E-Discovery basics: The EDRM

This week, I will discuss the EDRM.

What is the EDRM?

Well, the letters stand for the Electronic Discovery Reference Model.

Here's some background:

In the early-to-mid 2000s, and especially after the release of the Zubulake decisions, it became clear that the world of electronic discovery needed to agree on an intellectual framework. A common way to talk about e-discovery. A model to which everyone could refer. (A reference model!)

In 2005, some very smart people developed the Electronic Discovery Reference Model, and, courtesy of EDRM.net, here it is in all its glory:

The EDRM identifies six consecutive phases of e-discovery:

• Information Management
• Identification
• Preservation & Collection
• Processing, Review & Analysis
• Production
• Presentation

The model also shows us that as we progress through these phases, from Information Management to Presentation, the volume of data we are dealing with goes down. At the same time, the relevance of this data goes up. (Or, better said: the data becomes more relevant.)

Here are my brief summaries of each of the six parts of the EDRM:

1. Information Management
Every business must manage its information. Even if no litigation is pending, a company must keep track of its data. However, managing information with the EDRM in mind gives the company a big head start in dealing with future litigation.

2. Identification
Once a company and its counsel are aware of the need to conduct electronic discovery on the company's data, they must ask the question: Whose data? Which employees are involved?

3. Preservation & Collection
After the appropriate custodians are identified, counsel must ensure their data is preserved (and not lost or destroyed). Counsel begins to collect this data.

4. Processing, Review & Analysis
Once the data is collected, it is processed - this usually means putting the data into some form of e-discovery review tool. Duplicate copies of the same document are removed. Next, someone (usually an attorney) reviews the data for attorney-client privilege and relevancy to the case. Counsel analyzes the data, to understand what it is and what it means for the case.

5. Production
The data is turned over to the 'other side', or to whomever requested the data.

6. Presentation
The data is shown to a witness in a deposition, or used in a trial setting. At this point, there is relatively little data, but it is very relevant to the case.

This is the EDRM. A model that the industry has agreed upon, so that we understand each other when talking about e-discovery.

Next week: the 2006 changes to the Federal Rules of Civil Procedure.

Aftermath of the Zubulake decision

The Zubulake V decision was released in the summer of 2004.

(For background on Zubulake, see my earlier Zubulake posts here.)

In her decision, Judge Scheindlin created a framework for handling electronic discovery requests. She described what she considered the proper way to respond to requests for documents.

My electronic discovery group looked at the judge's recommended model, and liked what we saw. At the time, there weren't many 'best practices' guides we could rely on, and especially not any that were issued by a judge, and by a judge in our jurisdiction. So, we looked to the Zubulake decision to show us the way.

What is Judge Scheindlin's model?

She recommended the following:

1. A party's counsel should issue a 'litigation hold' to notify employees who may have relevant information that they could not destroy this information. The litigation hold should be kept 'fresh' in the minds of current employees, and be conveyed to new employees.

2. Counsel must talk to the 'key players' in the litigation, to make sure they understand the litigation hold, and to find out whether these employees store relevant information in non-standard or unusual places. If so, this data must also be preserved.

3. Counsel should obtain a copy of the relevant information. Counsel should also ensure backup media is kept safe, either by taking possession of it, or by telling the company's technology professionals that they must keep the relevant backup media separate from other media, and especially not to lose or destroy it.

What did we learn in following the model? Most importantly: Communication Is Key!

- A party's counsel must keep in close touch with the company's IT people, and keep reminding them they must make sure no relevant data is lost or destroyed.

- We also found that counsel must develop a deep understanding of how and where the company's data is stored. If you don't know what to include in a litigation hold, you may overlook vital information. As the defendant in Zubulake found, this is a bad thing.

- Finally, we had to ask key employees if they stored any data in any non-standard way. Different departments at my firm had different rules about storing company data, and it seemed as if each group of employees stored data in its own, unique way. It was a challenge to make sure we didn't overlook any source of relevant data. But it was a fun challenge, and each case was new and never a routine.

While the Zubulake case was the beginning of my discovery of e-discovery, it was only the beginning. In the weeks to come, I'll talk about other influences like the EDRM, the 2006 changes to the Federal Rules of Civil Procedure, and more...

(Post updated 5/28 to generalize names of companies.)

E-Discovery Basics: Zubulake V

The fifth and final decision of the Zubulake case was released in the summer of 2004, and it was a shocker.
(For background on Zubulake, please see my discussion in these four earlier posts.)

Zubulake had an effect on my world - it didn't exactly rock it, but it changed the way many legal departments viewed electronic discovery, including the department where I worked at Bear, Stearns & Co.
Some personal background: in early 2003, I had started working in the litigation group of the Bear Stearns Legal Department, just as the Zubulake decisions were being released. We read the Zubulake decisions as the judge issued them. The fifth Zubulake ruling worried a whole lot of people.
Why?
Well, the "adverse inference instruction" is why - the atomic bomb of the litigation world.
What happened here?

When Zubulake's legal team re-deposed certain UBS employees, as ordered by the judge in Zubulake IV, they found out about many more emails: emails that were improperly deleted and emails that were supposed to have been given to Zubulake almost two years earlier.
Even worse for UBS, the depositions showed that the deleted emails were important to the case, and they had been lost, forever. The back-up tapes that held copies of these emails were gone too.
The depositions also revealed that UBS had other emails important to Zubulake's case, but never turned them over to her.
Zubulake asked the judge for an adverse inference instruction to the jury. For the judge to grant this request, UBS had to have an obligation to preserve relevant evidence which was lost or destroyed, and it had to do so with a "culpable state of mind". In the court's jurisdiction, a "culpable state of mind" included not only willful or reckless behavior, but also negligence (although Zubulake then had to prove that the evidence lost was relevant to her case).

Ultimately, the judge found that UBS counsel did not do everything it should have done to preserve evidence, and that UBS employees also ignored much of the advice given to them by their counsel. Therefore, the judge decided that UBS willfully lost the evidence, and that under the legal standard, the lost information was presumed to be not only relevant to the Zubulake case, but favorable to Zubulake. The jury was instructed of this decision by the judge.

How did the case end? The jury awarded Zubulake almost $30 million in damages for her gender discrimination claim: about $10 million in compensatory damages, and about $20 million in punitive damages. A nice outcome for Zubulake, and not so nice for UBS.

Next week, I'll talk about my reactions to the Zubulake case, and some of the lessons learned...

(Zubulake V is cited as: Zubulake v. UBS Warburg, 229 F.R.D. 422 (S.D.N.Y. 2004))

E-Discovery Basics: Zubulake IV

In the last few weeks, I have been discussing the Zubulake case of 2003 and 2004. Background on the case can be found in my previous installments here, here and here. If you are not familiar with the case, it is worthwhile to review these earlier blog posts.

Picking up the story, a few months after issuing her third decision in the Zubulake case, Judge Scheindlin released her fourth decision.
UBS had been restoring back-up tapes related to the case and found that some tapes were missing.
Also, UBS discovered that emails important to the case had been deleted from the UBS email system, and were now available only on back-up tape. This was a problem, as UBS had told its employees to stop deleting any emails relating to this case.
Zubulake asked the judge to impose sanctions on UBS because of this issue.

The judge decided that UBS had a duty to preserve emails related to Zubulake, and that that this duty started when it seemed very likely that Zubulake would sue UBS.
The next question addressed was: preserve which emails? The judge decided that all documents relevant to the case belonging to "key players" had to be kept by UBS.

Zubulake had requested an "adverse inference instruction" from the judge. If granted, this would end the case, as UBS would have no reasonable option but to settle with Zubulake. An adverse inference instruction is extreme - a judge tells a jury to assume that if one side destroyed evidence, then this evidence must have been very harmful to that side's case.

Judge Scheindlin decided that Zubulake could not show that the lost evidence would have supported her claims, and so she would not issue an adverse inference instruction. However, UBS did have to cover Zubulake's cost for re-deposing some witnesses on issues raised by the missing back-up tapes.

Next week, I'll wrap up my discussion of the Zubulake case.

(Zubulake IV is cited as: Zubulake v. UBS Warburg, 220 F.R.D. 212 (S.D.N.Y. 2003))

E-Discovery Basics: Zubulake III

This week, I briefly continue my discussion of the Zubulake case, one of the most significant electronic discovery case of the 2000s.

Here's a brief recap: Laura Zubulake sued her employer (UBS) on grounds of gender discrimination. She asked for UBS emails relevant to her case. UBS said that emails relevant to the case were available only on back-up tapes, and claimed that it would be expensive to recover these emails. Judge Scheindlin, the federal court judge in the case, instructed UBS to restore a sample of back-up tapes, so that the judge could make a decision on whether to shift the costs from the defendants (UBS) to the plaintiff (Zubulake).

The judge's third decision in the case (known as Zubulake III) was released in July 2003, a couple of months after her first decision.
She applied her cost-shifting analysis and its seven-factor test (discussed in last week's post here), and decided that the plaintiff (Zubulake) would pay 25% of the costs of restoring the back-up tapes, while the defendant (UBS) would pay 75%.

However, it is important to remember that the restoration costs were much smaller than the costs of reviewing the emails to see what was there: UBS spent a lot of money to pay attorneys to look over the documents restored from the back-up tapes. The judge said that because UBS had control over these costs, it was responsible for all of them.

Zubulake III is interesting because the judge decided to shift some of the restoration costs to the plaintiff, but not the costs of reviewing the restored emails.

Next week, I'll continue my discussion of the case by looking at Zubulake IV, and hopefully, the last in the series, Zubulake V!

(Zubulake III is cited as: Zubulake v. UBS Warburg, 216 F.R.D. 280 (S.D.N.Y. 2003))