Common Oversights in Ediscovery

If you're getting married, buying a house, or doing anything you have little or no experience in, then consulting with people who have already travelled that road and have experience to share is a good idea. In a similar way, we turn to experts for legal or business advice. When it comes to ediscovery, there are many benefits in talking to technical experts.

My knowledge of ediscovery was low 3 weeks ago when I joined Kroll Ontrack as a summer intern. On a quest to learn as much as I could, I thought it would be useful to ask the experts around me at Kroll Ontrack for some advice on ediscovery best practices and how to avoid mistakes. So here are their technical tips on how to manage ediscovery projects slickly and efficiently in the future. In this first post I will focus on what not to do. In a follow up post I will offer some additional best practices.

Digging blindly for data

Don’t underestimate the importance of carefully considering which data sources and custodians you need to collect from and process. Ediscovery providers can provide great technology and services to reduce the amount of data that needs to be reviewed (for example, by providing Early Data Assessment services or by using keyword filters). They can also optimise the review process (by offering, for example, predictive coding technologies) BUT it is very likely that the more data you collect, the more documents you will need to review. You therefore need to make careful selections based on what you need to prove the case and meet your disclosure obligations, and always with the case budget in mind.

Cutting costs with data collection

While thinking carefully about where to gather your data from, at the same time you do you need to be as inclusive as possible when collecting data and avoid the temptation to exclude custodians simply to save money. It can be time consuming and expensive to go back and collect data that really is needed later on.

Focusing on reducing the review set at any cost

When preparing to process large volumes of data from multiple custodians, it might seem appealing to choose global deduplication over custodian deduplication simply because this usually results in a smaller document review set.

Choosing global deduplication means that some custodians’ data sets will be incomplete as duplicates from one custodian set are removed because they already appear in the set of another custodian.

It is important that the right approach is carefully considered based on anticipation of the needs of the case and future production requirements, and not simply on a desire to keep the review set to a minimal size.

Relying on Assumptions

Keep things clear! Everyone only knows what they are told so don’t cut corners. Lawyers or companies engaging with an external expert should make sure they provide all the necessary details of the case (case deadlines, procedural milestones, key facts and issues, important document / data types, what is at stake financially or otherwise) as soon as they are known.

Similarly, experts need to be clear about the assumptions used to assess data volumes and the cost of processing it. No-one should under- or overestimate expectations about data volumes and cost. As far as possible work with real metrics taken from the data itself.

Over complicating the review process

When designing your document review process, keep your category tree (the list of topics or issues than can be assigned to a particular document) as simple as possible. This directly affects the success, speed and quality of the review.

Tagging whole families consistently

When categorising families of documents it might seem sensible to categorise all members of the family in the same way – it makes sense that if an email is relevant, all the attachments should be relevant too, right?   Wrong. The chances are that only some members of the family will have relevant content and that may only be relevant to a particular issue. When you tag all members of the same family consistently you lose clarity as to the content of each individual document. It becomes difficult to collate precisely those documents relevant to a specific issue without re-reviewing documents.

I don’t think ediscovery is effortless – but with the right help you will be able to navigate its maze of complexities with much more ease! Like most relationships, the success of an ediscovery project and a case as a whole lies in good communication and trust between all parties involved.