The False Economy Of In-House ESI Processing

Tuesday, March 31, 2009 - 01:00

I like steak. I've become pretty good at grilling my own. Of course, I have to buy the steak and prepare the grill and season the meat and light the charcoal and build the fire and clean up the mess and make the side dishes, but the results can be quite tasty (if time-consuming). Of course, being an amateur, I also run a high risk of ruining the steak.

So, when I want a perfect steak, I go to Morton's. It may seem a more pricey option at first glance, but their professional grillmasters cook the steak, make the sides, serve it up, and clean up the mess. And if there is a problem with the steak, they fix it.

Could I make a steak like Morton's? Probablyif I invest in a broiler like theirs, find a reliable source of prime beef, experiment until I discover their seasoning mix and spend countless hours (and piles of red meat) practicing before I'd be able to prepare steaks as well as they do. Is it worth that much trouble and expense (especially given the limited amount of steak I'll be eating)? Probably not.

My point? There is a current push among some commentators for companies to bring electronically stored information (ESI) processing in-house, claiming that the cost savings will justify the expense. Certainly, there are aspects of managing ESI that can - that should - be handled in-house. But when it comes to processing and hosting, do any companies other than the largest need that kind of steak-grilling capacityand what is the true expense?

Bringing It In Is Rarely Practical

A couple of recent white papers by ESI consultants George Socha1and Brian Babineau2appear from their titles to suggest that companies should try to bring their ESI processing in-house. A closer examination of both papers, however, makes the costs and hazards of doing so rather obvious, and even Socha and Babineau concede these pitfalls.

Some of Socha's observations are unassailable. "The root of the problems discussed here - and the main reason pre-trial costs go through the roof - is that most organizations have no overriding concept or supporting structure in place to define and manage the relevant information that could be vital to their defense during litigation proceedings."3

However, he also warns that many organizations have unrealistic expectations about their ability to manage ESI in-house. They want a comprehensive solution system, but don't consider their lack of knowledge about such a system's price/value relationship, their unfamiliarity with the full scope of relevant regulations, or the hazards of procedural missteps.4

Since no outside counsel or service bureau can know how your organization stores e-mails, files, archives and backups, being able to identify, preserve and even collect potentially responsive data - all critical to the success of the litigation process - should require the involvement of in-house litigation support and IT personnel. Certainly, the in-house team must be in charge of preserving ESI and enforcing litigation holds. If the data is not to be collected by the IT team, they should certainly work with the ESI collection specialist to root out all potential sources of responsive information.

However, the stakes for these activities are incredibly high, and errors made during these steps can be irreparable. An example of this is the recent Louis Vuitton case,5which I analyzed in this space in February.6Louis Vuitton (LV) needed to extract their e-mails from a Kana Oracle database. Although LV's in-house IT team was not expert in Kana or Oracle, LV relied on them to search for potentially responsive data; the IT team claimed to have found none. The $15,000 savings that LV realized by refusing to hire experts was dwarfed by the adverse inference sanction and attorney's fees with which the court slapped them.

More Data, More Challenges

The larger the company, the harder it can be to manage ESI in-house. "[E]lectronic discovery has forced organizations to outsource most of the process to legal service providers," writes Babineau, "because the amount of requests, combined with the volume of data to collect, preserve, process, and analyze, is too much for internal IT departments to handle."7The expense becomes outrageous when companies do not have formal internal processes to gather data, so they simply send everything for processing that they believe could be relevant.8

Consider the recent D.C. Circuit decision in In re Fannie Mae Securities Litigation. 9A non-party to the litigation, the Office of Federal Housing Enterprise Oversight (OFHEO) was forced to spend $6 million - nine percent of their total annual operating budget - to comply with a subpoena for electronic documents. While the court's reasoning may be wildly draconian,10the overriding lesson was that OFHEO's trial counsel (who was not conversant with ESI) had agreed to overbroad production requests and search terms, and was therefore held to that agreement, no matter the cost or effect on the litigation11- or the fact that the agreement was based in ignorance.

On the other hand, there is no need to hand the keys to the store over to outside providers for all ESI management tasks. It's likely that not all data will require processing; and if it doesn't require processing, it doesn't require collecting. Outside counsel can help create sampling techniques for the in-house team to use to target the potentially responsive ESI,12and outside counsel, in turn, can (ideally) work with opposing counsel to settle on keyword search terms that will cut the size of the collection even further.

Parceling out ESI to multiple outside counsel and service providers creates a risk that the organization can lose control of their ESI.13Having the in-house team coordinate providers' activities internally will help to mitigate the risks of having "too many cooks."

Saving Money By Spending Money

Some argue that if the company does not avail itself of a centrally controlled repository, there is a danger of inconsistent processing and production of ESI. Clearly, having one place to park the data is preferable and will save quite a bit of cash. But should repository hosting be brought in-house?

Most companies are fiercely protective - and rightly so - of their internal computer networks. Among the triple threat of hackers, corporate espionage and disgruntled former employees, companies do not want to poke holes in their firewall by allowing any outsiders access to data on their internal servers, as those holes can be easily exploited.

Unfortunately, it's that same need to preserve the sanctity of confidential data that renders organizations poorly suited to host document repositories in-house. Service provider architectures are designed to permit external parties to access data (if they have appropriate permissions), while corporate firewalls are designed to protect all outsiders from getting in.14Outsourcing the repository hosting, therefore, resolves multiple issues (and avoids multiple hassles) of document management, security and infrastructure maintenance. It can also be substantially less expensive to take advantage of a vendor's economy of scale than to try to reinvent the wheel internally. The financial savings of outsourcing will be immediate, as companies will not need to dip into their capital budgets to fund the repository hosting, and the time cost saved by avoiding staffing and training issues is likely to be incalculable.

Similarly, when it comes to advanced culling, document processing and imaging, the flexibility provided by use of an outside service bureau far exceeds the false savings of trying to bring such activities in-house. Complete management of the ESI review and production process would require a large suite of different commercial software products15(presuming off-the-shelf software sufficiently addresses the company's requirements), each one of which requires training and a steep learning curve. Any company that chooses this route would need to pay to train multiple members of the in-house team and would need plenty of computer hardware to handle these processor- and memory-intensive tasks. The in-house IT workload (and budget) would swell in order to keep the new machines up and running, and the existing litigation support team would need to find the time in their busy workdays to assume these extra processing tasks.

Oryou can outsource them to a service bureau that already has the infrastructure and expertise in place, work with them to ensure that your exact expectations for processing and production are met, and pay as you go. It doesn't get much simpler than that.


Even commentators who argue for bringing ESI management activities in-house concede that only about one percent of organizations are prepared for full-scale e-discovery activities.16For the rest, there is no question that developing a process to keep tabs on where internal data is located, and how to access it quickly, is key. For a company with the resources to spare, searching and collecting relevant data is a task that is best handled in-house. Beyond that, however, there is no substitute for the expertise, experience, and infrastructure that already exists with outside counsel and service providers.

Judicious sharing of ESI management functions with outside providers can result in tremendous cost savings over both trying to manage everything in-house and trying to outsource everything. With the money you save by bringing some functions in-house and leaving others to the pros, you can treat your lit support and IT teams out to a nice dinner. (I hear they make a pretty good steak at Morton's.) 1 George J. Socha, Jr., Esq., "Bringing e-Discovery in-house: risks and rewards" (Feb. 2009), available at ediscovery_in-house.pdf

2 Brian Babineau, "Getting Control of Electronic Discovery", Enterprise Strategy Group (Dec. 2008), available at whitepaper3742 (registration required).

3 Socha, p. 3.


5Louis Vuitton Malletier v. Dooney & Burke, Inc., 2006 U.S. Dist. LEXIS 87096 (S.D.N.Y. Nov. 30, 2006).

6 Gary Wiener, "An In-House Review of Outside Review," The Metropolitan Corporate Counsel (Feb. 2009), p. 40.

7Babineau, supra note 2, p. 1.

8Id. at 4.

9 552 F.3d 814 (D.C. Cir. 2009).

10See Ralph Losey, "D.C. Appeals Court Affirms Order Requiring a Non-Party to Spend $6 Million, 9% of its Total Annual Budget, to Comply with an e-Discovery Subpoena" (Jan. 9, 2009), available from

11See Fannie Mae, supra note 7, at *7 ("Whatever the merits of these claims, OFHEO abandoned them by entering into the stipulated order.Having stipulated to a schedule for complying with the subpoenas, OFHEO can hardly complain now about being held to its agreement.").

12See Gary Wiener, "Breaking Through Babel Without Breaking The Bank," The Metropolitan Corporate Counsel (Dec. 2008), p. 46.

13 Socha, supra note 1, pp. 6-7.

14 Babineau, supra note 6, p. 7.

15 Babineau conducted a case study of a large corporation, and listed the nine commercial software packages they had purchased in order to bring all aspects of ESI handling in-house.

16Id. at 5. Despite promoting the shifting of ESI in-house, Babineau admitted that the corporation was unable to determine the amount of return on investment that these software packages, plus all attendant expenses, actually saved the company.

Gary Wiener is the Director of Litigation Services for Liquid Litigation Management. A former trial lawyer, he advises clients in the effective use of technology to streamline their litigation management practices.

Please email the author at with questions about this article.