As with any good sci-fi horror movie, the origins of e-discovery seemed innocent enough. Litigators needed a way to handle evidence contained in emails and word processing documents, so an industry of software providers and consultants was born. But like the science experiment that escaped the lab and now threatens all of humanity, e-discovery has become an end in itself, leaving little room for almost anything else (such as the merits).

To be sure, an army of Judges, lawyers, and legal technologists valiantly try to tame the burden of e-discovery; but their efforts only seem to make the monster stronger. Every new technology raises the standards that must be met. Every court opinion is quickly weaponized into the next sanctions motion.

When it comes to e-discovery, it has become increasingly clear that the only winning move is not to play. Courts can do two things to deprive the organism of the energy it needs to keep growing: (1) leverage the existing rules to consider substantive issues iteratively throughout the case, thereby narrowing the scope of discovery; and (2) utilize special masters more directly in data collection to cut through much of the waste generated by an overly adversarial process that is ill-suited for such a technical task.

Early last year, the special master’s opinion in the matter of Deal Genius, LLC v. O2COOL LLC in the Northern District of Illinois caught my attention. One of the parties in that case used Microsoft 365 for its company email, and – not surprisingly – used its built-in tools to review emails for potential relevancy.

The special master found that the use of Microsoft’s tools to identify responsive emails stored in its own platform was insufficient due to limitations within the technology. Instead, that party was required to collect all the emails from five individuals, run search times on those emails, and review the results. Presumably, such a process requires not just the use of expensive specialized e-discovery software to run the search terms, but also typically requires making copies of sensitive data and storing it in a completely new location for review (because cyber criminals don’t have enough targets). The special master proceeded to issue several more orders, addressing such topics as deduplication and elusion testing.

With such resources being dedicated to these technical nuances, one might think that the parties were well-financed, and the burden of e-discovery was justified. But that’s probably not the case, given that the matter was settled shortly after one of the party’s assets were sold at a UCC sale.

To be clear, the special master’s opinions were both legally and factually correct. But that’s exactly the problem. We have come to accept a system where the identification of potentially relevant evidence from five people requires specialized software and a working knowledge of data science, even when the financial burden may not be justified.

Another notable case is DR Distributors LLC v. 21st Century Smoking, Inc., where the court issued a 256-page order sanctioning the defendant/counterclaimants and their attorneys over e-discovery violations. Three years later, as a result of continuing violations, the court entered default against the defendants and dismissed their counterclaim. As the court itself noted in the opinion, the defendants/counterclaimants were judgment proof, with their legal fees being paid by insurance. The case had become almost entirely about, and its outcome determined by, e-discovery issues.

I would like to say that these two cases are just egregious examples, spawned by the unique conduct of the litigants. But as someone who has spent a fair amount of time working with law firms across the U.S. on e-discovery issues, I don’t think they are. An increasing number of cases involve lawyers and their experts battling over highly technical issues. Even the most diligent, well-intentioned lawyers can easily get caught up in it.

The e-discovery community takes a few approaches to try to reduce the burden of ESI: developing better software, promoting proportionality, and encouraging the use of experts. But neither the e-discovery software nor the experts can keep up with the advances being made in the larger world of data and technology.

Listen to this discussion with e-discovery experts who are struggling to figure out how to collect “attachments” from Microsoft. Even the concept of what constitutes a “document” seems to be in flux. One of the experts – who spends time listening to Microsoft developer podcasts – admits to having trouble finding where certain data is located. And this is just a discussion about Microsoft. There is an expanding universe of other data repositories (Google, iMessage, Slack, AutoCAD, etc.) that each have their own unique challenges.

E-discovery software providers keep developing new tools to try to keep up, but they will always lag far behind. The tools we use to communicate, collaborate and store information are evolving too quickly. And there are only so many experts who can devote the necessary time staying on top of all these developments. Even if there were enough of them to go around, how many litigants can realistically afford those experts?

The answer to these technical challenges is often contained in anachronistic concepts such as “proportionality” and “good faith efforts.” But such principles assume that parties actually know where their data is located; which may not be the case when they rely on platforms such as Microsoft and Google. The sheer amount and complexity of even the most basic data systems overwhelm most balancing tests, where parties are asked to “cooperate” in comparing the importance of information with the burden of accessing it. When such cooperation fails, the thought that a judge can read a few briefs and do better than the experts in sifting through it all is simply not grounded in reality.

We need to let the courts focus on the things that courts do best, and bring in the technologists on a more selective basis to handle matters that are not appropriate for the adversarial system.  More concretely, courts can utilize the discretion they have under most rules to consider substantive issues earlier, and decide what discovery is truly necessary. When technical expertise is needed to map data and determine the best way to extract it, the courts can employ special masters to take a hands-on approach in doing so.

Assume a key issue in a case is whether one party fraudulently induced the other into entering into a deal. Under the traditional approach, the first party would identify all individuals with potentially relevant information, collect data from all the sources used by those individuals, and review (either manually or using technology assisted review) that data for production. Such a process often takes months, as well as specialized software for review, and potential experts to assist in collection and review.

An alternative might be for the court to allow limited document discovery from the key decision makers from both sides, followed by short depositions. In my experience, there are many instances where it turns out that the “fraudulent omission” was actually something that was known to both parties. And even if that’s not the case, the court might be able to determine that a much smaller universe of custodians have potentially relevant information than previously thought. Bear in mind that the court’s decisions do not have to be a final adjudication on the merits; but rather could be for the purpose of setting appropriate discovery limits. And of course, to work, this approach needs to be done quickly (we all know that bifurcated proceedings have not exactly been a cure for inefficiency).

The rules in most courts expressly allow for such an approach. Under the Federal rules, for example, courts have broad discretion to determine the timing and sequence of discovery, to separately try individual issues or claims, to adopt special procedures for managing potentially difficult actions, and to prioritize the presentation of evidence at trial.

Keep in mind that this suggestion is not all that earth-shaking. In preliminary injunction proceedings, the parties engage in expedited (and often limited) discovery followed by an evidentiary hearing where the court makes preliminary rulings on the merits. Such preliminary rulings (or their imminence) often end up resolving the entire case. By way of another example, the parties in multi-district litigation utilize bellwether trials to adjudicate some claims early, while other non-bellwether cases have undergone only limited discovery. Some MDL’s are starting to employ a “legal bellwether” approach to address key legal issues before discovery. For well over a decade, federal courts have considered the merits of case when considering class certification, often before merits discovery has been completed.

To be sure, there are going to be many cases where preliminary rulings do not obviate the need to dive into the morass of technical issues described above. This is where strategic use of special masters comes in. In my prior example (the fraudulently induced deal), the negotiations and motion practice over the collection of data traditionally takes place through a game of telephone. The client talks to the expert who talks to the lawyer; who then turns all of the information into an argumentative letter/brief.

It is possible to cut through much of the posturing by empowering a special master to talk directly (as a surrogate of the court) to the individuals who have the most direct knowledge of a party’s data systems. Something as simple as sharing a screen to see how data can be accessed, or the structure of a file system, may allow the special master to cut through briefs and understand the possibilities and problems with collecting and searching data. By way of illustration, compare trying to trouble-shoot a problem with your computer by reading a technical article, versus having someone show you step-by-step what to do.

This approach may sound anathema to defense lawyers and corporations who are highly protective of their data. But I am not suggesting unfettered access to a party’s systems. I am merely suggesting a conversation directly between the individuals who can most likely understand each other.  It would minimize the use of technical issues as a legal bargaining chip.