I have received a lot of questions recently regarding whether it is defensible for a corporation to allow its employees to self collect relevant documents for litigation. Self collection refers to a collection process where in-house or outside counsel articulates to each employee (“custodian” of records) the scope of relevance and that custodian takes on the task of identifying any documents they may have that are potentially responsive and isolates them for use in the litigation.
Though self collection continues to be popular as a collection methodology, some argue it is not defensible and that collection should instead be completed using technology solutions that search data at the custodian, department or enterprise level to identify relevant documents.
My take on this is that there is a time for each methodology and often a case may call for a hybrid approach. Self collection is defensible in the right case if planned and executed correctly. There are some types of cases where self collection does not make sense. For example, if your company is sued because a custodian was sending harassing emails, it may be “grossly negligent” to depend on the employee who allegedly sent the harassing emails to locate, preserve, collect and turn those emails over for review and production. You may recognize this fact pattern from Jones v. Bremen High School District, 228, 2010 WL 2106640 (N.D. Ill. May 25, 2010).
However, for most commercial litigation, the primary hurdle is developing a defensible plan and executing that plan properly. This is especially true when a large organization has frequent litigation involving so-called serial custodians.
I will note that I prefer the term “managed self collection” to “self collection” because the term self collection brings to mind the types of scenarios that have given rise to some of the case law in this area: custodians instructed to preserve and collect relevant documents with little guidance on scope or relevance, inadequate training on the technology and unclear instruction on who to contact if they have questions. It should not be surprising that under these circumstances custodians may perform a less than reasonable collection.
The structure of a defensible managed self collection protocol will vary with the complexity of the matter. However, as you develop your plan, consider including the following to bolster defensibility:
- draft a document outlining the scope of relevance with specific examples tailored to each department so custodians know which documents they should include/exclude from collection;
- develop a checklist outlining the steps (IT process) the custodian should take to identify and collect the relevant documents;
- provide technical training for custodians so they are able to complete the collection steps correctly and without destroying metadata or inadvertently failing to collect documents they intend to collect;
- draft an interview form to gather information from each custodian about their involvement in the matter and the locations of documents in their possession;
- set up a conversation between key custodians and a legal representative to talk through the document outlining scope of relevance, the process of collection and to complete the interview form;
- establish points of contact in IT and legal in the event a custodian has questions about process or substance;
- verify custodians are following the process by sampling results and confirming that if a custodian noted during the interview that they had documents in multiple locations that the custodian actually provided documents from each of these locations.





Thursday, May 24, 2012
Daryl –
I agree that self-selected collection isn’t dead. In fact, for smaller organizations it may be the only possible affordable alternative. One case I really like, and just about the only published opinion I know of approving self-selected collection, is Calixto v. WABO, 2009 U.S. Dist. LEXIS 111659 (S.D.Fla. Nov. 16, 2009). The opinion supports your “managed self-collection” analysis while providing some much needed practicality in the process. I would caution, however, that as the size of the organization and number of custodians (and potential witnesses) grows, the ability to sustain the burdens of defending self-selected collection grow to where a more automated solution likely becomes necessary. In Calixto, there were only seven employees who knew anything about the underlying facts or opposing parties. The risk of sloppy, stupid or more sanctionable conduct would likely increase were the number to go to 77 or 777 custodians – and the cost of managing that many self-selecting custodians could get ridiculous without the use of automated search and retrieval technology. There still is no substitute for picking the best tools for the job.