ECA And Over-Collection Solutions In E-Discovery

Monday, November 2, 2009 - 01:00

Editor's Note: Benton Armstrong and Andy Ruckman are principals in Deloitte Financial Advisory Services LLP (Deloitte FAS). Andy leads the Analytic and Forensic Technology (AFT) practice in the Southeast and Benton is the practice's global and West Coast leader. Deloitte's practice consults with global corporate clients and law firms regarding approaches to addressing technology and process challenges related to e-discovery, record retention, and enterprise information management.

Benton and Andy were involved in developing Deloitte FAS's White Paper entitled Early Case Assessment: Finding Value Beyond Your Next Lawsuit. Because the conclusions in the White Paper are of such great importance, we asked to interview them. Both Benton and Andy participated in answering each of my questions.

Editor: Tell us about the costly results of the repeated export of information in conventional e-discovery systems.

Armstrong/Ruckman: The volume and complexity of electronic data that is being stored presents challenges and great opportunities to reduce e-discovery costs and to mine that data for valuable business information. Today conventional e-discovery systems often require collecting large volumes of data from data stores throughout the organization for export and processing. The costs can be immense - one of our clients had hundreds of thousands of SharePoint sites within its organization. When they did a collection that touched a given department, they felt compelled to pull data from every SharePoint site to which anybody in that department had access. On a recurring basis, they were collecting enormous amounts of data for culling and review, knowing full well that the vast majority of it would not be relevant to the matter on which they were working, but they were still having to either handle that data internally or pay an external vendor to do so.

Even if that data is programmatically culled to remove irrelevant data, you still have to go through the collection process, and you may get large numbers of false positives. I don't think Andy or I have ever worked on a case where the initial set of keyword search terms that was devised by the client, counsel and consultants would not have been revised based on the knowledge we gained by actually reviewing the data.

Most of our clients incur significant internal costs around collection while having processing done externally. Many companies that have invested in an early case assessment solution use it to cull the data down to what is likely to be relevant and hand that subset to a vendor to process, with the review and production being handled by outside lawyers.

The biggest piece of the cost for almost every client we have worked with has been the document review piece - paying attorneys to put eyeballs on documents.

Editor: How can this problem be solved?

Armstrong/Ruckman: There are appliance solutions on one end of the spectrum that have a very defined mission - intelligently culling the data you have already collected. At the other end you have enterprise content management systems, which control how data is created, stored, and retired - how we retire it, when it's no longer of use to the company - many of which have e-Discovery modules which address the processing and review of data within those systems.

Some companies are truly committed to managing that data. They have an information management or records management function within the organization that actually has some clout and works closely with IT, legal, risk management, compliance, finance and other stakeholders within the organization to manage and use their data effectively.

In the middle you have hybrid solutions that have information management capabilities, but which should not be confused with full enterprise content management solutions. Hybrid solutions point at your data stores as they sit within the organization, including e-mail archives, collaboration sites and any other significant data storage you have within the organization and are able to index those systems as documents are created and altered to provide the company with an ongoing index - so you are able to search for possibly relevant data where it sits instead of having to go through a very cumbersome and costly collection exercise.

Editor: What is driving companies to explore hybrid or full enterprise content solutions?

Armstrong/Ruckman: Many companies want to do a better job of controlling their e-discovery costs because they're getting a lot of pressure from finance and from the CEO to find ways to manage their litigation spend more effectively, given the current economic environment. To assess how much a particular solution is actually going to help you, you have a clear understanding of what you're spending for review, of the cost model that you're using, of the numbers of documents you're handing to reviewers and of how that work is getting done.

Over time the mounting costs of doing e-discovery the old fashioned way may force more and more companies to graduate to a solution that enables them to index and mine the data while it's sitting in place in the organization rather than collecting it and shipping it off to another location. However, the initial costs of a hybrid solution are currently high and can run into seven figures.

Editor: How can management be convinced to spend such a large amount in today's environment?

Armstrong/Ruckman: Assuming I was the head of litigation in a Fortune 500 company that had a consistently large litigation portfolio, and we were spending $10 million a year on discovery and found out that we were spending a lot of money to have outside counsel review nonresponsive documents, I would want to stop that from happening.

If I was spending that kind of money on an ongoing basis, it would be very valuable to develop an ROI calculation that would support at least a hybrid solution. I personally think we've got a lot of companies that are just kind of dipping their toes into this space right now, and they're investing in appliance solutions because they are quick and easy to implement, and do not require a large upfront investment in costs or IT infrastructure. Yet, general counsel realize that the cumulative amount is high and that a large part of that cost represents the fact that you have to re-collect data every time a big new case is brought against your company.

There are also risks in not going the hybrid route. Every time you collect data for e-discovery you are preserving a corpus of data that sits out there until that litigation goes away and that data is potentially discoverable on any future litigation you may have. One of the challenges we find some of our clients facing is they get a new matter and in addition to searching all the data stores we would normally search, they say here's a matter that we did for the SEC a year ago, it's still ongoing, we collected data from 47 custodians and we probably have to search this data as well. So the idea of having technology that enables me to search for indexed documents and preserve them in place becomes far more attractive than searching my business de novo every time I have a new matter.

Think of the advantages such a system offers. Before a meet-and-confer we would have adequate time to do some internal analysis of that data, we would have seen where we think that data takes us, and we would have an idea of the volume of data involved. The hybrid approach would put us in a position where we could limit the scope of discovery because we could show, for instance, that out of 100 custodians requested by opposing counsel, only 25 were involved in the issue being litigated.

We're seeing few companies buy into a full enterprise content management solution on the basis of e-discovery savings alone. However, legal could make a convincing argument in partnership with IT that there could be a substantial ROI from a litigation perspective in better organizing corporate data at the enterprise level. There are all kinds of IT implications to having technology within your firewall that's pointing at all your big data stores, and that certainly can justify spending a significant amount to make that happen.

Editor: What is the role of an ECA tool?

Armstrong/Ruckman: ECA tools can reduce the spend on review by greatly reducing the number of documents that make it through to the attorney-review phase. Let me give you an example. Traditionally in e-discovery, we might have 100,000 documents of which 20,000 end up being responsive, and we have to review all 100,000 of those. With our ECA tool we might reduce to 40,000 the documents that need to be reviewed and only 20,000 of them are responsive, so we might be able, using ECA, to cull out about 60 percent of the documents you would traditionally have to review.

For those companies that have adopted the hybrid approach the savings could be even greater because bots can actually crawl the network and create an index of all the documents and the emails that are out there without having to collect the data. When litigation arises, you can just run searches for possibly relevant data and then apply the culling tools. The fact that the information is indexed and can be searched in place means that you can quickly get a good handle on the documents at the beginning of a matter. This would give you the ability to do some analysis of the data on your side of the matter and see how much information you have to support your case. You would have the information upfront to make a settle-versus-litigate decision and to properly scope discovery as part of the meet-and-confer process. You would be dealing more often with documents that are highly likely to be relevant as opposed to possibly relevant. Yes, you still have additional costs upfront, but you have slimmed down that data so much that you could experience substantial savings by not paying attorneys to look at documents that never had a shot of being relevant to the matter to begin with. Because you have done a better job of culling information, you did not over-collect to begin with and you did a much more robust job of searching the information and analyzing it before it ever got to the point where somebody had to process it or put eyeballs to it.

We think it's going to take some time to evolve to the full enterprise content management level, but we believe that a lot of our clients need to get to a point where they have at least a hybrid solution in place for e-discovery, and that's going to require a couple of things to happen. It's going to require those companies to make their technologies more robust, demonstrate that they've got good connections to most of the typical data sources within their companies and do a better job of explaining the ROI. It's going to require the company to take a cross-functional approach to solving this problem. If it's just Legal attempting to attack this issue, it's not as likely to happen because to make these solutions work you have got to have your records management function, IT, finance, compliance and everybody else who would benefit on board because it's a significant investment and one that touches a lot of different departments even though the primary beneficiary is the legal department.

This publication contains general information only, and Deloitte Financial Advisory Services LLP and its subsidiaries is not, by means of this publication, rendering accounting, business, financial, investment, legal, tax, or other professional advice or services. This publication is not a substitute for such professional advice or services, nor should it be used as a basis for any decision or action that may affect your business. Before making any decision or taking any action that may affect your business, you should consult a qualified professional advisor. None of Deloitte Financial Advisory Services LLP or its affiliates or related entities shall be responsible for any loss sustained by any person who relies on this publication.

Please email the interviewees at or with questions about this interview and visit to download their white paper.