What Is a Custodian in eDiscovery?

A Custodian doesn't have to be at the center of the dispute. It could be the employee everyone assumes is relevant, their manager, or someone three steps removed who happened to be copied on the right email chain and never thought twice about it.

The term shows up in a second, unrelated context too. In IT and data governance, a data custodian is the role responsible for the servers, backups, and access controls a company's data actually sits on, distinct from a data owner who sets policy or a data steward who manages quality. That's a real title in a lot of organizations. It just isn't this one.

Custodian identification sits at the front of the EDRM, in Identification and Collection, well before processing or review ever begin. But it isn't a task that finishes once and stays finished. The list a case opens with is rarely the list it closes with.

The Types of Custodians

Not every custodian carries the same weight, and treating them all the same wastes time and money. Most matters sort custodians into a few groups:

• Key Custodians: People with direct, central involvement in the events at issue. They're often likely witnesses, and they should receive litigation hold notices first.
• Departmental Custodians: People with access to relevant material because of their role or department, even if they weren't directly involved. Their data still needs preservation and review for relevance.
• Representative Custodians: In a large department where many people hold similar, largely duplicate data, a sample of custodians can stand in for the rest, keeping cost proportional without ignoring the group entirely.
• Non-Custodial Data Sources: Not everything relevant sits with a person. Shared drives, databases, and systems can hold data nobody individually owns, and they get identified and collected separately from any one custodian's set.

Identifying and Interviewing Custodians

Custodian interviews happen early, ideally before the Rule 26(f) conference, so counsel actually knows what data exists before negotiating scope with the other side. A good interview covers the custodian's role, their data sources, the devices and apps they use for work, and whether anything auto-deletes.

The single most useful question in any custodian interview is the last one, who else might have relevant information? Custodian lists rarely come from an org chart alone. They grow because one interview leads to the next name, and the next, until the picture is complete.

Document every interview. If the custodian list is ever challenged, that record is what shows a court a good-faith, reasonable effort was made, not a guess.

Where the Traditional Custodian Model Breaks Down

The traditional model assumes each custodian's data lives in a clean, attributable place, their inbox, their laptop, their home drive. That assumption is cracking. In Slack, Teams, and SharePoint, a single message or document can be co-authored, shared, and hyperlinked by a dozen people who never show up on an org chart as "involved."

Miss a custodian and the risk isn't an inconvenience, it's a spoliation problem, since data that was never identified was never preserved either. Add too many custodians and review costs climb for data that was never going to matter. Both mistakes trace back to the same cause, treating the custodian list as a one-time task instead of something to revisit as a case develops.

Why Getting the Custodian List Right Matters

A missed custodian means missed evidence, and to a court, that looks the same whether it was deliberate or not. Sanctions, adverse inference instructions, and lost motions all trace back to the same failure, nobody stayed on top of the list as the case moved. Custodian identification isn't a task to hand off and forget, it has to stay part of a defensible legal hold process from the first interview through the final release.

One Custodian List. Nothing Left to Chase.

Venio Legal Hold builds custodian tracking, interview documentation, and hold notices into one workflow, so the list doesn't live in a spreadsheet or a project manager's memory. See how it fits into your legal hold process, or book a demo to see it manage your own custodian list.

‍

Frequently Asked Questions

What is a custodian in eDiscovery?

A custodian is a person, or sometimes a system, with possession, custody, or control of electronically stored information relevant to a legal matter. Once litigation is reasonably anticipated, a custodian's data has to be preserved and, eventually, collected and reviewed.

What's the difference between a custodian and a data custodian?

In eDiscovery, a custodian is a person whose data is relevant to a case. In IT and data governance, "data custodian" means something different: the role responsible for managing the technical systems that store and secure an organization's data. Both terms are real, they just come from different fields.

What is a key custodian?

A key custodian is someone with direct, central involvement in the facts of a case, often a likely witness. Key custodians typically receive litigation hold notices first and get the closest attention during collection and review.

How do you identify all the custodians in a case?

Start with the people obviously connected to the dispute, then interview each one and ask who else might have relevant information. Custodian lists are built through this chain of interviews, not from an org chart alone, and they should be revisited as the case develops.

‍