 |
| Case Study: Early Case Assessment |
|
| Client Profile |
| The client is a medium-size law firm representing one of the largest chip manufacturing companies in North America. The chip manufacturing company is publicly traded on NYSE and has annual revenue in excess of US $2 billion. |
|
| |
| Case Background |
| The law firm’s client suffered shareholder’s derivative law suit with allegation of back dating of stock options leading to fiduciary violations under Corporations code and violations of Securities Exchange Act. |
|
| |
| Cruz Solution |
The task for Cruz eDiscovery was enormous since the allegations required the client attorneys to go back in time up until 12 years (1996) and pull any and all records relating to all the stock options that were given to all the employees of the company. The attorneys particularly wanted to get a quick handle on the case facts so that they can suitably advise the client. The senior management of the company was particularly affected by the derivative law suit and hence the sensitivity.
Cruz eDiscovery provided the desired early case assessment solution to the law firm attorneys. The early case assessment method provided timely legal analysis to attorneys to formulate the case defenses and suitably advise the client. The lawyers found it particularly useful to quickly review the data labeled into different categories. The Cruz platform offered lawyers to utilize automatic dynamic multidimensional categorization (about 29 different categories including foreign language emails) of the data index in guiding them through the maze of thousands of pages of data. In this respect, note the data was automatically segregated based on sender domains, amounts, concepts/topics and financial quarters of the company and since most of the alleged violations had taken place only in last 8 quarters, the lawyers immediately focused on the relevant data and that reduced the review span from hundreds of thousands of docs to few thousands. Applying script manager and decision tree techniques, Cruz was able to provide the attorneys a viable navigation method that quickly culled the data to arrive at handful set of relevant documents (few thousands). Cruz wrote specific customized first pass scripts in java to accurately cull the data such as the script to obtain all documents pertaining to “any analysis, computations, reports, schedules, general or sub ledgers relating to stock compensation expenses for the fiscal years 1996 to 2006”. This targeted culling was particularly helpful in data set reduction and gave the confidence to the legal team because many aspects of search were covered and not just few key words or search term phrases. Often the processing terms are focused and result in limited dataset reduction leaving out either too many important documents (under inclusive) or encompassing too many unimportant documents (over inclusive), clearly either of the scenario is undesirable.
Cruz closely worked with the law firm and provided processing with legal analytics including online collaborative culling to the attorneys of the law firm. Cruz de-duplicated, searched by file types, keywords, terms, and used fuzzy, proximity, wild card searches to cull the data.
After processing and review, the data was loaded on to the collaborative culling hosted secured platform for legal analytics, that provided the attorneys an opportunity to perform legal analytics for 2 months at no extra costs and data was sent to folders before creation of load files for the online review platform.
Our client was able to successfully defend itself in the case.
|
|
| Case Statistics |
| A company's initial collection total was 500,000 documents for review. Using an early assessment tool, that number was dramatically reduced through de-duplication (20%), using fast multi-dimensional categorization, scripting, decision tree technique to sift through data - the lawyers were able to quickly and broadly identify the documents that either weakened or strengthened the case and formulated the case strategy. Soon the lawyers used search terms and criteria to identify relevant data causing 25% reduction in data, privileged documents were excluded and caused further reduction by another 10%, and identifying the responsive data set caused 70% reduction. The final number of documents loaded onto the content analytic tool was: 80,000 (or just 16% of the original collected documents). |
|
| |
