Do companies have repeatable e-discovery process or dedicated staff in place?

Market conditions in 2009, according to The 451 Group, an independent technology industry analyst company, reflected new levels of cost sensitivity among customers, as tighter recessionary spending bit into budgets, and corporations embraced the in-sourcing of e-discovery as a repeatable business process. Opportunity in the market has not diminished, though.
 
As The 451 Group’s 2010 E-discovery User Survey indicates, most companies still have no repeatable e-discovery process or dedicated staff in place, with the overwhelming majority continuing to ‘do it themselves’ using existing internal tools and staff. Moreover, half of all businesses indicated in their responses that they plan to purchase e-discovery products or services in the next year, although few have finalised their decision of vendor or product. These customers’ future buying habits will help determine the fate of the market.
 
In order to know more about the current status, Text Analytics News’ Ritesh Gupta recently spoke to Nick Patience, director - Information Management, 451 Group. Excerpts:  
 
Can you provide an update on text analytics technologies such as categorisation, clustering and named entity extraction in the context of eDiscovery? What new developments have you witnessed in this arena?
 
Nothing major has happened regarding the underlying technologies in the past year. 
 
Recently, an industry professional told me: The most interesting developments seen are on the edges of e-discovery, in early case assessment and for what are essentially investigative functions.  Call these areas “making the case”.  They are areas where text analytics can really shine, in applying information extraction and data mining technologies to discover actionable or litigation-relevant information in corporate electronic stored information. How do you assess the situation?
 
I'd agree. e-Discovery has been a major driver for search and text analytics technologies for the past four-five  years, mainly in the US. However, these technologies only succeed when they have been part of an application that has the necessary workflows and processes that enable legal professionals to exploit them. Early case assessment - the process of examining case facts at an early stage to determine legal strategy - has been a major driver within the e-Discovery market, as organisations have sought to reduce redundant and irrelevant data to lower overall volume and expense at the attorney review stage, where clients are often charged by the gigabyte.
 
ECA tools typically use keyword and metadata search to help users locate the most relevant data, either after processing, or in the case of some indexing tools, as early as the identification stage. Some also feature analytics such as concept search or clustering, and may include further data de-duplication for culling. It's a market ripe for the application of text analytics, though so far many of the main text analytics players haven't paid the market much attention. 
 
Last year, you told me: You’re never going to get 100% precision and recall from technology alone. These issues are being tackled by a combination of technology and legwork by humans. What do you think still are the major challenges as far as the adoption is concerned?
 
Well, the adoption problem is slowly being solved in certain areas, most notably in voice of the customer applications. There, the buyer is increasingly in the marketing department and they're becoming ever more aware of the need to listen to the conversations going on the web regarding their products and services and tap into and react to them. Delivering such an application via a SaaS model has been particularly appealing to this sort of buyer.
 
Regarding precision and recall specifically, that hasn't changed and isn't likely to in the foreseeable future - I'm not sure people ever would - or should - 100% trust software to make key decisions for them without some human consideration. 
 
In terms of language problems, how has the technology progressed to encounter some of the issues? 
 
We've seen a lot more use by vendors of anaphora resolution in the past year, which is key for understanding longer blog posts. 
 
Last year you had pointed out that most eDiscovery challenges centers on unstructured data held in emails and documents. And in terms of accessing it across silos, federated search is becoming an increasingly central part of eDiscovery, rather than searching each silo separately. Where do you think the industry still needs to improve upon be it from technology/ solution?
 
Yes, I do feel federated search needs to become more prevalent. But this largely depends on connectors and I think many vendors are acting upon that this year. 
 
How do you think companies are using technology originally purchased for reactive eDiscovery for proactive risk management or information governance strategies?
 
It's very early days still but we're seeing some use of identification and collection-type technologies (think EMC-Kazeon, StoredIQ or newcomer Digital Reef) being used more for general information governance and that's where the trend will begin to manifest itself. 
 
What are your expectations going forward? What according to you are going to be the major developments?
 
As more and more companies bring e-Discovery in house, eventually they’re going to find themselves with a complex management problem on their hands. I foresee there being a resurgence of interest in e-Discovery service providers helping to manage the process but only for those that change their pricing model from a pure volume-based model to something that is linked more closely to their client's success.
 
I also expect the software vendors themselves to add more management tools to manage the in-house complexity this in-sourcing movement will create.
 
6th Annual Text Analytics Summit
 
6th Annual Text Analytics Summit is scheduled to take place in Boston (May 25-26). Nick Patience, director - Information Management, 451 Group, is scheduled to speak at the two-day event.
 
For more information, click here:
 
http://www.textanalyticsnews.com/text-mining-conference/view-the-agenda.shtml
or
contact:
 
Ben Satchwell
Text Analytics News
T: +44 (0) 207 375 7163