eDiscovery (e-Discovery) is short for "electronic discovery" and refers to the processes of identifying, collecting, processing, analyzing, reviewing, and producing electronically stored information (ESI) in response to an investigation or lawsuit. ESI can be emails, SMS messages, documents, databases, social media, images, audio/video files, financial transactions, etc.
The EDRM (Electronic Discovery Reference Model) is the widely accepted model that conceptually diagrams the eDiscovery process. Their website is a great starting place to learn about eDiscovery: https://www.edrm.net/
The following is a super explanation of how data moves through the eDiscovery process, aka the "eDiscovery lifecycle".
When litigation or an investigation is pending, an organization issues a legal hold to preserve all forms of relevant information and documents. This means that custodians must take the necessary steps to avoid deletion or destruction of the relevant data. Failure to abide by a legal hold, or spoliation of evidence, can impose severe legal consequences.
After relevant data is identified and preserved, the data is forensically collected from the targeted custodians. The collection process can happen onsite or remotely. During collection, the data is extracted from the custodian’s device to an external or network device to be processed.
After the data is collected, it is processed so that it can be imported to a review tool for reviewers to review. Processing involves culling, deNISTING, deduplicating. It is essentially transforming the raw extracted data into a format that is usable by the software review tool, while preserving the metadata. Another goal during processing is to reduce the amount of data promoted to review.
After the data is processed, it is imported to a software product or tool for review. The documents and their related metadata are linked and stored in databases and reviewed by attorneys and supporting staff in a review tool such as Relativity. During this process, workflows are used to ensure documents are reviewed and coded efficiently. "Review" is the biggest part of the eDiscovery process.
Documents that are deemed relevant or responsive to the litigation/investigation are produced to the opposing party. "Producing" means that the relevant documents are exported from the review tool in a format agreed upon by the parties at the onset of the case (during the meet and confer) and are in a format permissible by the court.
Analytics is a process that can be applied throughout the eDiscovery process (collection, processing, review). Analytics is essentially using algorithms and machine learning to analyze the data. While humans are capable of making many of these analyses on their own, computers can do this much, much faster and often more accurately, thus saving time and manpower.