The Challenge
In the case of Kim v. Cushman & Wakefield U.S., Inc., 2026 WL 1353455 (C.D. Cal. Apr. 24, 2026), defense counsel faced a discovery crisis that's becoming increasingly common: they designed a reasonable email search strategy, ran it against their email repositories, and certified their production as complete. The problem? They hadn't searched Microsoft Teams at all.
The plaintiff, Ms. Kim, alleged pregnancy discrimination after termination following maternity leave. During testimony, it emerged that Teams was "one of the primary communication methods used at the company." Defense counsel then ran their existing email search terms against the Teams repository, producing 47 pages of messages. Only two were relevant to the core pretext analysis.
This should have closed the gap. Instead, it raised a broader question: if the same search terms that worked for email only captured two relevant messages in Teams, what else did they miss?
The Environment and Constraints
The defendant's search terms all required "Connie Kim" as an anchor, such as "Connie Kim" NEAR "terminat!". This reflects standard email search design, where formal communications use full names and explicit references.
Teams data doesn't work that way. Managers might reference "the Smartsheet" or "Brooke's workload" without mentioning the plaintiff by name. The court recognized this structural difference: "Keyword searches alone, without more advanced and thoughtful search techniques, will be inadequate for Teams data, a medium where conversations are shorter, more informal, and less likely to include full names than email."
The court noted that the root cause traced back to a pro forma Rule 26(f) conference. Had the parties conducted a substantive ESI discussion identifying repositories, custodians, and communication platforms at the outset, the Teams gap would've been caught months earlier.
The Approach the Court Required
The court rejected both the plaintiff's demand for "all reasonably likely repositories" without limitation and the defendant's position that running email terms against Teams satisfied its obligations. Instead, the court held that the defendant had undertaken a Rule 26(g)(1)(B) certification obligation the moment it elected to search rather than move for a protective order.
The court declined to dictate specific methodology, stating that "whether through custodian-based collection, refined keyword queries, or technology-assisted review" remained the defendant's choice, as long as the search was reasonable and complete. But it was clear that simply rerunning email-centric terms didn't meet that standard.
This ruling matters because it shifts the burden. You can't just check the Teams box by running your email Boolean strings against chat data and calling it done. The court expects you to design search techniques appropriate to the repository's communication style.
Results and What They Reveal
The court ordered supplemental production. While specific metrics from that supplemental search aren't detailed in the opinion, the two relevant messages already found in the initial 47-page production showed that responsive communications existed in Teams, and the Rule 26(b)(1) proportionality analysis weighed in the plaintiff's favor as a result.
eDiscovery expert Tom O'Connor of the Gulf Coast Legal Technology Center identified a critical technical limitation: Microsoft's native Teams search index "prioritizes exact word matches," "does not index message metadata as richly as Outlook," "often misses partial-word matches," and "returns fewer results when the term is too specific."
In other words, even well-crafted Boolean terms underperform when run against Microsoft's native Teams index. You're not just dealing with informal language, you're dealing with an index that wasn't built for legal search precision.
What They Would Do Differently
The obvious answer: identify Teams as a repository during the Rule 26(f) conference. But that's only the starting point.
You'd also need to test your search terms against Teams data early, before certification. Run a sample. See what comes back. If your email terms anchor on full names and formal phrasing, design parallel terms that capture how people actually talk in chat: project codenames, first names only, pronouns without antecedents, emoji reactions that constitute the entire response.
You'd also need tools that don't rely on Microsoft's native Teams index. The court referenced "more advanced and thoughtful search techniques" and explicitly mentioned technology-assisted review as an option. That suggests iterative search design: test terms, review samples, refine, repeat, all before you commit to a production set.
Finally, you'd document your methodology. The court gave the defendant flexibility to choose their approach, but that flexibility comes with an obligation to show your work. If you're going to argue that your Teams search was reasonable, you need to demonstrate that you understood the platform's communication patterns and designed your search accordingly.
Takeaways for Your Team
Identify collaboration platforms at Rule 26(f). Don't treat this as a checkbox exercise. If your organization uses Teams, Slack, or similar tools for substantive business communication, those repositories belong in your ESI discussion from day one.
Don't assume email search terms translate to chat. The court held that rerunning email-oriented search terms against Teams data is "structurally flawed." You need platform-specific search strategies that account for informal language, partial references, and conversational context.
Test before you certify. Run sample searches against Teams data early in the discovery process. Measure what you're capturing. If your terms return almost nothing, that's not proof there's nothing there, it's proof your terms don't match how people communicate in that medium.
Consider tools built for the job. Microsoft's native Teams search index wasn't designed for legal discovery. Solutions that build their own indexes of Teams data, enabling Boolean operators, proximity searches, and consistent execution across both email and chat, give you the precision the court expects.
Document your search methodology. The court gave the defendant discretion to choose their approach but made clear that choice comes with a Rule 26(g) certification obligation. If you're going to argue your search was reasonable, you need contemporaneous documentation showing you understood the repository's characteristics and designed your search accordingly.
The Kim decision doesn't create new law, it applies existing proportionality and reasonableness standards to a repository type that's now ubiquitous. But it makes explicit what many legal teams have been learning the hard way: your email playbook doesn't work for Teams. You need search strategies designed for how people actually communicate in collaboration platforms, and you need to identify those platforms before discovery gaps become court-ordered supplemental productions.
