Unstructured content is growing at an unprecedented rate, reaching 650% over five years, with Fortune 1000 companies managing petabytes of data. With electronically stored information (ESI) being formally covered under the Federal Rules of Civil Procedure (FRCP), organizations need new tools to manage, analyze, and review ESI effectively. This article presents eight techniques and technologies that can be used to lower costs and improve litigation success.
Early Case Assessment.
Early Case Assessment (ECA) enables attorneys to manage cases from a strategic top-down approach, providing information to make decisions on prosecuting or defending a case within the first 90 to 120 days. By examining the cause of action, key facts, allegations, applicable laws, key players, potential liability, and potential issues with ESI, organizations can make better decisions, reducing the time and costs spent on eDiscovery. ECA has reduced the cost of cases by 50-90% while improving litigation success to 76%. Traditionally, ECA is performed after processing and collection; however, new tools enable this to be performed before collection, with the data at rest.
Data Mapping.
Data maps provide documentation of an organization’s sources of ESI, a useful solution for eDiscovery, business continuity planning, and information security. For eDiscovery, data maps satisfy the FRCP requirement to supply "a copy of, or a description by category and location of, all documents, electronically stored information, and tangible things" to the opposing party. Traditionally, spreadsheets were acceptable solutions, but the growing quantities of ESI are bringing prominence to dedicated software, which improves the management and freshness of the data map. Advances in data mapping enable integration with legal hold notification, remote collections, and preservation.
Investigative eDiscovery.
Investigative eDiscovery brings a top-down strategic approach to eDiscovery, through “matter-based culling.” Traditional culling tools only support large-scale culling using secondary attributes such as custodian, date, and file type. This may miss ESI critical to the case and require going back to the data source late in the process, an expensive proposition that grows as the amount of ESI grows. Investigative eDiscovery enables organizations to find relevant ESI earlier in the process using a scalable search engine. Matter-based culling also solves the one traditional issue with culling file systems since files are not typically associated with custodians.
Concept Search.
Concept Search finds ESI using meaning instead of keywords, providing a solution when reviewers do not know the specific keywords used, as is often the case with secret code words, project names, and simple user behavior. In one case, attorneys and paralegals missed 80% of potentially discoverable documents because opposing parties used terms such as “unfortunate incident” and “disaster” to describe the same incident. Concept search uses techniques such as Latent Semantic Indexing, Probabilistic LSI, and Latent Dirichlet Allocation. An alternate solution to concept search is a synonym search, a type of query expansion, which uses a thesaurus to add similar words to a query.
Non-Linear Review.
Review is typically the most expensive component of eDiscovery. Traditional eDiscovery review is performed linearly in the order the ESI was brought into a case. This method is fast for systems but slow for reviewers due to context-shifting overhead, which occurs when the subject matter changes frequently. Non-linear review groups related ESI together so attorneys can improve review performance by up to 3,100% to 5,900% compared to traditional review. Non-linear review relies on technologies such as clustering, concept search, and faceted search.
Parallel Search.
Complex eDiscovery searches can take over a month to develop and involve hundreds of query terms. Running these searches can then take days to return results, increasing costs, and delaying case preparation. Parallel search enables searching 10,000s of terms against billions of documents in seconds to minutes. Using parallel search, organizations can move from trial and error approaches on a few terms to a simulation approach on many terms. Detailed queries of custodians, patents, products, codes, and other terms can be run efficiently, allowing the organization to find the best data.
End-to-End eDiscovery.
With over 600 eDiscovery software and services providers today, and forecasts that over 25% of providers will disappear by the end of 2011, vendor longevity is a legitimate concern. The traditional approach of using a vendor for a single or couple of stages of the EDRM process is giving way to vendors offering End-to-End eDiscovery. Single Vendor End-to-End eDiscovery provides organizations with an easier procurement process; however, the solution may have been built through acquisitions, leading to integration and usability challenges. Single Platform End-to-End eDiscovery provides additional benefits via a single technical solution, reducing costs and time by eliminating the need to transfer ESI between stages and enabling attorneys to "go back to the well" cost-effectively.
Cloud Computing.
The scale and project nature of eDiscovery makes it ideal for cloud computing, a distributed, utility-based deployment model that provides low cost, on-demand, and easy-to-manage computing power. The promise is so compelling that leading analyst firms forecast a $150 billion market by 2013. At the same time, recent surveys found that concerns regarding security and SLAs have restricted adoption to only 3% of organizations. The key to adopting cloud-based benefits while maintaining control over the data and infrastructure is the use of private clouds, which enable organizations to receive the same benefits on infrastructure they control. Private cloud-based eDiscovery solutions will provide greater scalability and lower costs while meeting security and SLA requirements.