This is yet another term we toss around when discussing electronic discovery.
We borrowed the term “cull” from other meanings. If you look up the definition, it will mention (1) reducing the population of a wild animal or (2) selecting from a large quantity; obtain from a variety of sources.
When I teach newbies, I simply define “culling” as creating a subset of documents from a larger dataset.
Typically, the most common usage, within the electronic discovery workflow, for “culling” to take place, is after we have collected a bunch of data and before we begin a document review. The goal is to reduce the volume of documents we need to review for production, aka the document review stage. However, we also perform a “culling” process at any point during the discovery process when we start with a larger set of documents and eventually end up with a smaller subset of documents that are the most helpful.
Many service providers will use a workflow where they will charge us to process the electronic data, but then they will allow us to spend time in their database software, “culling” the data (at no charge), so that we end up with a smaller subset that is transferred to the hosted document review database, which incurs a monthly hosting charge based on the volume of data.
There are different ways to “cull” the data, using keyword searches, or filtering on metadata fields. It is based on user preferences and/or the needs of the matter.
Hopefully, this helps clear up the definition of another buzz word we use in our industry.