Our second to last episode of the year is analysis of the decision on Roy v. FedEx Ground Package Sys., Inc., 2022 WL 17343846 (D. Mass. 2022) by United States Magistrate Judge Katherine A. Robertson. This decision is from November 30, 2022 and we discuss whether a request to produce data from a proprietary database for all class members is proportional to the needs of the case.
Keep reading or watch the video to understand the eDiscovery issues.
Good morning and welcome to Episode 91 of our Case of the Week series published in partnership with ACEDS. My name is Kelly Twigger, I am the founder and CEO of eDiscovery Assistant, as well as the principal at ESI Attorneys. Thank you so much for joining me today.
Okay, just one event to make note of. Registration for the University of Florida eDiscovery Conference happening on February 8th and 9th, 2023 is now open.
We have a limited number of seats that are live in Gainesville at the campus of the University of Florida, the law school — there’ll be 150 seats — as well broadcasting the event virtually and virtual registration is free. Sign up! Get a group together at your organization to be able to watch the event. We’re hoping to provide a lot of really great practical content for you.
In the post or comments, you’ll see a link to Doug Austin’s (eDiscovery Today) post on the case that we’re discussing today, as well as a link to our 2021 Annual Case Law Report published in partnership with eDiscovery Today. Check those links out.
Let’s dive into this week’s decision. This week’s decision comes to us from the Roy v. FedEx Ground Package Systems. This is a decision dated November 30th, 2022, and it comes to us from United States Magistrate Judge Katherine Robertson.
This case is a pretty straightforward analysis regarding the production of data from a proprietary database. On this motion to compel, Magistrate Judge Robertson grants the plaintiff’s motion to compel and orders the production of the requested data on hours for all 450 plus named plaintiffs in the class action from a proprietary database that is managed by FedEx.
As always, we include the issues on each of the decisions in our eDiscovery Assistant database, and the issues for this week’s decision include failure to produce and proportionality.
Let’s talk about what are the facts of this case. We’re before the Court, as I mentioned, on a motion to compel data. The data that is at issue comes from scanners, and that data is maintained in a database by FedEx that shows the hours worked for the individual class members. FedEx produced data for 204 of the 450+ opt-in plaintiffs but has refused to produce that same data for the remaining class members.
We are before the Court here on a Fair Labor Standards Act class action in which the plaintiffs, who are all delivery drivers, allege a single claim for unpaid overtime, basically that they worked more than 40 hours a week and were not paid for that additional overtime. The Court has conditionally approved the class and there are 483 opt-in plaintiffs as of the date of this decision. That means for purposes of this discovery that there are 483 separate reports or data exports that need to be done in order to meet this current request from the plaintiffs.
The data that is at issue is captured by the delivery drivers using a scanner that is a handheld scanner that records their daily activities. You’ve likely seen that scanner if you’ve ever had UPS or FedEx deliver you a package and you’re required to sign your name to it, or they’re scanning it as they hand over the package to you.
The scanner data for each date includes the driver’s name and his or her ISP, a destination terminal, vehicle number, and on-duty and off-duty times. The data recorded by the drivers’ scanners is automatically uploaded to FedEx’s servers and is continuously maintained by FedEx. This is not a situation where we’re talking about any data loss or spoliation. The motion here is purely concerned with whether or not FedEx is required to provide the data for each individual plaintiff in this conditionally certified class.
Plaintiff’s first set of requests for productions asked for, “scanner data including time records and route information,” for the plaintiffs and each opt-in from February 19th, 2015 through the present. It’s a considerable period of time but with a fixed number of metadata fields for each individual plaintiff.
FedEx agreed to produce the plaintiff’s scanner data and vehicle weight information, but initially objected to producing scanner data for all of the opt-ins. Then FedEx changed its position and provided information for roughly 204 of the opt-in plaintiffs who were designated by FedEx or plaintiffs for initial discovery.
Really, what we’re talking about is the difference between the 483 opt-in plaintiffs minus the 204 that have already been provided. Plaintiffs have now moved to compel FedEx to produce the scanner data for those remaining opt-in plaintiffs.
FedEx objects on four different grounds:
- That the additional scanner data would not provide relevant information;
- That the request for individualized information for all opt-ins is not appropriate in an FLSA collective action;
- That the burden and cost of producing the scanner data outweighs its benefits and is disproportional to the needs of the case; and
- the GVWR data is unduly burdensome to produce.
The fourth is not really one that we’re considering here on our discovery decision today because the GVWR data is not actually sought by the plaintiffs on their request for production and the court notes as such.
What’s the Court’s analysis on this motion to compel? Well, as always, the Court begins with the requirements of Rule 26 that data sought must be relevant and not duplicative. The Court also notes that the party bringing the motion must demonstrate the relevance to the case. That’s the plaintiffs here and once that showing of relevance is made, the opposing party, so here FedEx, must show why the request is improper.
There’s not really a dispute here that the scanner data shows a driver’s time in and out of the FedEx terminal each day, and the plaintiffs argue that the scanner data can potentially be used to analyze the hours worked by those who have opted into the collective class action using a common methodology. That’s the relevance is of the data from the plaintiff’s perspective. We can take that data and determine how many hours they worked each week and then put that together with their pay stubs to determine whether or not they have, in fact, been paid that overtime.
The Court then looks at FedEx’s arguments in response, having deemed that the information the plaintiff seek is relevant and says that first, the fact that FedEx argues that there are deficiencies in the data that suggests that its relevance may be compromised is not sufficient. As we’ve talked about multiple times on Case of the Week, the credibility of the data does not negate its relevance for production, and that’s what the Court finds here.
The Court then addresses FedEx’s argument that the data requested is duplicative of data that was produced for the other 204 class members, and finds that argument is really premised on the inaccurate contention that plaintiffs seek scanner data to show that FedEx controls the drivers who deliver the FedEx packages instead of how many hours the drivers worked. Because of that mistaken impression, essentially, or inaccurate contention that FedEx is making, the Court finds that the data is not duplicative, but it’s likely to have material and new information for the additional opt-in plaintiffs for whom data has not yet been produced.
Next, the Court considers whether the data is too individualized to be appropriate at this stage of a class action that has been conditionally certified. This particular argument — this is going to be one of our key takeaways for this case. The Court here looks to another similar FLSA matter regarding drivers brought against FedEx and points out that in that case, FedEx was required to produce data for 30,000 opt-in class members, and that the data was much more burdensome to produce than the standard data that was requested here.
In that case, as in this one, FedEx is the single source of this data for all the opt-in plaintiffs and can produce it in a usable, reasonable format. What we have is another decision against FedEx in a different jurisdiction being used against it here on this motion to compel. The Court also then goes on to reject FedEx’s proportionality argument that the value of the evidence is outweighed by the burden and expense to provide it.
This is another key takeaway from this case: instead of demonstrating the burden of providing the information sought in this case — the actual data from the scanner — FedEx’s counsel points to data from a previous case, which is the one we just talked about, in which the court rejected FedEx’s claims that it didn’t have to produce data that provided a greater burden than the data produced here. Instead of providing information about what the burden would be to produce the additional 250 reports for opt-in plaintiffs from this scanner data, they tried to produce information about how burdensome it was to produce information in this other unrelated case.
The Court notes the following:
Rather than attempting to support its claim of undue burden by estimating the time and cost that would be expended to comply with the plaintiff’s request for scanner data for the opt-in plaintiffs in this case, for whom such data has not been produced, FedEx points to the 16 weeks that it took to compile the scanner data for approximately 60 times that number of opt-ins. As plaintiffs had demonstrated, FedEx has not shown that producing scanner data for all opt-in members of the collective for whom it has not produced such data would be unduly burdensome.
As such, the Court rejects FedEx’s arguments on proportionality grounds, and having rejected all four of FedEx’s arguments, grants the plaintiff’s motion to compel the production of the scanner data for all of the remaining opt-in plaintiffs within 30 days.
What are our takeaways? Well, we talked about a couple of them, but I think one of the things that’s important here is that I try to look at these decisions and read between the lines of what are the parties or counsel thinking when they bring these motions. Because this is yet another motion, and we’re seeing a pattern of this on cases that we address each week, where it seems that the cost of the motion is not really justified when you look at what the arguments are to be made on the motion.
Reading between the lines here, it seems that either FedEx wanted to fight this motion to delay having to provide the information, or perhaps the counsel didn’t have a full understanding of the electronically stored data that was at issue, what its relevance might be, or what it would take to provide it. Instead, it seems like FedEx relied on arguments from another case in which it had been ordered to provide a much more onerous set of data, exactly the opposite of what they wanted to have happen here.
So a couple of things. First, the position that you take in other motions will impact you going forward. Here on a previous motion to compel FedEx to produce a different, more burdensome type of data had been granted by another court, and this court looked at that motion as part of the basis for granting this motion to compel.
Essentially, the Court said — look, you had to provide a heck of a lot more evidence in this other case, and it was all relevant, and it’s very similar to what we’re talking about here. And you’ve given me absolutely no basis to understand why what would be provided here would be more onerous to provide than what was in that case.
So you need to think carefully about your strategy and discovery in every litigation because it will impact you in other cases going forward. Decisions that you make to bring motions that result in written decisions are out there for anyone to use and leverage.
Now, it’s not really clear here whether the plaintiffs had the previous FedEx decision and they’re the ones that provided it to the court or whether the court went out and found that information on its own. Either way, it starts to become a pattern of conduct by a particular party if they’re engaging in similar motions in different cases across the country. So be wary of that. Take that away.
Another takeaway from this case. The question really running through my head in reading this case is whether or not counsel ever found out what it actually would take to run a report with the limited data that was requested here for each opt-in plaintiff. There would have to be some detailed discussions with business folks who manage this database as to what data is available and in what format. But generally speaking, a structured database like the one capturing this data would have easy reporting functions. Those are things that need to be explored by counsel in detailed discussions — ones that are hard if you’re not sure you know what you’re talking about, but the only way to learn is to have those conversations. Now, in this type of case, we’re talking about several hundred reports that would need to be run.
Now, many developers will tell you there are easy ways to write scripts to automate a report if you set up the parameters of those reports. And each system that you talk about will have different functionalities. So it really does depend on you going either to the business person who knows and understands the ins and outs of the database and its reporting functions or to IT or business information systems (however it’s defined in the organization) to understand what are the ways for us to be able to get the information we need in the most cost-effective manner.
It doesn’t seem to me that there was ever a good argument for FedEx that this information would not be produced. And so the most cost-effective strategy would have been to figure out how to use the technology that existed with the database to provide that information. Even if someone had to sit and run 483 separate reports, ultimately, that’s what the court ordered them to do anyway.
So if counsel had taken this into their own hands at the outset of litigation and come up with a viable solution for the client that was cost-effective, a lot of this motion practice could have been eliminated. Instead, once again, we have those significant costs on motion practice with the end result being them having to undergo that process and provide it with a 30-day turnaround.
Now, I mentioned this a little bit, but data from databases is structured data. It all sits in individual fields which allows for the export of those fields of information. Think about when you use a review platform to export a privilege log and you decide what are the fields that you want to have exported and it exports it to you in either a CSV format, which is the generally accepted form, and you can open it in, say, a program like Excel, Numbers if you use Mac, even Google Sheets if you’re using Google applications, any spreadsheet program is going to allow you to open a CSV format. And you could see those individual fields. Well, that’s what we refer to as structured data. And structured data is a lot easier to work with than unstructured data like email, documents, etc.
So if you have a situation like this where you have to produce data from a structured database and you don’t know what your options are to understand how it can be provided, go to the business folks, talk to whoever manages the database, go to IT, whoever you need to within the organization, and tell them what it is that you need to be able to produce. Say it simply, “Here’s what we need to provide, and I want to find the most economical way to be able to do that so we don’t disrupt business and we can provide what we need for purposes of the litigation.”
You need to be detailed. You need to be down to the measure for each report that’s exported, for example “we need either the plaintiff name or a number that can be easily correlated to the plaintiff name as the file name so that the information can be provided in a way that’s reasonable and that we can use it.” All of that can be set up to be run automatically. You just need to understand and really define the parameters of what you need on that data.
The only way to learn how to do this is to do it. You’ve got to start having those conversations. Go and learn what you need to know for each individual client, and then be able to provide that information and discovery in a cost-effective manner.
Talking to your clients — whether it’s legal, IT, or business folks — about that data is what helps you learn how to solve these types of problems. This is where we’re seeing counsel needlessly engage in motion practice and spend resources that end up getting them right back to where they started, which is having to produce information anyway. Save the motion expense, learn how to deal with the data, and provide it in discovery.
All right, that’s our Case of the Week for this week. Thank you so much for joining me. We’ll be back again next week with our final episode for 2022 and another decision from our eDiscovery Assistant database. It’s hard to believe we’ve come to the end of another year.
If you’re interested in doing a free trial of our case law and resource database, sign up to get started.
Thanks, have a great rest of the week and Happy Holidays!