#CaseoftheWeekCase LawVideos

#CaseoftheWeek Episode 11: Pitfalls of Not Understanding What’s in Your Data

On this episode of our eDiscovery case law livestream, ACEDS and eDiscovery Assistant chose as the #CaseoftheWeek Barrow v. Dayton, 2021 WL 222930 (Ohio Ct. App. 2021), January 22, 2021. We focus on pitfalls of not understanding what’s in your data and whether sanctions are appropriate following failure to obey a discovery order when the parties don’t agree on search terms, and the ability to recover costs on a sanctions motion. 


Good morning and welcome to our #CaseoftheWeek for February 2nd, 2021. I am Kelly Twigger, CEO and founder of eDiscovery Assistant and the principal at ESI Attorneys. As you know, if you’ve watched our broadcast, we have teamed up with ACEDS to bring you a weekly video series on eDiscovery case law with the goal to keeping you up to date on how courts are treating ESI and discuss the practical implications of decisions, and what you and your clients want to be thinking about.

Each week we select a case to discuss and highlight key issues in eDiscovery, how courts are assessing parties’ obligations, what those decisions mean for you and your clients. These decisions are drawn directly from our curated database of eDiscovery decisions in eDiscovery Assistant. Each of the more than 14,800 cases are manually tagged by our content team with our list of eDiscovery issues for searching purposes. This week’s case is one that includes multiple issues from our database, including sanctions, failure to produce costs and fees, native format, privilege log, attorney client privilege and protective order. Also includes search terms, you can use any of those tags to be able to sort by cases within eDiscovery Assistant.

You’ll see the link to our case, which is titled Barrow v. Dayton in the events page on LinkedIn. If you’re watching this broadcast on another platform, on YouTube or Twitter, you’ll want to connect over to the LinkedIn page to grab that link. That is a public link for you to be able to view the case on eDiscovery Assistant. Also, in the events page on LinkedIn, you’ll see the link to the University of Florida eDiscovery Conference. That conference is one that I’m proud to be a part of the planning committee for, and we’re working hard to bring you a fantastic practical event on March 18th, 2021. Coming up in another month and a half, the event is free. It will be all day. I’ll be participating in two of the sessions and we’re just going to have some fantastic content. We’ll talk about that a little bit more. The agenda will be going live next week. You click on the link that’s in the events page, you’ll be able to sign up to get information as soon as that registration is live. With that information, let’s get into our case for this week.

A lot of times much of our case law in eDiscovery Assistant comes from the federal courts versus state courts. I’ve had a lot of discussions with many of you about why that is. Generally, if you think about it, it’s because the discovery decisions happen at the trial court level, and unless and until those decisions go up on appeal, they’re not published at the state court level. They’re not accessible to most any research platform, eDiscovery Assistant included.

Oftentimes we will get trial court decisions from our users. If you are working on a case and you get a trial court decision in state court that otherwise would not be published, we’d love to be able to make that reasoning in that decision available to our subscribers. If you could send it over to us, you can always email us with any information, questions, other ideas that you’d like to see in the platform at support@ediscoveryassistant.com.

Today’s decision is one of those state court cases. It’s a decision from the Ohio Court of Appeals authored by Judge Michael Hall. This is a case where we don’t get a tremendous amount of information about the facts of the case in terms of the underlying claims, but we do get a lot of facts about the discovery dispute and because we’re talking about eDiscovery case law, it’s really helpful for us. This particular opinion is one that I want you to pay attention to and potentially share with fellow colleagues of yours who may not be as adept at eDiscovery or eDiscovery processes. It really shows the pitfalls of not taking advantage of or knowing and understanding what’s in your client’s data. Let’s dive in.

This decision is titled Barrow v. Dayton. Our site is 2021 Westlaw 222 930 and it’s from the Ohio Court of Appeals. The date of the decision is January 22, 2021. Just from a week or two ago and as I mentioned, is authored by Judge Michael Hall.

What are our facts here? This case is an appeal from a trial court order requiring the plaintiff and his attorney to pay fees as a discovery sanction. Trial court ordered the plaintiff and his counsel to pay fees as part of a discovery sanction for failure to obey a discovery order. Following that decision, the plaintiff appealed.

The background facts. The plaintiff’s name was Barrow, was an alleged author and a member of the Living Word Dayton Church. In May of 2017, the plaintiff filed a lawsuit against the Living Word Church for multiple claims, including tortious interference with contract. The court, as I mentioned, specifically states the facts of the case are irrelevant to the issue before it. So, it really focuses just on the facts that pertain to the discovery dispute. Seven months later, in December of 2017, Living Word filed a motion to compel production of the plaintiff’s emails in native format.

In its discovery responses, the plaintiff had identified 68 witnesses that may have information and multiple email accounts that might include responsive information. For reasons that are not articulated in the decision, Living Word proposed that Barrow provide the account and password information to Living Word’s expert to allow them to download all of the emails. There’s no information in the decision on the capabilities of Barrow’s attorney or why he agreed to allow the other side’s expert to access email accounts and download information. There’s lots of alternatives that we could consider. Maybe the lawyer did not know and understand how to engage eDiscovery service providers or attorneys, or maybe his client didn’t have any money to pay for it. We just don’t know.

Under the proposal by the defendant, the emails would then be searched for relevance based on the names of the witnesses that Barrow provided in his discovery responses. Now, if you’re tracking with me here, you know that 68 people, in terms of one individual’s email, even on multiple counts, is, first of all, a lot of witnesses, a lot of people. Secondly, to consider that all of the emails to and from those 68 people in these email accounts would all be relevant to the claims of this case is not generally a position that we would take without looking at data. Likely there needed to be some other limitations placed on that data, but at this point nothing is. That was the proposal on the motion to compel, the court granted the motion to compel and ordered the production of the information using the search terms compiled of those 68 people’s names.

Barrow then appealed in a separate ruling and this court, the appellate court, reversed and remanded that case because there was no process to deal with attorney-client privileged information as part of the production. On remand, following that initial appeal, the parties agreed to a new protective order. This is the discovery order that becomes an issue in this case.

The new protective order provided for a two-part process. Again, both parties agreed to this and the court signed off on it. Step 1 would include that the expert for Living Word would access all of the email accounts, download the information and search all of those emails for the attorney names on a list, and responsive emails would then be given to Barrow’s attorney for him to review and assess privilege. Now, if you’re thinking that’s a pretty big red flag that only attorney names would be used to assess privilege, then you’re on the right track. When you’re creating privilege filter by client and by matter, it isn’t just the attorney names that you need to include. We’ll talk about that a little bit more in our takeaway section.

As far as this process goes, any of the emails deemed privileged would then be put on a privilege log with information that was specified in the discovery order, and any non-privileged emails would be produced. That’s step one, basically reviewing documents that are responsive to attorneys’ names on the search, sending those to counsel, letting them review for privilege, adding them to a privilege log, and then producing any subsequent emails, any non-privileged emails from that set.

Step 2 would be to search all of the remaining emails that did not hit on the privilege attorney names, based on a list of terms provided by Living Word. Living Word, the defendant here, is coming up with a list of search terms to search the plaintiff’s email, which this decision states mostly included the name of the 68 witnesses. At this point, there’s no information that Living Word had access to the information that its expert had collected, but it’s simply coming up with a list of terms based on the names that were provided to it by the plaintiff in its discovery responses.

What were the results of that two-step process? Step 1 produced 3,200 emails. They were reviewed by counsel. 2,700 of them were deemed to be privileged and added to the privilege log. There’s nothing in the decision that says whether or not the remaining 500 emails were produced. Step 2, the search results of the 68 witnesses, produced over 50,000 emails. This is where it starts to get interesting.

After the search results came in, the expert provided those 50,000 emails to the plaintiff’s counsel and plaintiff’s counsel took a day and said, “This is too many, this is overbroad.” They wanted to refine the search. Unfortunately, they’d already agreed to the process and the search terms to be applied in the process in the protective order.

They went to opposing counsel, and opposing counsel said, “OK, if you want to, you can amend your discovery responses to reduce the names, to allow us to refine the search terms that we can then reapply.” Barrow’s counsel spent two weeks not responding and then insisted that they search only for “material information” rather than relevant information, which I think we pretty much all know that that’s not a standard. Relevant is the standard that we’re always looking for discovery on and material is a standard that is applied in certain claims, but it is not applied as it pertains to discovery.

Barrow also refused to agree to review any emails prior to 2013, arguing that they were not relevant. In response, Living Word provided multiple alternatives for the plaintiff to consider to deal with the volume of the 50,000 emails, offering: you can have extensions of time to review them, you can do rolling productions, you can do batches with deadlines for production or you can turn over all of the emails subject to a clawback provision to prevent any inadvertent disclosure of privileged information. All of those were great alternatives to be able to deal with the volume problem, as a result of the overbroad search terms. Barrow’s counsel just didn’t respond at all to any of those alternatives.

All of that happens up to March 29th, 2019, almost two years to the date of the case being filed in May, when Living Word filed a motion for sanctions with multiple bases. Their motion for sanction asked for sanctions for failure to provide a sufficiently detailed privilege log as required by the order. Remember that we said that the protective order on file had very specific requirements as to what went into the privilege log. Generally speaking, Rule 26 also provides what goes into a privilege log.

The second basis on the motion was the refusal to provide emails prior to 2013. They also argued that the sanctions were appropriate based on the refusal to review and produce any emails responsive to the second search and any failure to respond to the alternatives proposed by defendant. Basically, defendants motion was, “Look, we’ve tried everything here. We came up with a process, we engaged them in the process, we followed the process, gave them alternatives to try and deal with the volume associated with the process, and we’ve gotten nothing, no response, no production of information.”

Barrow’s counsel argued on the motion that the search was not a real search and that the search terms were constructed in a way so as to not actually constitute a legitimate basis for search. I’m not really sure where they came up with that argument. With respect to the pre-2013 emails, Barrow’s counsel insisted that he did not know the defendant at all prior to 2013 and that emails from that time are not relevant. On the other hand, Living Word provided two independent bases for why emails prior to 2013 was relevant and the court found that they were.

The trial court granted that motion for sanctions and subsequently ordered counsel and plaintiff to pay just under $12,000, $11,835 in costs and fees. That is the decision from which Barrow appeals. There were three separate hearings in which the court issued sanctions and then asked for information on costs and fees and then issued information on costs and fees.

Now we’re before the appellate court on appeal. On appeal Barrow has asked the court to challenge the finding that he failed to comply with the discovery order, the finding of contempt, and the award of sanctions and attorney’s fees award. He’s essentially appealing all three of those things.

All three of the issues that are before the court on appeal are dealt with in relatively short shrift and I think we’ll see why, but we’ve already talked about the fact that there was a discovery order, and it was very clearly ignored.

In terms of the analysis, first we have to look at the failure to comply with the discovery order. The order that we’re talking about was the protective order that the parties agreed to, and the court signed on remand that laid out the two-step process. The appellate court found that the scope of that order included all emails that contain search terms on the list, and that Barrow had admitted he had refused to provide those emails or review them. Basically, the court said, “You don’t get to carve out anything new because you agreed to the process and the discovery order and you refused to follow it.” The appellate court also noted that the trial court’s order did not give Barrow any discretion as to which emails to review and if he had thought compliance would propose an undue burden or expense that needed to be raised at the trial court level instead of just ignoring the court’s order.

Looking at Ohio Rule of Civil Procedure 37, which is equivalent to FRCP 37, that rule gives the Ohio courts the authority to sanction an attorney for failure to comply with a discovery order and allows the court to order the party, the attorney or both to pay reasonable expenses unless that failure was substantially justified. The court reiterated the analysis that Barrow and his counsel willfully failed to comply with the court’s order and that his failure was, and I’m quoting, “contemptible”, and as such, the sanctions and award were valid. The award of fees was then based on evidence that was presented at the hearing and there was no abuse of discretion. Again, all dealt with relatively short shrift in terms of analysis by the court, but a fairly clear-cut case that there was a willful disregard for the court’s order regarding discovery to be produced, and then a basis under Ohio Civil Rule 37(b)3, I believe, that gave them the authority to provide that sanction and the cost.

Now, one thing to note, and we’ve covered several cost recovery cases already on a #CaseoftheWeek in past weeks, and the costs here are low. $12,000, considering the scope of what was ignored and the scope of putting together a sanctions motion, is pretty low. I think that’s something to take into account here.

It seems to me that there are a number of underlying issues that are probably at stake in this case that never come out with respect to these kinds of decisions. We talk about that the law is 98% other facts and 2% about the law. I think there were probably a lot of 98-percenters here.

So, what are our takeaways from this case? First and foremost, I want to reiterate that all of the problems that Barrow’s counsel identified with the emails could have been identified prior to engaging in process if Barrow’s counsel had been the one to direct the collection, search and review of the emails.

If Barrow’s counsel had even gone on to the individual email accounts with their client, not even collecting them and putting them in a different review platform, and done the analysis of “which of these 68 witnesses really has relevant information”, he could have helped to narrow the search terms by either amending the discovery responses or by explaining that he had reviewed those emails. The first step you should always take is to understand your client’s data before agreeing to any process or search terms.

If you are doing a process or agreeing to search terms without looking at data first, you are literally putting the cart before the horse. That is, of course, with regard to your own client’s data, it’s very difficult when you’re on the other side and trying to come up with search terms and you don’t know what terms will be responsive to the client’s data. A lot of times in those cases we look at doing sampling or using search term hit reports from a platform that will tell you what your search results are and narrowing those things by custodian or having different search terms based on different places in the organization where a custodian might be. Our search terms for I.T. would be different than our search terms of the marketing department, or the H.R. department.

In this particular case, as noted by one of the alternatives that was given from the defendant, the counsel could have narrowed this issue entirely by amending their discovery request. We talked about that. If you could have just reviewed the data, gotten in there, understood with his client what was actually involved, he would have been able to narrow probably the 50,000 emails to a much more easily reviewable set.

We’ve talked in the past what it cost to review information, and if you did a strictly linear review on 50,000 emails using no sort of computer assisted learning or any other technologies, you’d probably pay about a dollar a document probably be at about $50,000. This doesn’t seem to me like a case that would support that kind of expenditure. Generally, attorneys like to have clients, especially individual clients, pay up front for those kinds of costs so they’re not left holding the bag. It may just be that that wasn’t something the plaintiff here could afford. In which case, as an attorney, you do have an obligation to still meet the requirements of that order. Barrow’s counsel here needed to go and figure out how he could deal with the eDiscovery process.

There are a number of the 98% of facts that I mentioned earlier. Although there’s no information in this case, it strikes me that there are a number of options that could have come up, or a number of obstacles that could have come up. Either one, that plaintiff’s counsel was not very familiar with the eDiscovery process at all, that he didn’t have any technology to assist in the review of the emails from the expert, or like I mentioned earlier, that maybe his client wasn’t paying him. Ethically speaking, it doesn’t matter. If you take a case that involves ESI, you have an ethical obligation to know what you were doing, and you will be held accountable by the court.

You have options that were outlined originally by the California Ethics Committee a number of years ago in that you can: refuse to take the case, you can hire somebody to be able to help you with the case or you can get up to speed. Those are really your three options when you have cases that involve ESI. You have to start learning what those obligations are and you need to understand what you need to know for the cases you take on and what resources you need to have available to you. If you need to go out and start talking to software folks or looking at providers to be able to assist with that, then that’s something that you need to do prior to bringing a case into the firm.

In addition to being able to narrow some of these processes, I think that there are things that Barrow’s counsel could have done to better control the process, even if he didn’t have the ability to collect the information. There are very limited set of facts in which I would consider having my client’s data collected by an opposing party’s expert. The cost to collect data these days, even from multiple email accounts, is not so egregious that it would be worth having the other side do it. I also want to be able to handle that data in a way that allows me to search it.

In this particular case, if Barrow’s counsel could have better controlled the process under the order, he could have done a couple of things differently in agreeing to the process under the protective order. He could have provided some additional protections for his client. He could have asked to review the set and determine whether further revisions to the search terms were necessary or that specific terms could be excluded. This is just another level of protection he could have added. He could have also utilized any one of dozens of eDiscovery platforms that he would have been able to get up and running within a couple of hours to physically look at the information and understand how he could narrow search terms. It did seem like defense counsel was very open to being able to get to an appropriate set of information, and they really just wanted the data. Those are ways in which you could have better control of the process.

With regard to the privileged information, the privilege filter that was applied here instead of just including attorney names, should have been broader. It should have included standard terms that are used in privileged emails, attorney-client and privilege. Those are kind of a starter. We also use domain names for law firms, because those will often pull up emails that are not sent by the attorneys themselves but would come from the attorneys’ offices and would still be privileged. You need to make sure that the filter you’re creating for privilege is complete. Think about all of the different ways in which you have email in your inbox that you would consider privileged. How do you account for those in terms of setting up a privilege filter? How do you account for those in terms of identifying email addresses and people and terms that would be used? What are the issues that were discussed with attorneys that you can use as search terms to be able to further identify privileged information? All of that needs to be done very early on in the process of looking at your data. Those are additional steps that counsel here could have taken to have a more inclusive privilege filter.

Second to last takeaway is that you need to raise these issues with the trial court when you look at a situation and realize that you don’t have the resources to do what’s needed to be done to comply with the order. You’ve got to find a process or a resolution between opposing counsel and the court that means that you’re not in violation of a discovery order. Rule 37, whether it’s in Ohio or under the Federal Rules of Civil Procedure, or the many states that mirror the Federal Rules of Civil Procedure, all provide under Rule 37 that failure to comply with the discovery order is a basis for sanctions. The question is, what is the sanction you’re going to get and how much is it going to cost you? It’s really important that you know and understand and raise those issues at the trial court. You cannot just ignore order even if you’re having difficulty complying with it the way that counsel did here.

We talked a little bit about some of the underlying factors here in counsel’s behavior and, was it the cost of review in terms of linear review that was going to be too expensive? We’ve talked in the past about the ability to use technology to identify groups of information that would have significantly reduced the email that even needed to be reviewed here. There’s one technology that would allow you to review the data faster than linearly. There’s also technology that would allow you to identify large groups of information that aren’t relevant or witnesses of that 68, who really don’t have any relevant or responsive information or whose information you could look at in bulk tag as “produce this”. That process you could have done through rolling productions to be able to satisfy counsel.

There’s another potential explanation, which is that upon looking at some of the 50,000 emails, counsel found information that was damning but still subject to the search terms. You can’t hold those back once you’ve agreed to a process identifying search terms and you don’t know what’s responsive to those search terms, you don’t have another option to be able to hold that information back.

Those are our takeaways for this week. That is our #CaseoftheWeek for this week, February 2nd, 2021. Thanks so much for joining us. Please remember to sign up for the registration for the University of Florida eDiscovery Conference. We’ll be back next week with another exciting decision in an eDiscovery.

Again, I’m Kelly Twigger. Have a great week.

Watch Next Episode Live

#CaseoftheWeek streams live on the ACEDS YouTubeFacebookTwitter, 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.


Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from - Youtube
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google
Consent to display content from - Spotify
Sound Cloud
Consent to display content from - Sound