6 Practical Search Lessons Learned from I-Med Pharma, Inc.
TASA ID: 4285
Within the last year, there have been many decisions from the bench providing guidance on the handling of electronically stored information (ESI). One particular case that stands out is I-Med Pharma, Inc. v. Biomatrix, Inc., No. 03-3677 (DRD), 2011 WL 6140658 (D.N.J. Dec. 9, 2011) which will be referred to here as simply the I-Med Pharma case.
In this specific case, the United States District Court for the District of New Jersey reached a decision affirming a prior judgment by Magistrate Judge Michael Shipp. The defendant appealed the original decision by Judge Shipp not to force the plaintiff to search and review an estimated 95 million pages of records found in the unallocated space on its computer systems using keyword search terms previously agreed upon by both parties.
The reasoning used in the original decision and affirmed here was that the original search terms that were agreed upon were too broad, making the costs associated with performing a full privilege review burdensome. The likelihood of any of this data being responsive was especially low because it included searches on the unallocated space of the plaintiff’s computer system further causing the projected costs to be disproportionate to the expected gain of this search and review.
Search mistakes like those identified by the court decision here still happen frequently. As the volume of ESI increases in all types of cases, these types of issues will continue to be a burden for the foreseeable future.
However, this doesn’t have to happen in your next case. Here are three lessons that can be learned by the parties in the I-Med Pharma case to help you avoid making these same mistakes as you negotiate the pitfalls of e-Discovery in your practice.
1) Know what you don’t know – The first sentence of this opinion says it all: “This case highlights the dangers of carelessness and inattention in e-Discovery.” Some basic mistakes were made in this case. First, the lack of search filtering based on custodian and date, and second, the plaintiff’s agreement to the search criteria without having an idea as to what the estimated result would be.
The lesson for those still getting their feet wet in cases containing e-Discovery is to know your limits. Electronic data can be full of surprises for even the simplest data set, so if you don’t know, ask. Or better yet, find an expert that can help guide you along the way. The e-Discovery field is full of knowledgeable professionals that can act in an advisory role to make sure your client’s interests are met.
Also, spending a little time and money in the beginning can save a lot by keeping you out of situations like the plaintiff in this case got themselves into.
2) Test your terms – Keyword searching has been a part of e-Discovery for quite a while. Unfortunately, people still struggle with the selection and implementation of sound keywords in discovery. If you include yourself in this group, don’t be discouraged. You are not alone. There is hope for those who don’t know how to navigate the minefield of selecting keywords. Even the experts were once in your shoes. They have benefitted from the experience of trial and error that you are now experiencing.
When you get your data, make sure you get it into a searchable form and then test your assumed keywords to see what they return. Based on what you find, tweak your terms and rerun your searches and investigate the results again. If you find brand new keywords along the way, go ahead and analyze those documents separately to see where they lead. You may have to run through many iterations of this in order to really understand your data, but it will be to your client’s benefit. After all, no one gets married on the first date. Spending the time to learn about an individual’s intricacies can save heartache in the dating world and learning about the data for your case can save time, money and possibly even sanctions in the world of e-Discovery.
3) Too much may really be too much – Selecting keywords is as much art as it is science. With almost any term you are bound to get false hits. The key is trying to find a reasonable balance ensuring you find the privilege or responsive documents you are looking for while still limiting the number of extraneous files. Learn about and use the different keyword search options available to you. By using search tools to perform what is called Boolean Searching with syntax operators and options such as OR, AND, NOT, wildcard, proximity, fuzzy, or stemming can help you to limit your results while still ensuring you find all of the relevant data. Many of these tools should also have guides to help in your search efforts as well.
4) Filters are your friend – Narrowing your scope through the use of filtering, also known as culling, can be an effective way to separate out the chaff during searching. Some commonly used criteria for filtering is by limiting your results to specific date ranges, custodians or file types. During the 26(f) meeting filtering criteria can be discussed to make sure all parties are in agreement as to what is appropriate to ensure irrelevant data is excluded, so both sides can focus on the most relevant data and cut costs.
5) Sampling isn’t just for science – Statistical sampling may sound complicated, but it doesn’t have to be. If you return a larger than expected result set from your keywords, creating a sample set from the results to evaluate can ensure you aren’t getting an inordinate number of false hits.
If you do, examine the difference between the accurate hits and the false hits to determine how search options like those mentioned above can help to limit the false hits going forward. This may seem time consuming, but considering the cost of reviewing those extra documents can potentially save a significant amount of money.
6) Unallocated space is not for everybody – In many, even most, cases searching the unallocated space for data is rarely worth the difficulty and cost.
Unallocated space is the space on a drive that is not used for storing active files, but may contain pieces of files that are active or have been previously deleted from the drive. There is an instinct by many that because there may be pieces of deleted file, that it is automatically assumed those pieces would be useful since the custodian could have deleted important data. Although this is, of course, a possibility, the likelihood of finding something useful is still slim.
One aspect that people fail to consider is that most of these files really are just bits and pieces of the full file that was deleted. This means that all you may have to review is merely a segment of the original file which reduces the likelihood of finding relevant data.
There are many examples of parties insisting on reviewing the data in the unallocated space of a custodian’s drive even though malicious or even accidental deletion was not an issue in the case. This data poses issues for review due to the majority of it consisting of complete file information. In other words, just because it is there, doesn’t mean it will be beneficial.
This is not to say that the information contained in the unallocated space of an individual’s computer drive won’t ever be useful. There are cases where these bits of information can work strongly for your client and are necessary to your case. Just be prepared that the cost associated with the forensic efforts and the review of this data may not be insignificant. Unless the malicious or accidental deletion of relevant files is at issue, it may not be worth the effort and expense common in these instances.
This article discusses issues of general interest and does not give any specific legal or business advice pertaining to any specific circumstances. Before acting upon any of its information, you should obtain appropriate advice from a lawyer or other qualified professional.
This article may not be duplicated, altered, distributed, saved, incorporated into another document or website, or otherwise modified without the permission of TASA.