Man vs. Machine?

The Round Table w/ Avalon

October 2012 – By JP Midgley, President

While attending the International Legal Technology Association (ILTA) 2012 conference in August, I sat in on a panel discussion debating the hot topics of eDiscovery today. As many anticipated, predictive coding was brought up early in the discussion. If I took only one thought away from the conference it was from a comment made by George Socha (one of the founders of the EDRM model). I don’t remember the exact comment, but the takeaway was evident. Predictive coding isn’t meant to be a battle of attorney vs. machine.  Predictive coding is meant to be machine helping the attorneys.

Many of the comments we have heard early on in predictive coding’s life revolved around the fear that this new technology was going to replace the billable hours accumulated during document review. The attorneys were (to some extent) worried that this could put them out of business. This technology has now been around for a year or two and it is evident that it’s not going to replace the reviewers. This technology is meant to help the reviewers and save the client money. The explosion of data and the massive increase in document review projects has provided job security for many years to come.

While it may have originally been marketed as a way to avoid reviewing all documents before production, we can now see through practical applications that it is used more as a prioritization tool. Let’s start with a hypothetical document set of 1,000,000 documents to be reviewed. After a senior attorney reviews 1000 to 2000 documents with the aid of predictive coding technology, the software will reach a certain level of confidence in its ability to determine a responsive document vs. a non-responsive document.  Every data set is different, but let’s assume that it has identified 100,000 documents as being very likely to be responsive and 900,000 as highly unlikely to be responsive. In most cases, we would not simply produce those 100,000 and ignore the rest. We would still want to know what is being produced and review the 900,000.  The technology gives us is the ability to prioritize the review by allowing us to focus on the 100,000.  We could review the 100,000 with in-house reviewers that bill at a higher rate. This gives them the ability to immediately start working on documents that are likely to be relevant to the matter. The 900,000 documents can then either by passed to contract reviewers that bill at a much lower rate, or if agreeable by the opposition, ignored for the time being.

Using this technology could potentially save the client millions of dollars by putting the relevant documents in the hands of the attorneys who are directly working on the matter. When this technology first came out, even I was skeptical of its use.  Today, the cost benefit analysis of using it as a way to prioritize a review makes it hard to ignore as a possibility on every matter.

Additional Round Table Articles by J.P. Midgley

October 2012 | July 2012 | April 2012 | January 2012 |

October 2011 | July 2011 | April 2011

Comments/Questions for DIGITS LLC?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s