There is an ongoing debate in the industry around how to treat document families. But, what is a document family? Well, there are several kinds of families whether this is an e-mail and its attachments or a zip file with the contained content.
But the usual debate is around e-mail families. The e-mail is the parent and each attachment is a child, hence the term family!
Why is there a debate about this? An email is just an email right? Well, anyone trying to control costs and reduce risk within an eDiscovery exercise very much cares about this topic.
Consider an e-mail with 10 attachments. I have some targeted keyword searches and I identify a hit in only one of the attachments to the email. Does my reviewer need to review the email and 10 attachments or just the one attachment that has the hit? The approach taken will have a large impact on the overall review costs and the approach taken will very much depend on the matter.
When I come to produce my data to the opposition, do I need to include the whole family or just the single attachment that was deemed to be relevant to the case? I don’t want to disclose more than I need as this could be detrimental to my case. However, I also don’t want the completeness of my production to be questioned and enter into a lengthy back and forth and incur the cost and time penalties associated with it.
The problem exists as there are no hard and fast rules. During the Abu Dhabi Commercial Bank v. Morgan Stanley & Co, Inc. case, the Special Master needed to investigate the appropriate approach and found that although there are guidelines, there is nothing to dictate the required approach:
The Special Master concluded that the “best practice” was for parties to discuss and settle issues surrounding emails and attachments in advance.
The typical approach with regard to producing data is the data should be produced “as kept in the usual course of business”- complete families. There needs to be a placeholder to represent any data that needs to be withheld due to Privilege or Commercial sensitivity for example. And lets not forget data unavailable due to technical reasons – these items also need to be represented.
The eDiscovery application chosen needs to be flexible to meet your differing case requirements. From a review perspective you need to be able to review at an item level, family level or perhaps a combination of the two.
When you produce data, you need to have the option to include placeholders that clearly denote the reason for their placement whether it is due to privilege or any number of technical reasons.
