Imagine you’re working on a relatively straightforward litigation matter involving a contract. You’ve agreed that “delivery” is a key search term for discovery. But as you’re reviewing your opponent’s production, you come across a single text message that asks, “Will you accept that delivery?”
And then—nothing. The discovery doesn’t include any messages responding to that question. When you ask your opponent to fill in the blanks, they respond that they’ve already produced all responsive messages because they’ve produced everything that hit on your agreed search terms.
If that doesn’t sound right to you, you’re in good company. That was the issue in Sandoz, Inc. v. United Therapeutics Corp., No. 19-cv-10170 (D.N.J. June 16, 2021), where Special Discovery Master Jose L. Linares ordered a party to “produce additional context-related cell phone text messages.” Here’s what happened.
The Facts: Text Messages and Their Context
Defendant United Therapeutics Corp. (“UTC”) sought an order to compel plaintiff RareGen to produce additional text messages in discovery. UTC argued that RareGen had, as in the example above, only produced those text messages that “hit on specific search terms” without providing the surrounding messages that would give those texts helpful context. UTC pointed out that it and plaintiff Sandoz had already produced contextual text messages and that it was “simply asking that all of the parties be treated in the same manner.”
RareGen, perhaps sensing a losing battle, didn’t attempt to argue that the requested contextual text messages wouldn’t be relevant. Instead, it objected that UTC’s request was untimely, as UTC filed its motion to compel after the close of fact discovery.
The Special Master’s Ruling: What’s Good for the Goose …
The Special Master was unmoved by this assertion. The court noted that it had, in an earlier ruling, ordered UTC to produce context-related text messages. UTC quickly responded to the order the next day by asking RareGen to produce its own context-related text messages. This request preceded the end of fact discovery on April 2, 2021. However, the parties were unable to reach an agreement, leading UTC to raise the issue with the Special Master.
The Special Master concluded that UTC engaged in no “undue delay” but instead “moved promptly to try to resolve the issue.” Therefore, “in the interest of fairness to the parties,” the court extended its prior ruling and ordered plaintiff RareGen to “produce context-related text messages for each of the [identified] messages … including the preceding text message or responding text message, if they exist.”
“in the interest of fairness to the parties,” the court extended its prior ruling and ordered plaintiff RareGen to “produce context-related text messages for each of the [identified] messages … including the preceding text message or responding text message, if they exist.”
The Earlier Ruling: Text Messages Need Context
The Special Master had considered the question of contextual text messages previously. Before the March 29, 2021 order, the only messages that UTC had provided were those that specifically hit on search terms. There, plaintiff Sandoz argued that the surrounding messages were critical to understanding the context of those produced messages. Sandoz, therefore, asked the court to order UTC to produce text messages that, while not directly responsive, provided the necessary context for responsive messages.
The Special Master agreed, noting that there were “gaps in the UTC production” that could be remedied by the production of “surrounding context-related text messages.”
Key Takeaways on the Importance of Context in Ediscovery
Text messages aren’t the only type of data for which context matters.
With the rise in remote work, companies are increasingly turning to collaboration platforms like Slack to keep their teams on the same page. As with text messages, Slack messages tend to be short and incomplete. The only way to understand them is to review them in the context of their surrounding messages.
Additionally, most Slack discovery approaches rely on JSON exports, which are hard to read generally, much less in context.
Check out other case law related to Slack ediscovery.