The credit crisis is sending the economy into a tailspin. Meanwhile, legal professionals await the litigation fallout from the issuance of so many bad mortgages.
It is this anticipated abundance of litigation that experts believe could wake a sleeping giant, known as structured data, that might forever change the scope of e-discovery. Structured data is information that is managed and stored in an organized system, typically a database, as opposed to unstructured data, such as e-mails and Word documents.
“You are going to see a lot of litigation from shareholders and investors arising from the credit crisis, and a great deal of information in these matters is going to be from structured data sets,” says John Montana, general counsel of The Pelligroup Inc., a records management consultancy.
It’s likely that once structured data is thrown into the e-discovery limelight, it’s there to stay. As plaintiffs’ firms learn to request this kind of information, corporations better be prepared to produce it.
“Historically, the cost benefit of requesting and producing structured data wasn’t understood,” says Erik Post, senior managing director at FTI Consulting Inc., which has a team of professionals that consult on producing structured data. “Now the plaintiffs’ bar realizes a lot of information lies in these systems. In-house and outside counsel need to understand how these systems are set up and how to review them.”
No company is exempt from the structured data issue. Even a one-person company is subject to its difficulties if the owner uses a program such as Quickbooks to handle his accounting. That’s because structured data is any information that is stored and managed in a database.
Countless systems, both proprietary and off-the-shelf, fit this definition. For example, HR, accounting, content management, customer relations and inventory control systems all fall within the realm of structured data. In fact, according to the Data Warehousing Institute, an educational organization that caters to business and IT professionals, 47 percent of all enterprise information is structured, with 22 percent being semistructured data that exhibits both structured and unstructured qualities such as spreadsheets. However, when it comes to e-discovery, structured data is all but absent from discussion.
“I’d say 80 percent or more of discovery is devoted to e-mails and attachments, while the remaining 20 percent focuses on unstructured information stored on hard drives and servers,” says George Socha, e-discovery consultant and co-founder of the Electronic Discovery Reference Model (EDRM) project, the think tank that created the standard workflow chart ubiquitous among e-discovery professionals.
The legal profession has shied away from dealing with structured data in part because lawyers are still getting their arms around unstructured data. But the overriding reason why structured data has had little mention in e-discovery circles is due to one thing–fear.
Structured data poses numerous difficulties that unstructured data does not.
First, databases can be huge. Consider a database of employees at Microsoft Corp. The software company employs about 79,000 people worldwide. If within the database each employee had five fields attributed to him, that would equal 395,000 data fields that would need to be collected, processed and reviewed to respond to a discovery request.
There’s also the issue of retrieval. Discovery requests often require the company to extract certain types of information from the system that aren’t requested during the ordinary course of business. “You can’t just run your standard business-function reporting and generate material for discovery,” Montana says.
Consider accounting software. Normally a company can run a standard report using the software to see how the business is doing financially. But for a discovery request, a requesting party might need to know whether a specific employee played a role in a number of transactions. That’s not necessarily data the software can generate on its own and may necessitate the use of forensics.
Finally, reviewing structured data takes a certain degree of knowledge that an ordinary contract attorney doesn’t possess. This means special experts, either recruited from within the company or outside, need to be brought in to make sense of the data and determine what needs to be produced.
“Instead of trying to find the needle in the haystack as with e-mails, reviewing structured data requires complex analysis to identify what information within these large datasets is important,” Post says.
Fortunately, in-house counsel don’t always have to honor requests for structured data. Under the amended Federal Rules of Civil Procedure, requesting parties only have a right to reasonably accessible data. So when receiving a request for structured information, counsel must sit down with technical experts to assess the burden the request creates.
“In-house counsel need to work with outside counsel to determine whether the request is actually germane to the dispute,” Montana says. “If it is unreasonably burdensome, you should negotiate it down to a reasonable scope or talk to the judge and make your case.”
As plaintiffs’ attorneys and in-house counsel alike become more familiar with structured data, the meet-and-confer process will likely lead to a much less painful production. This learning experience will likely come out of the credit crisis fallout, where contents of e-mails matter less than the numbers in digital ledgers.
“With unstructured data, you might find an e-mail that tells the tale of what happened,” Socha says. “But if you really want to see the corporate DNA, you better be looking for structured data.”