What Is the EDRM?

Every matter needs a shared map. Without one, legal, IT, and outside vendors end up describing different parts of the same process in different words, and scope conversations turn into translation problems. The EDRM, spelled out, the Electronic Discovery Reference Model, is that map, and the EDRM meaning is easiest to grasp exactly this way. It's the reason "processing" means the same thing to a paralegal in New York and a vendor across the world.

What Is EDRM in eDiscovery?

The EDRM model organizes electronic discovery into a defined set of stages that legal teams can follow across any matter. It's a conceptual model, not a rigid checklist, so it describes what happens to data without locking anyone into one tool or workflow. Practitioners use it to plan discovery and to scope work the same way with opposing counsel, clients, and vendors.

The term carries two meanings worth separating. It refers to the model itself, the diagram of discovery stages, and to EDRM the organization, the professional community that maintains the framework and publishes related standards. This entry is about the model. Teams apply the EDRM framework to plan discovery, and it sits at the center of modern eDiscovery practice, underpinning how organizations build a defensible discovery process.

How the EDRM Model Works

The EDRM diagram is usually drawn flowing left to right, but it isn't a strict sequence. Teams move back and forth between stages as new data surfaces or scope changes, and several stages often run in parallel.

The left side of the model is about process and readiness. Information governance, identification, preservation, and collection, the groundwork that determines what data exists and secures it against loss. The right side is about the data itself as it gets refined toward production: processing, review, analysis, production, and presentation. Volume typically shrinks moving rightward, since each stage filters the dataset closer to what's actually relevant.

The Nine Stages of the EDRM

Each stage, sometimes called a phase, has a distinct job in moving raw data toward defensible evidence:

1. Information Governance: Setting policy for how data is created, stored, and disposed of before any dispute arises, which reduces both risk and cost long before litigation is on the table.
2. Identification: Locating potential sources of electronically stored information, custodians, systems, devices, and cloud accounts, and scoping their breadth and depth before anything gets collected.
3. Preservation: Protecting relevant data from alteration or deletion, usually through a legal hold notice that puts custodians and IT on notice to pause routine deletion.
4. Collection: Gathering preserved data from its sources in a forensically sound manner that keeps metadata and chain of custody intact for later use as evidence.
5. Processing: Reducing and converting collected data into a reviewable format, culling, deduplicating, and extracting metadata and text along the way.
6. Review: Examining documents for relevance, privilege, and confidentiality, typically the most labor-intensive and expensive stage in the entire model.
7. Analysis: Evaluating content for patterns, key facts, and context across the dataset, often surfacing connections a document-by-document review would miss on its own.
8. Production: Delivering relevant, non-privileged documents to opposing parties in an agreed format, native, image, or a near-native hybrid.
9. Presentation: Displaying the resulting evidence for depositions, hearings, or trial in a form a judge, jury, or witness can actually follow.

‍

Why the EDRM Isn't a Straight Line

Treating the EDRM as a one-way checklist is the most common misreading of it. New custodians surface mid-review. A processing job flags something that sends the team back to identification. Analysis turns up a gap that means collection isn't finished after all. None of that means the model failed, it means the model was built for exactly this.

Following it still matters, because a consistent, repeatable process is what makes a production defensible. Courts and counsel recognize the structure, which makes it easier to show that data was preserved, collected, and produced through a sound method, not an improvised one. It also controls cost. Because each stage filters the dataset, disciplined execution keeps expensive reviews focused on a smaller, genuinely relevant population, and a shared vocabulary across legal, technical, and vendor teams prevents gaps and duplicated work.

One Platform Across All Nine Stages

Most teams don't run the EDRM on one system. Legal hold lives in one tool, processing in another, review in a third, and every handoff between them is a place data, or time, can get lost. Venio's unified eDiscovery platform supports teams across the full EDRM lifecycle instead, in a single system with one dataset moving through every stage rather than being re-imported at each one. Book a demo to see how it fits your own workflow. 

‍

Frequently Asked Questions

What does EDRM stand for?

EDRM stands for Electronic Discovery Reference Model, a framework that maps the stages of the electronic discovery process. The same acronym also names the professional community that maintains the model.

Who created the EDRM?

George Socha and Tom Gelbmann developed the EDRM to bring structure to a fragmented eDiscovery industry, starting as an effort to standardize how legal teams describe and execute discovery. The framework is now maintained by the EDRM community and updated with industry input.

Is the EDRM a linear process?

No. Although the diagram flows left to right, teams routinely move between stages and run several in parallel. The model is iterative, reflecting how discovery adapts as new data and issues emerge.

What's the difference between the EDRM and the IGRM?

The EDRM maps the full electronic discovery lifecycle. The IGRM (Information Governance Reference Model) focuses specifically on the information governance that underpins the EDRM's first stage, detailing who owns data and how it should be managed and disposed of. The two models align, with the IGRM setting the foundation the EDRM builds on.

‍