As an e-discovery lawyer at a large global law firm, I field many questions. The questions often repeat. When facing the same Frequently Asked Question (FAQ) for the umpteenth time, snappy retorts may spring to mind, usually in the form of a question back. It is important to remind oneself, however, that it is probably just the first time each particular person is posing the question. The questioner deserves a polite and thoughtful answer.

Here are three FAQs in e-discovery, the replies to avoid and some more appropriate responses:

Q: My client has a situation that may lead to litigation or a government investigation. Should they implement a legal hold?

Avoid: That depends. Does the client enjoy responding to sanctions motions or prefer not to?

A: In most U.S. jurisdictions, preservation obligations attach as soon as litigation or an investigation is “reasonably anticipated.” If the client is asking counsel the above question, it probably is reasonably anticipating litigation.

Of course, different courts have interpreted the standard in different ways, ranging from when litigation is “imminent” to when the potential for litigation is merely “foreseeable.” However, there are at least three factors that should lead most companies to err on the side of implementing a legal hold whenever in doubt:

1. It is usually not possible to know in advance what judge(s) or other official(s) may later be evaluating the company's compliance with its preservation obligations, or how narrowly or broadly they will interpret and apply the standard.
2. It typically only becomes an issue in retrospect, in cases that actually proceed to litigation or an investigation. Hindsight is always 20/20.
3. The cost of implementing a legal hold is usually far less than the potential liability for failing to issue a legal hold should a judge or other official later decide that a hold was warranted.

There are occasional exceptions, where companies can reasonably argue that, at the time of the decision, it appeared that the potential for litigation was relatively small and the cost of implementing a hold exceeded any potential benefit. However, that approach can be risky if an adverse party has a basis to make contrary arguments. At a minimum, companies need to be prepared to document and defend the bases for such decisions in light of the prevailing legal standards.

Q: I've heard that computers can now replace human beings to screen documents for relevance and privilege using predictive coding technology. Can we use predictive coding on my case?

Avoid: If the computer is so smart, why don't you ask it?

A: There are a number of exciting new technologies currently being developed and refined to reduce time and costs in document review. We are not, however, at a point where computers can entirely replace human lawyers in performing document review. Predictive coding software needs to be trained by human decisions on any given population of documents and a large number of those documents, typically across many sample sets, need to be reviewed before the software can achieve acceptable accuracy levels. Even so, most developers of predictive coding software are not recommending reliance on the software for final privilege determinations, as those determinations sometimes require complex analyses, and mistakes can have serious negative consequences.

To date, predictive coding has been approved for use in only two published court cases. In one of those two cases, Da Silva Moore v Publicis Groupe & MSL Group, both parties agreed to its use. Even in that case, the plaintiff then objected to the protocol approved by the court. In the other case, Global Aerospace Inc. v. Landow Aviation, the court allowed the use of predictive coding over the plaintiff's objection, but included little discussion or analysis in the opinion. There is no case, as yet, where predictive coding technology actually has been applied and found to achieve acceptable accuracy levels, so lawyers and technologists are closely following the cases where the technology is now being tried.

The bottom line is that predictive coding technology holds a great deal of promise, but there are hurdles yet to overcome, and it would be risky to rely on predictive coding technology as a substitute for human lawyer review absent the agreement of adverse parties or judicial approval to use the technology.

Finally, perhaps the most common FAQ, and my personal favorite:

Q: We're handling a new matter where there will be a lot of e-discovery. We don't know the volume yet but we have to prepare a budget. How much will it cost?

Avoid: How much you got?

A: Many of our firm's 60-plus e-discovery attorneys have been doing this kind of work for a long time. We have the data to price by the hour, by the document or even by the gigabyte.

Until we know the volume of information likely to require review, we won't be able to provide an accurate cost estimate. The good news is that the cost savings that we are now able to achieve through advanced technology, optimal workflows and low-cost but experienced e-discovery attorneys is likely to make that cost estimate far more palatable than it otherwise would be.

We will be glad to try to help you come up with an estimate, however, if you first just answer a few of our FAQs!

As an e-discovery lawyer at a large global law firm, I field many questions. The questions often repeat. When facing the same Frequently Asked Question (FAQ) for the umpteenth time, snappy retorts may spring to mind, usually in the form of a question back. It is important to remind oneself, however, that it is probably just the first time each particular person is posing the question. The questioner deserves a polite and thoughtful answer.

Here are three FAQs in e-discovery, the replies to avoid and some more appropriate responses:

Q: My client has a situation that may lead to litigation or a government investigation. Should they implement a legal hold?

Avoid: That depends. Does the client enjoy responding to sanctions motions or prefer not to?

A: In most U.S. jurisdictions, preservation obligations attach as soon as litigation or an investigation is “reasonably anticipated.” If the client is asking counsel the above question, it probably is reasonably anticipating litigation.

Of course, different courts have interpreted the standard in different ways, ranging from when litigation is “imminent” to when the potential for litigation is merely “foreseeable.” However, there are at least three factors that should lead most companies to err on the side of implementing a legal hold whenever in doubt:

1. It is usually not possible to know in advance what judge(s) or other official(s) may later be evaluating the company's compliance with its preservation obligations, or how narrowly or broadly they will interpret and apply the standard.
2. It typically only becomes an issue in retrospect, in cases that actually proceed to litigation or an investigation. Hindsight is always 20/20.
3. The cost of implementing a legal hold is usually far less than the potential liability for failing to issue a legal hold should a judge or other official later decide that a hold was warranted.

There are occasional exceptions, where companies can reasonably argue that, at the time of the decision, it appeared that the potential for litigation was relatively small and the cost of implementing a hold exceeded any potential benefit. However, that approach can be risky if an adverse party has a basis to make contrary arguments. At a minimum, companies need to be prepared to document and defend the bases for such decisions in light of the prevailing legal standards.

Q: I've heard that computers can now replace human beings to screen documents for relevance and privilege using predictive coding technology. Can we use predictive coding on my case?

Avoid: If the computer is so smart, why don't you ask it?

A: There are a number of exciting new technologies currently being developed and refined to reduce time and costs in document review. We are not, however, at a point where computers can entirely replace human lawyers in performing document review. Predictive coding software needs to be trained by human decisions on any given population of documents and a large number of those documents, typically across many sample sets, need to be reviewed before the software can achieve acceptable accuracy levels. Even so, most developers of predictive coding software are not recommending reliance on the software for final privilege determinations, as those determinations sometimes require complex analyses, and mistakes can have serious negative consequences.

To date, predictive coding has been approved for use in only two published court cases. In one of those two cases, Da Silva Moore v Publicis Groupe & MSL Group, both parties agreed to its use. Even in that case, the plaintiff then objected to the protocol approved by the court. In the other case, Global Aerospace Inc. v. Landow Aviation, the court allowed the use of predictive coding over the plaintiff's objection, but included little discussion or analysis in the opinion. There is no case, as yet, where predictive coding technology actually has been applied and found to achieve acceptable accuracy levels, so lawyers and technologists are closely following the cases where the technology is now being tried.

The bottom line is that predictive coding technology holds a great deal of promise, but there are hurdles yet to overcome, and it would be risky to rely on predictive coding technology as a substitute for human lawyer review absent the agreement of adverse parties or judicial approval to use the technology.

Finally, perhaps the most common FAQ, and my personal favorite:

Q: We're handling a new matter where there will be a lot of e-discovery. We don't know the volume yet but we have to prepare a budget. How much will it cost?

Avoid: How much you got?

A: Many of our firm's 60-plus e-discovery attorneys have been doing this kind of work for a long time. We have the data to price by the hour, by the document or even by the gigabyte.

Until we know the volume of information likely to require review, we won't be able to provide an accurate cost estimate. The good news is that the cost savings that we are now able to achieve through advanced technology, optimal workflows and low-cost but experienced e-discovery attorneys is likely to make that cost estimate far more palatable than it otherwise would be.

We will be glad to try to help you come up with an estimate, however, if you first just answer a few of our FAQs!