Episode 105: How the Language of Your ESI Protocol Can Force Your Hand

In this week’s episode of ACEDS’ Case of the Week, where the matter involves a breach of contract action with both parties seeking multimillion-dollar judgments. This episode focuses on an objection to the Magistrate’s discovery order that compelled the plaintiff to review documents responsive to search term hits prior to production based on the language of the ESI protocol. Kelly Twigger discusses the importance of ESI protocol language and how parties can be held to what is in that document. She also emphasizes the need to deal strategically with review costs by reducing the number of documents or engaging with technology. 

Don’t miss out on this important episode!

Introduction

Hi, and welcome to this week’s episode of the Case of the Week series, published in partnership with ACEDS. This week marks our 105th episode of the Case of the Week series, and thanks so much for joining me. My name is Kelly Twigger. I am the Founder and CEO of eDiscovery Assistant, as well as the Principal at ESI Attorneys.

We’re going to dive directly into this week’s case, which comes to us from McCormick & Co. v. Ryder Integrated Logistics, Inc. This decision is from March 8th of 2023, so just a little bit more than a month old. It comes to us from United States District Judge James Bredar. Judge Bredar has 19 decisions in our eDiscovery Assistant database.

As always, we tag our cases with issues relevant to eDiscovery. Those issues for this week include failure to preserve, manner of production, ESI protocol, legal hold, proportionality, and search terms.

Background

What are the facts underlying this case? Well, we’re before the Court on a breach of contract action in which both parties brought separate cases that were then consolidated before the court. Both parties seek multimillion-dollar judgments; McCormick seeks $3 million in damages, and Ryder claims $2 million in damages for this breach of contract.

Facts

What are the facts that come to this decision today? Well, the parties exchanged requests for production in March of 2022, so a little bit more than a year ago, and then cooperated to develop a plan for responding to those requests as they involved extensive amounts of ESI.

As part of that process, the parties agreed to search terms to run across the ESI and to a joint protocol for the discovery of ESI. That protocol, which they submitted to the trial court — to the magistrate judge — included a subsection titled “No Presumption of Responsiveness.” Under that heading, the following language was included.

A party’s obligation to conduct a reasonable search for documents in response to discovery requests, shall be deemed to be satisfied by reviewing documents that are captured by utilizing the methodology provided for in this protocol.

and

The fact that a document is captured by a search pursuant to this protocol does not mean that such document is responsive to a discovery request or otherwise relevant to this litigation, and the parties may exclude such non-responsive documents from production.

That’s the language that we’re dealing with on the ESI protocol. That’s really important to this decision, and it’s going to be really important to our takeaways.

During the same time that this protocol was being proposed, McCormick, the plaintiff here, learned that despite a litigation hold having been in place, custodial files for a custodian named Willa Blasingame, who was a key custodian, were deleted when she left the company. To account for that loss of data, McCormick notified Ryder (so they notified the other side) and searched the ESI of six additional custodians as well as running a search for the name Blasingame across all of McCormick’s ESI. McCormick then ran the agreed upon search terms across the additional custodians and search data for Blasingame and included that total number of documents in its entirety of its pool that we’re discussing here.

After the search terms were run on both sides’ documents, Ryder had approximately 37,500 documents, which rounded out to about 50,000 documents with families. McCormick had roughly 49,000 documents which amounted to about 68,000 documents with families. Not too different despite the additional custodians.

According to Ryder, of those numbers, approximately 20% of the hits for McCormick are due to the additional searches to address the loss of Blasingame’s data. There’s nothing else in the decision that talks about how Ryder came to that information. If that’s true, subtracting that 20% of McCormick’s number of documents to about 38,000, so just under 39,000, it means that they have about 1,200 more documents than the total of Ryder’s search results, even with the additional six custodians that they searched.

The case went to the Magistrate Judge, who granted the motion to the extent that it sought for the ESI protocol to be entered as an order of the Court. The Magistrate denied the motion that requested a declaration that the ESI protocol did not require a manual review of the documents. The Magistrate Judge found that the ESI protocol expressly contemplated a manual review of the documents based on the language that I quoted to you previously, and rejected McCormick’s assertion that a proposed cost of $240,000 for review was not proportional to the needs of the case.

We are now before the court on McCormick’s objection to the Magistrate’s ruling on that discovery order. McCormick essentially wants the District Court to overrule the Magistrate Judge’s order that they are not required to conduct a manual review of those documents prior to production.

Interestingly, this raises the same issue that we saw on last week’s Case of the Week in the SinglePoint decision — basically, whether a party has an obligation to review search term result hits for relevance prior to production. This is also a situation that we’ve seen previously on Case of the Week.

Recall that in SinglePoint from last week, there was no ESI protocol language that the court relied on in ordering a review of all the search term hits for responsiveness. In that case, the producing party argued that a search term was overbroad and that it should not be required to review those non-responsive documents that the term hit on based on relevance grounds.

Citing case law, Judge Teilborg in SinglePoint found that the plaintiff’s relevance objection was overruled and that according to case law, the fact that running a search term that may hit on non-responsive documents does not waive relevance.

That’s a different situation than we have here with McCormick and Ryder, because McCormick is arguing that they simply don’t have an obligation to review documents post the search term review under the protocol. Their argument is, hey, the protocol requires that we run search terms. We ran the search terms. That’s all we have to do. Ryder disputes that and says, you still have to review the documents for relevance and produce to us only what is relevant.

We’ve seen this situation in other cases where parties are arguing about either the cost of manual review. Let’s dive into what’s key here, which is the ESI protocol language.

Analysis

The Court starts its analysis by looking at the standard of review. The standard of review over a magistrate judge’s ruling is clearly erroneous, which means that the District Court is required to determine whether the Magistrate Judge’s findings here are reasonable and supported by the evidence. Again, as we’ve discussed on Case of the Week, under Rule 72, this is a high bar to be able to get over.

The Court then addresses the language of the ESI protocol that formed the basis of the magistrate judge’s ruling, which I cited to you earlier. Judge Bredar looks at the language and found that the Magistrate’s order stating that the ESI protocol contemplated a document-by-document review was not erroneous.

He cited specifically to the protocol language that states, “A party’s obligation to conduct a reasonable search for documents in response to discovery request shall be deemed to be satisfied by reviewing documents that are captured by utilizing the methodology provided for in this protocol.”

McCormick argued that the language states that a party satisfies the protocol by running the search terms. It argues that running the search term and then conducting a manual review of each document requires two levels of review where the protocol only provides for one.

Well, Judge Bredar rejects that analysis and points again to the language of the protocol, which states that “A party’s obligation to conduct a reasonable search for documents in response to discovery request shall be deemed to be satisfied by reviewing documents that are captured by utilizing the methodology provided for in this protocol.”

Reviewing documents, that’s what we’re focused on here. The Court states that the language says a party discharges its discovery obligation by reviewing documents captured by search terms and notes that “It defies logic, that the ESI protocol would provide for a party to review documents but not to exclude irrelevant documents identified in that review.”

The Court also rejected McCormick’s argument that the Magistrate did not fully consider whether the cost of review was proportional to the needs of the case, noting specifically that McCormick’s failure to preserve data for a key custodian is what created the mess in the first place.

Of note is this: the Court states that the Magistrate Judge who initially denied the motion, Judge Copperthite, did not conclude that there exists a per se duty to conduct a manual review. The Magistrate concluded that in this case, the ESI protocol required the manual review, and that was sufficient for the District Judge to uphold the Magistrate judge’s ruling.

What is key on this issue of whether manual review is required is this quote from the Court:

Of course, in the absence of the ESI protocol, Judge Copperthite may (or may not) have concluded differently, and the cases cited by Judge Copperthite and the parties reflect that a manual review may or may not be appropriate, depending on the circumstances.

With that, the District Court overruled McCormick’s objections to the Magistrate’s ruling and required a manual overview of the documents prior to production.

Takeaways

Okay, so let’s move to our takeaways. If you’re here with me regularly on Case of the Week, you know what I’m going to say right now: the language of your ESI protocol is critical, and you will be held to what is in that document. Make sure that you know what you’re doing and agreeing to when you draft it.

McCormick easily spent $60,000 on motions here that were never going to succeed. Ironically, they asked the Magistrate Judge to enter the ESI protocol with the language in it requiring manual review. At the exact same time, they asked not to have to do manual review. If I was doing this analysis in writing, I’d be adding the shaking my head emoji.

What’s interesting to me about this case, and last week’s decision, is that the parties are coming to the courts wanting to avoid review of documents based on cost. That cost can be considerable. There’s no question about that. But if you want to do that, you have to draft language in a protocol that allows it, but that also meets your obligations under Rule 26. Another solution is not to draft a protocol where you agree to any strategy for complying with Rule 26 at all.

There’s no requirement as to what has to be included in an ESI protocol. There’s simply the obligation under Rule 26 that you provide all relevant information in response to a request for production. You got to look at Rule 26, you got to look at Rule 34, and you’ve got to meet those obligations. You do not have to put down in a protocol how you’re going to meet those obligations.

By putting down language in a protocol, you are committing yourself to a course of action that you’re going to have to engage in. We’ve seen time and time again here on decisions that we’ve discussed on Case of the Week, and that you can search for in eDiscovery Assistant using the ESI protocol tag, that courts are going to enforce the language that the parties agreed to in their ESI protocols.

Go back to the Raymond James decision that I looked at during episode 78 of the Case of the Week. In that situation, the defendant did not conduct a manual review and instead dumped all of the documents that were responsive to search terms on Raymond James. Raymond James then spent the equivalent of the review costs, I think it was around a quarter of a million dollars, to weed out the non-responsive information and then went back to the court on a motion for sanctions to get it back. They were very successful. They got all of it back.

If you’re going to be involved in litigation, you will have to bear the costs of review. We use the American Rule in litigation, and as much as it may not be fair in certain litigation situations, it’s what we have. There are ways to lessen the burden through savvy search term analysis, proportionality discussions, using technology before you draft an ESI protocol, but you have to get into the data.

There’s a discussion on LinkedIn based on an article that Doug Austin from eDiscovery Today drafted about considerations or things that you need to think about in terms of your ESI protocol. Take a look at that situation — that comment, that thread on LinkedIn — and weigh in there because there’s some very constructive feedback on what needs to happen.

You can also take a look at our ESI protocol series that’s included on our blog because this is a very, very critical issue. If you’re going to draft an ESI protocol, you need to know what you’re getting yourself into before you do it. To do that, you need to get in the data. You need to understand what the issues are before you draft that protocol.

If review costs are going to be a concern, you need to figure out a way to deal with that strategically by either reducing the number of documents that you have, making sure that the request for production are proportional to the needs of the case, and narrowly tailored, or engaging with technology. There are a lot of solutions to that problem, but there’s still a cost that’s going to be at issue. Motion practice usually just adds to that cost.

Conclusion

Okay, that’s our Case of the Week for this week. Thanks so much for joining me. I am off to the Kentucky Derby this week, so if you’re planning to be there, please hit me up and maybe we can connect. It’d be fun to be back in my home state for such a great event.

We’ll be back again next week with another decision from our eDiscovery Assistant database. If you’re interested in doing a free trial of our case law and resource database, sign up to get started. Thanks so much. Have a great week.