The collaboration app Slack sounds, from its full name, like it should be designed for discovery in litigation—“Slack” is actually an acronym for “Searchable Log of All Conversation and Knowledge.” Isn’t that what we’re aiming for in ediscovery?
What can we say: sometimes names can be deceiving.
While Slack is searchable, and it’s often positively packed with discoverable information, it’s not like the structured email data that ediscovery practitioners have grown accustomed to working with. While your existing ediscovery playbooks give you a written, repeatable, and defensible plan for identifying, preserving, collecting, processing, culling, reviewing, and ultimately producing discoverable data, those approaches don’t work well with Slack data.
Here are the top five reasons why you need a dedicated playbook to help you make sense of your Slack data and successfully shepherd it through every stage of ediscovery.
1) SLACK—LIKE OTHER INNOVATIVE TECHNOLOGIES—IS HERE TO STAY
via GIPHY
You may be thinking you’ll just sit this one out. After all, if Slack data is so difficult to manage, surely you could simply prohibit its use and find or wait for a different collaboration tool, right? And Slack has been posting multimillion dollar losses for the last year or so; maybe it’s not worth figuring out how to manage it in ediscovery. You don’t need a playbook for Slack if you don’t use Slack!
But here’s the thing: Slack is the direction that technology will be evolving in, whether your company evolves along with it or stagnates with familiar older technologies. Because the upcoming generation of workers places a high value on communication and collaboration, there will only be more, not less, of these instant message board apps in the future.
What’s more, companies that seek to attract new talent also make themselves more attractive by demonstrating their ability to embrace innovation, which includes adopting new technologies. While you could choose to ignore innovative technologies like Slack, you’re likely to find that that’s a future-limiting move as technology in the workplace—and the generation manning that workplace—continues to change at a breakneck pace.
Practice tip: Don’t try to “policy your way out” of using Slack. Instead, create workable policies that ensure you know what apps your employees are using to communicate and what sorts of communications happen where.
2) SLACK DATA IS POTENTIAL EVIDENCE, WHICH MEANS IT’S DISCOVERABLE
via GIPHY
Under Federal Rule of Civil Procedure 26(b)(1), the scope of discovery includes “any nonprivileged matter that is relevant to any party’s claim or defense and proportional to the needs of the case.” Information does not need to be admissible as evidence to be deemed discoverable.
In businesses everywhere—including 65 companies on the Fortune 100—much of the internal business communication that used to happen on email has now moved to Slack. But, as with other types of evidence, if any parts of those communications are relevant to a litigation matter, they’re still discoverable. That means you need to know about the data you have within Slack and be prepared to identify, preserve, review, and ultimately produce that data in ediscovery. Not to mention that, aside from discovery you provide to an opponent, you’re going to want to be able to analyze Slack data for your own use in litigation.
Practice tip: Incorporate Slack—and other communication and collaboration tools—into your standard custodian questionnaires and legal hold notifications so that custodians are prompted to disclose their potentially relevant Slack communications.
3) SLACK DATA IS SUBJECT TO SPOLIATION—AND ATTENDANT SANCTIONS
via GIPHY
This is the flip side of the discoverability coin: because Slack data is potentially discoverable, you may be obligated to preserve it from the moment that litigation becomes reasonably likely. Fail in that duty, and you could find yourself subject to sanctions for spoliation, up to and including dismissal or default judgment of a case or claim.
Several aspects of Slack raise specific spoliation concerns. First, if you’re using the free version, you can only access the most recent 10,000 messages (and you’d be surprised how quickly you’ll hit that limit). If you have 9,950 messages in Slack when you learn that messages 15 through 30 are subject to a legal hold, you’ll need to act quickly to either stop the addition of messages or to gain unlimited message access.
Slack also has a setting that allows users to edit or delete their individual messages, opening up another avenue for potential spoliation through later manipulation of conversations. And Slack’s internal ediscovery tools are blunt instruments at best, as they don’t allow users to implement targeted legal holds on specific custodians. (We’ll get into the trouble with custodians more in a minute.)
Practice tip: If you allow for edits and deletes, make sure that your Slack administrator has set Slack to preserve edits and deletes.
4) SLACK DATA ISN’T ALWAYS EASY TO ACCESS
via GIPHY
How do you get data out of Slack, anyway? While ediscovery professionals have learned how to deal with email and other established data types, it can be surprisingly difficult to get Slack data out of the app and into an ediscovery review platform.
As we mentioned above, this is especially true in the free, limited-function version of Slack. But even within paid versions, Slack’s native export function uses JSON files that are hard to navigate—each day in each channel is saved in its own file—and consequently slow down review.
Practice tip: If you’re involved in more than one litigation matter at a time, you need to upgrade to Slack Enterprise Grid so you have access to Slack’s discovery application programming interface (API), which gives you more options for exporting data.
5) SLACK DATA IS UNSTRUCTURED—WHICH MEANS YOU NEED A SEPARATE PROCESS FOR MANAGING IT
via GIPHY
Most data in ediscovery is structured—organized into “bins” by custodian. Slack data is different. It doesn’t sort data into mailboxes or files by custodian. Instead, its messages primarily operate as a virtual bulletin board, posting information chronologically for anyone in a channel to see. Messages are marked read as soon as a user scrolls past them, but Slack doesn’t know who’s actually read what. This message organization dramatically redefines “custodians” for purposes of ediscovery; everyone who has access to a channel is a custodian for all of the messages in that channel.
The unstructured nature of Slack data makes it entirely unlike the ediscovery data you’ve grown accustomed to working with. Not only do you need to rethink who is a custodian, but you also have to create new rules for scoping discovery due to the chronological, rather than conversational, presentation of messages. The context for a Slack message may take several screens—interrupted and punctuated by irrelevant remarks and separate conversations—to be fully revealed. This, more than anything, illustrates why you can’t just fold Slack in to your existing ediscovery playbooks. You need a whole new approach that accounts for the unique features of Slack data.
Practice tip: Plan on using an external preservation method to implement legal holds on Slack data. Why? First, because Slack data is not organized in a way that allows for targeted legal holds on individual custodians, so you have to preserve entire channels. Second, because Slack’s internal information governance and ediscovery tools are binary rather than granular—they either preserve all messages in a channel forever, or they delete all messages after a set period of time. In combination, this means it’s easiest to preserve messages for a legal hold outside of Slack and let Slack itself delete messages in accordance with your records retention policy.
