1. Baker & Mckenzie Epiq Systems, Inc.
Second Requests
case study
Increasing speed and confidence in the second request response process
Baker and McKenzie utilized innovative processes to defensibly exclude a large number of potentially relevant
documents from time-consuming, expensive individual review, completing a large Second Request on time. This led
to a successful transaction close for their Fortune 500 client.
The Challenge: Three million documents, an eight-week
deadline, one large second request with an acquisition hanging in By the numbers
the balance
Baker & McKenzie was retained to help complete a response to a 3M Collected documents
Second Request from the U.S. Department of Justice. The collected
data set was approximately 650 gigabytes of electronically stored 650 Gigabytes of data
information (ESI), estimated to be more than 3 million documents.
The Solution 8 Weeks until production
Epiq worked with Baker’s legal team to identify and remove “junk”
e-mail messages from domains like ebay.com and espn.com that
could be attacked as a group. Epiq’s consultants found additional
60 Review attorneys
patterns of junk identification as well, and reduced the volume of
documents requiring review by one million (to 2 million). 90 Estimated percentage time savings
Traditional review begins
Baker began document review using traditional methods. A review 95 Estimated percentage cost savings
team comprising 60 lawyers focused first on email messages and
applied keyword searches. Epiq used near-duplicate identification
and e-mail threading solutions to group substantially similar and 1 Acquisition successfully closed
related documents, so that relevance determinations could be
applied to larger chains of documents.
IQ Review™ deployed to speed review
With 750,000 documents remaining and a very tight schedule, Epiq deployed its predictive relevance application, IQ
Review™, to help complete the project within its original timeframe. Since Epiq was already hosting the documents,
IQ Review™ could be quickly applied to all remaining documents in the data set without sacrificing prior work
product. Additionally, the ongoing review could continue, even as batches of documents were assembled through IQ
Review™.
2. Documents prioritized for review
A senior attorney with thorough knowledge of the nuances of the case reviewed small batches of sample documents,
rating them as responsive or not responsive. After each batch, the classification engine within IQ Review™ compared
the expert’s classifications with its own predictions, while constantly tuning its ability to assess document relevance
for the case. When it found that it could learn nothing more from the document collection, the system terminated
the training process, and applied its analysis to the entire document collection. This took about 10 hours of attorney
time, relatively a very small time investment.
Of the remaining 750,000 documents, approximately:
20 percent received a high relevancy score
60 percent received a very low score
20 percent received a score between high and low
The documents in the top-scoring group were found to have near-universal relevance, and every document in that
grouping was included in the production to DOJ. Documents in the median cluster were individually reviewed, and
only a few were ultimately selected for production. For the lowest ranked documents, two separate methods were
used to probe for potentially relevant materials. First, a variety of keyword search terms were used. Second, every
500th document in this subset was individually reviewed. These two processes provided a quality-control verification
of the predictive coding technology’s results. Neither approach identified any responsive documents for production,
and these materials were set aside.
The Results
A large number of potentially relevant documents were defensibly excluded from time-consuming, expensive individual
review, and the review was completed on time with substantially less effort. Using IQ Review™, review of the final
35% of the document collection took only about 10% percent of the total project time. The final 35% of the project
incurred approximately 5% of the total project costs. The teams estimated that the review could have been completed
in less than half the time required by standard document review had predictive coding been applied from the outset
of the project. IQ Review™ also provided noticeably greater consistency in document classification than that found in
documents reviewed traditionally. The DOJ made no material objections to the ultimate document production, and
the transaction successfully closed as planned.
For more information contact: Myriam Schmell
212 225 9251 or mschmell@epiqsystems.com