself. Once this was done, the other pieces began to fall into place as all the data was categorized correctly.
Crunch time
Now that we had all 9.5TB of data, we began working with client considerations to process the data. This included standard time restrictions and search terms. Using client supplied search terms yielded a set of 4.5million documents to review. Our job was to reduce the data to review as much as possible, so we began an analysis of our own and provided an updated set of search criteria that reduced the data set to just 450k documents. A 90% reduction in data.
We then used ALP in two ways, first in the ECA workspace to test search terms for accuracy, make sure we were not missing anything, and to reduce the total reviewable population. The second way we used ALP was in the review workspace for a prioritized review. This allowed us to review fewer than 30% of the potentially responsive documents, reducing overall review costs significantly.
Signed, sealed, delivered
The timeframe is what’s notable here. Normally, a 9.5TB processing job would carry a time estimate of roughly 4+ weeks, but we did it in than half that time. Which goes to show that managing data isn’t a brute force enterprise. You don’t need a huge team, what you need is real expertise, the proper tools, and like Archimedes’ famous lever, you can move mountains.
