#CaseoftheWeek Episode 27: Requesting a New Search following Self-Collection

In Episode 27, we have chosen Maker’s Mark Distillery, Inc. v. Spalding Grp., 2021 WL 2018880 (W.D. Ky. 2021), April 20, 2021 as our #CaseoftheWeek. We analyze whether and when a party can request a producing party to do a new search on data collected by employees under the supervision of counsel where searches for email are done using Outlook.

Good morning and welcome to our #CaseoftheWeek for June 8, 2021. My name is Kelly Twigger. I am the CEO and founder of eDiscovery Assistant, as well as the Principal at ESI Attorneys. Thanks so much for joining me this week. As you know, if you tune in with us each week through our partnership with ACEDS, we choose a case from eDiscovery Assistant case law database to talk about the practical implications of what it means for you, your clients and your practice moving forward.

For each of the platforms that we’re live on today, you’ll find links in the various platforms for the event to the public link for the case that we’re going to discuss today, which is the Maker’s Mark Distillery case, as well as links to two other cases that are kind of an issue as a result of this decision. One is the H Guys case and M1 5100. The M1 5100 case is actually a case that we covered on week one of our #CaseoftheWeek back in November 2020, which seems like a really long time ago, but actually wasn’t.

We’ve also just launched a new website at ediscoveryassistant.com, so please hop over and take a look. You’ll find a new events calendar where issues are coming up as we move back to in-person events, as well as some webinars that are happening. You can check that out. You can also sign up for our blog at ediscoveryassistant.com/blog.

All right. Let’s get into our #CaseoftheWeek for this week. Our case is titled Maker’s Mark Distillery vs. Spalding Group. It is a decision from April 20, 2021 from Judge Lanny King, who is the United States Magistrate Judge in the Western District of Kentucky. This case raises some important issues, and I think kind of muddies the waters a little bit on on where we are with regard to self-collection.

There are some key issues in this decision to be discussed. As we always talk about, in eDiscovery Assistant, we tag all of our decisions with issues and the issues for this case include search terms, self collection, failure to produce and metadata, as well as proportionality.

The licensing agreement was renewed a couple of times, and in 2013, Maker’s Mark notified Spalding that it was terminating the license under the terms of the agreement. We’re looking at a time period—from a business perspective for Spalding—from 1996 prior to when it entered into the licensing agreement all the way up until December 31, 2015. That comes into play later when we talk about production of financial information.

Under the termination, Spalding then had until June of 2016 to dispose of its remaining inventory of the licensed cigars. After the termination of the license, Spalding began to sell a different cigar that the complaint alleges was also seasoned with Maker’s Mark Bourbon and the packaging of which Maker’s Mark alleged was intentionally designed to evoke an association by customers between the Bourbon Cigar, which is what the new product was called, and the Maker’s Mark prior licensed cigar. Essentially Maker’s Mark was claiming that the Spalding group advertised its new Bourbon Cigar in a variety of ways to associate it with Maker’s Mark and the previously permitted licensed cigars despite the cease and desist and the termination of the licensing agreement.

The allegations are essentially that Spalding caused actual confusion among retailers as to the relationship between Maker’s Mark and the Bourbon Cigar, the subsequent product produced after the termination titled “The Bourbon Cigar,” which should not have contained Maker’s Mark and was outside the licensing agreement. Maker’s Mark claims that there was actual confusion. That’s really what the issues are before the court.

In the context of this decision, the parties had proceeded with discovery and, in responding to requests for production from Maker’s Mark, Spalding allowed employees to identify documents for collection using search terms that Spalding devised without input from Maker’s Mark, ran those terms across Outlook, which is the email client from Microsoft, and then exported the results for review and production. There are some other facts that we’ll get into in the case. The key here is that that Spalding really permitted employees to do self-collection, and there is no ESI protocol that’s entered on the docket between the parties.

There was also a motion for a protective order which was originally denied by the court and has been renewed on this particular decision, although, it’s not really addressed.

What we have now is a motion to compel brought by Maker’s Mark regarding the production of documents by Spalding, and Maker’s Mark raises a couple of separate issues, one of which has multiple steps to it. The first of the two issues raised is that Spalding’s sought additional 30(b)(6) testimony.

This is not really an issue, except to the extent that when you’re entitled to 30(b)(6) testimony in the scope of the 30(b)(6) testimony that you’re entitled to is important. That was kind of a minimal issue here in this case.

The more important issue from an ESI perspective, and what I want to discuss today, is that Maker’s Mark claimed that Spalding’s search for responsive materials was inadequate. There were a number of reasons that Maker’s Mark gives alleging that search was inadequate. There are separately five different reasons in which they give.

I want to go over those, but more importantly, at the very outset of its analysis, the Court says immediately that Maker’s Mark has waived any objections that it has to Spalding’s responses to the requests for production because they failed to bring them in a timely manner. The Court states that as a “general rule, when a party fails to object timely to interrogatories, production requests or other discovery efforts,” which here must mean responses to those production requests in terms of the documents, “objections there to are waived.” The Court notes that Maker’s Mark waited six months to bring the issue to the defendant’s attention once documents had been produced and nine months to bring this issue to the attention of the Court.

The Court then says that federal courts have found more timely objections constituted waiver, and that Maker’s Mark’s failure to object earlier constituted a waiver. Now, what the Court does not say is what that waiver practically means in the context here. What I think is interesting is as we go through the next elements of this Court’s analysis, that failure to bring the issue to the court’s earlier impacted the court’s perception of Maker’s Mark’s argument.

We’ve talked about this multiple times, but the timing of when you bring your objections to a court is very key. It is really difficult as we engage in litigation to be able to sit down and come to terms with all the arguments that need to be made following a production in a really timely fashion. The reality is that you have to do that. You don’t have that luxury of waiting for several months until you want to get ready for depositions and then discovering what the flaws are in the production that you’ve received. So timeliness.

Back to the adequacy of search. Maker’s Mark argued that the defendants, “chose to employ a flawed search process without making any attempt to meet and confer with Maker’s Mark beforehand, as required by the report of the parties Rule 26 planning meeting and request of the court and ask the court then to grant its motion to compel documents and have Spalding develop a new search protocol in conjunction with Maker’s Mark.” Essentially, Maker’s Mark is saying, “hey, you went ahead and decided what you should search for and you didn’t even talk to us, and our Rule 26 report requires that you talk to us first.” Again, no ESI protocol according to the docket.

Maker’s Mark then provides five categories of what it calls evidence to demonstrate the flawed process from Spalding with regard to its search. They alleged that defendants ignored the email account of at least two custodians, that the key word searches were used were inadequate, that Outlook is an inappropriate program for discovery purposes, that the employee self collected and made relevance determinations as opposed to an attorney doing it, and that the low number of emails produced implied withholding.

To each of those individual points — what did the court say. With regard to the custodians, one of the custodians the court looked at and found that based on his position and information that had been produced, that custodian really had no relevance to the matter. However, Spalding had offered to run the search terms across that custodian’s email, the same search terms that had already run across other custodians’ email, and produced them within three days. Maker’s Mark said, “no, we still think your process is flawed and we don’t want you to do that. Well, what happened is the court said that the process was fine, and because Maker’s Mark had refused the offer from Spalding, it got no new documents. The Court essentially said, “they offered you rejected. As a result, you get nothing else.”

We’re going to talk for a minute in which the court does the analysis of how it found that the key searches were adequate. I do think it’s worth noting here that this opinion kind of jumps around and is all over the place. It’s a little bit hard to dissect, just reading it verbatim. When you’re taking a look at this, you’ve got to really take into account a lot of the facts that are piecemeal distributed throughout the opinion. They’re not clearly stated at the outset. It is really diving into a specific analysis of this decision. As you are going through these ediscovery decisions, when you’re writing your motions, make sure that you are looking at all of the individual facts and making sure that you’re drawing analogies between the cases that you’re trying to cite to support your arguments and the facts as they exist in this case. In some instances here at least, the Court does not address where Maker’s Mark may have made some of those arguments based on previous case law.

The second argument with regard to the flawed process for searching by Spalding was that the keyword searches were inadequate. There’s not a listing within the case of what the keywords were that Spalding used.  What’s really interesting here is that the Court kind of drops the whole discussion of whether or not the search terms were adequate into a footnote. That’s a little bit problematic because whether the search terms are adequate to identify relevant and responsive information is a pretty key point for purposes of discovery.

What Maker’s Mark argued is that Spalding didn’t use any search strings with Boolean terms. They didn’t create any search strings at all, but rather used single words like bourbon. Bourbon is the only term that’s mentioned within the footnote or briefly mentioned within the case. The Judge actually kind of glosses right over the search term discussion, which is, I think, a little bit problematic because it’s it’s key to how you’re going to identify relevant documents, particularly if you’re going to use a search tool as would be built into Outlook. Again, the Judge just kind of glosses right over that, doesn’t really address that the parties should have had any sort of collaboration or cooperation regarding search terms, and he doesn’t address the Rule 26 Report argument that Maker’s Mark raised. Of course, Maker’s Mark is arguing that the lack of search terms would have missed responsive documents and that Spalding’s search terms were unreasonably narrow and unsophisticated. Again, the Court really just glosses right over that and says, “based on the fact that the product was called Bourbon’s cigar, it feels like it would have to encompassed exactly what was needed.”

The next argument that Maker’s Mark raises is that Outlook is an inappropriate program for searching. This, again, is dealt with by the Court in a footnote, which is kind of surprising given the importance from an ediscovery perspective of considering whether the tool that the parties use to search for data was really an appropriate tool. I think without some actual test results showing the difference in search results between Outlook and another type of program—where you would have collected a larger amount of data and input it into a review platform and then compared the search results you would get from Outlook to what you would get from the review platform—it’s going to be hard to be able to demonstrate to this court based on this ruling that Outlook is an insufficient tool. The Court found, because the search is utilized were sufficient, that it really didn’t see any reason why Outlook was not sufficient. Maker’s Mark did make multiple arguments as to why Outlook was not an appropriate search tool. All of them were really dismissed by the Court. Even where Maker’s Mark asserted that that limited search results would appear from Outlook, the Court identified that employees would be able to be aware of additional search results in Outlook.

The argument is not well made or well analyzed by the Court with regard to Outlook. It feels like—and perhaps the briefs on this case would have given us more evidence—but it feels like there could have been more built into the technical arguments of why Outlook is not a great search tool. Again, I really think that you’d have to do a test demonstrating the efficacy of search term results between Outlook and a review platform in order to be able to show some empirical evidence to the Court that Outlook was not an appropriate tool.

The next issue that Maker’s argues, is the failure to supervise. Essentially, it argues that the employees from Spalding were allowed to decide what emails were relevant and to be collected and produced. Maker’s Mark asserts, “that a proper search for discoverable documents requires careful planning, oversight and monitoring by the party’s counsel,” and they argue that, “by having the employees, and not the attorneys, determine which documents were relevant to the plaintiff’s request, that they fail to provide the proper oversight and need to start again with a new search.”

Now, this is where we kind of get a little bit dicey, the court agreed that attorneys and not employees are the ones to make the legal determination of relevancy in discovery. But the court says, “however, while related, collecting documents and making relevancy determinations are distinct actions.” That statement is problematic for me as someone who’s constantly working with clients to make sure that we’re identifying search terms, relevant information, and the scope of documents to be collected, because collecting documents is essentially a physical process of pulling out a copy of information that you have identified is relevant and potentially responsive to a request for production or to a preservation requirement. Deciding what’s relevant comes later. The scope of what is identified for collection should be slightly broader than what is relevant. I think that the Court loses track of that a little bit here, or maybe it’s language that I’m interpreting differently than the Court intended. It feels like the Court is making a big distinction between the identification of information, the collection of it, and then a subsequent relevancy determination. They kind of go together. You have to do some of that relevancy analysis when you’re identifying information before you collect it. That statement from the Court is a bit problematic for me.

Now, Maker’s Mark cites to the case called H Guys, and we’ve provided that public link to you in the comments on each of the platforms that you’re looking at. That’s also a case that’s in eDiscovery Assistant. That’s a case from the Northern District of Illinois in which the Court did order a second review of documents following self collection by the client by saying that there were smoking gun documents in that case.

The Court here distinguishes the H Guys case because it says that in H Guys there was evidence of improper searching that prompted the new search being ordered. The Court also found that here, the attorneys instructed the clients on what was relevant, where he did not find that evidence of that instruction was available in the H Guys case. Essentially, the Court says, look, here Spalding’s counsel instructed the employees on what they should look for and that was sufficient as far as supervision goes. In H Guys, the Court found that although there was empirical evidence of smoking gun documents that were not produced, there also was nothing in the decision that specifically stated that counsel had instructed the employees on what to look for. Those are the two ways in which the Court distinguishes the H Guys case.

I still think that we’re kind of walking a fine line here with regard to self-collection, and we’re getting into what is required by counsel to meet its Rule 26(g) obligations with regard to supervising collection of information.

I want to point you back to that M1 5100 case, which is also included in the links in event. In that case, which again, was our first #CaseoftheWeek. If you’re a user of eDiscovery Assistant, you can go to the M1 5100 case and watch that #CaseoftheWeek within the platform. In that case, Judge Matthewman from Florida went through in great detail what the lawyers obligations are under Rule 26(g)(1), under Rule 26(a)(1) – (a)(3) and that, “Rule 26(g) imposes an affirmative duty to engage in pretrial discovery in a responsible manner that is consistent with the spirit and purposes of Rule 26 through Rule 37.” They also quote the Sedona Principles on best practices.

I think we’ve got a little bit of a difference here in what constitutes self-collection and what the attorney’s responsibilities are with regard to providing oversight to the client. Is it sufficient to simply say here are the things that may be relevant without looking at any of the data and making any decisions about what keywords are relevant? Is it appropriate to allow the client to come up with search terms without looking at the data, which it feels like is what happened here in this Maker’s Mark case?

I think this ruling is a little bit different than the other rulings that we’ve seen on self collection with regard to the M1 5100 case, with regard to the H Guys case. There are also other decisions that are cited by Maker’s Mark in their motion, which the Court distinguishes, and the judge here essentially holds that you have to have some kind of empirical evidence and not just be arguing about process. You have to be able to show there are smoking guns, or that there weren’t very many documents produced. You have to be able to show something that allows you to get a footing with regard to arguing that the process for self-collection was not appropriate.

That seems to be a different bar than what we’ve seen in previous cases. In the M1 5100 case, there was evidence that the responding party had only produced twenty two documents here. Here there’s also evidence put forth by Maker’s Mark that a minimum number of documents were produced. I think it was more than twenty two, but it was still a minimum number of documents. The question becomes, what is the scale that we’re operating on to be able to test process as it relates to self-collection?

There’s not enough facts, I think, in this decision for us to be able to evaluate that effectively. I think we need some more guidance. I think we need to know exactly what it is that a lawyer is responsible for doing.  I think that the M1 5100 case sets that out in a more succinct manner than this particular decision in the Maker’s Mark case.

Final issue in the case was about the proportionality of sales and financial information. Maker’s Mark had asked for additional information stemming back to prior to the licensing agreement, so back to 1996. We talked early on about the time period, a licensing agreement being 1996 to 2015. And Maker’s Mark said we want to see your business from back from the start of the licensing agreement all the way up until after the licensing agreement expired in order to be able to assess damages. The Court agreed that those documents were appropriate and that providing them did not create an undue burden on Spalding to create them.

What are our takeaways from this case? We’ve talked about a couple of them already. One, don’t delay in reviewing your documents and filing your objections. In this particular instance, six months, which doesn’t seem like a very long time in discovery, was long enough for the Court to say that you’ve waived any objections. I think that did have an input into the Court’s analysis as to whether or not a second search should be ordered for Spalding.

Second, we talked a lot about this already. Does the court set a new bar on self-collection? What level of supervision is really permissible by clients? Is simply a conversation where they instruct the clients on what’s relevant going to be sufficient? That doesn’t seem to be sufficient, in my estimation, based on all the work that we’ve done in reviewing data with clients. You’re really shooting fish in a barrel by guessing at what search terms would be without having looked at data. But again, we don’t have a very detailed recitation of what the facts were as to what counsel’s level of supervision was in this case.

I do think that the Court’s decision here that the receiving party has to provide some evidence of misdeeds versus just contesting the process is an issue. I think that the Court really does a de minimus analysis of whether or not there was sufficient supervision. The only note by the Court is that some of the employees testified that counsel gave them instructions on what was relevant.

Is that sufficient? I don’t think so. We talked about the M1 5100 case. I do encourage you to go back and look at that as it pertains to self-collection so that you can understand exactly what your duties are in the jurisdiction where your existing.

I think the next issue from a take away perspective is when you’re challenging the software that has been used to conduct searches or the search terms that have been used, you’ve got to be able to provide empirical evidence for the court as to why those methods are not sufficient. It doesn’t sound like Maker’s Mark was able to do that here.

I would have the same challenges to this searching that Maker’s Mark had with regard to the use of Outlook. There are empirical tests that would have to be done to be able to provide some information to the Court as to why they were not valid. The question that I have in my mind with regard to all these issues is why the parties never engaged in putting together an ESI protocol to set out who the custodians were, what the search terms were.

There’s one important point that I haven’t mentioned yet, and that is that the Court does state two things. It says, one, that the parties have not cooperated at all during the course of discovery. But it also states that Spalding has been very forthcoming about the custodians and the search terms and the methods used to conduct a search during discovery. That’s a key point for the Court here, because it believes that by sharing all of that information, Spalding has put it on the table and if Maker’s Mark needed to object to it, they have to provide some evidence as to why it would be objectionable beyond just saying that the process of self-collection was not sufficient.

That’s our #CaseoftheWeek for this week. Thanks for joining me. I will be back next week with another edition of our #CaseoftheWeek from eDiscovery Assistant. If you’re an ACEDS member and interested in using eDiscovery Assistant, there is a discount available to current ACEDS members and a trial for folks who are preparing to take their ACEDS exam.

If you’re interested in either of those, please reach out to us at ACEDS@ediscoveryassistant.com and one of our team will get back to you. If you’re interested in doing a free trial of our case law resource database, please jump to ediscoveryassistant.com and the upper right hand corner and sign up to get started. Have a great week. Stay safe and healthy and I’ll see you next week. Thank you.

Watch Next Episode Live

#CaseoftheWeek streams live on the ACEDS YouTube, Facebook, Twitter, and LinkedIn company pages. Subscribe to the ACEDS social channels to be notified when we go live.

Catch the Replay

You’ll be able to read episode summaries and watch the recorded show right here on the eDiscovery Assistant blog or access the recorded versions from the ACEDS social media channels.