A grim look inside the e-discovery problem

Is the game worth the candle?  Not a new question for class litigation defendants and large organizations generally who are regularly targeted with complex litigation involving extensive discovery.  However, the price of candles has been going up, thanks in part to the massive increase in electronically stored information (ESI) retained and stored by parties.

Even highly sophisticated technical organizations are choking on the sheer volume of data having some arguable relevance to the litigation they face.  Take for example Microsoft.  Here is a company that knows a thing or two about technology and would appear to be at least as well positioned to deal with the challenge as the next guy.  While that may be true, Microsoft has joined a consortium of other similarly positioned companies to push for clearer and less burdensome discovery rules.

As Microsoft reported to the Federal Rules Advisory Committee this past summer, the company maintains a dedicated data center for the preservation of ESI for ongoing cases.  The data center currently holds the equivalent of 740,000 banker’s boxes worth of documents.  For the average case, the company preserves 48 million pages, of which it collects and processes 13 million pages, actually reviews 645 thousand pages, and produces 141 thousand pages.  After all of this effort, Microsoft estimates that only 142 documents are actually used in the case.

A recent study performed by the RAND Corporation highlights the costs associated with the activities of preserving, collecting, reviewing, and producing such large quantities of data.  As the RAND paper also notes, while corporations are spending all of this time and money to preserve data that no one will ever use they continue to feel very insecure about the defensibility of their actions.  The result: even broader preservation, higher costs, and increased distraction of employees from their core business functions.  Indeed, business decisions such as whether to adopt certain information technology products must now factor in the impact of the latest E-Discovery legal decisions, rather than the straight forward business proposition of the technology investment.

So what’s wrong with amending the Federal Rules of Civil Procedure to clarify the obligations of parties and perhaps lessen the current burden?  Two primary concerns are the law of unintended consequences and a generalized concern for “justice.”  The debate is nicely summarized at a recent Forbes article (here).

Whatever happens with the effort to amend the Federal Rules, one thing is clear, nothing will happen anytime soon.  Those in the know estimate it will be a minimum of five years before any sort of change would go into effect.  In the meantime, companies like Microsoft will continue to put away data, armies of paralegals and technical consultants will continue to sift it, and employees will continue to be diverted from their core business functions. 
The critical difference between the companies who will drown in the process and those who will move past will not be the amount of data they store.  Realistically, the digital title wave is not something anyone can outrun.  They key is to recognize the special problem data creates for modern litigation and to deal with it head on. How?  (1) Build in defensible and workable procedures now, before the next case; (2) Educate, train, and remind staff on carrying out these procedures; (3) Develop trusted vendor and outside counsel relationships; and (4) Invest in cost appropriate technology.

Easy to say.  Not so easy to do.

Search this Blog

Media Contact

Recent Posts

Jump to Page

Necessary Cookies

Necessary cookies enable core functionality such as security, network management, and accessibility. You may disable these by changing your browser settings, but this may affect how the website functions.

Analytical Cookies

Analytical cookies help us improve our website by collecting and reporting information on its usage. We access and process information from these cookies at an aggregate level.