At the simplest level, you can reduce e-discovery technical challenges to scalability, performance, and data security. This article also covers other considerations of return on investment (ROI) and minimizing risk.

Review platform challenges

Scalability

The first challenge associated with litigation review platforms is knowing what scale you need. Like many applications, litigation review platforms suffer from large spikes in demand. However, unlike many applications where the spikes are fairly predictable (end of month, 9:00, or 5:00, etc.), litigation is almost impossible to predict. Litigants file lawsuits without warning, large-scale discovery may or may not occur, and cases can settle or be dismissed at any time.

The size of the spike a single large case can generate has increased significantly in recent years. A large case can now mean sorting through hundreds of gigabytes or even terabytes of data. Getting through this data quickly, and with no downtime, is paramount as many lawyers make up review teams, each billing hundreds of dollars per hour.

Historically, review platforms were built to run on a single server, or even a desktop machine, and law firms and corporate legal departments ran them in house. This worked fine when discovery involved a few boxes worth of documents, and presented manageable challenges to IT departments in terms of space, power, administration, cost, etc. It does not work so well in the current environment.

The specific characteristics of legal review have spawned the creation of grid-based platforms that can be accessed on a software as a service (SaaS) model.

(Software as a Service is a new delivery model where companies don't own the software itself, but pay for using it.)

Tens of networked servers to ingest data, index it, and serve it up to multiple users can comprise SaaS systems. They reside in glass house fortified data centers and are backed by 24x7 support. This model can provide virtually unlimited scalability, high levels of performance that don't deteriorate even with massive volumes of documents or multiple concurrent users, and mission-critical levels of availability.