This is part two of the continuing series on two-filer document culling. Please read part one first.

Every attorney has a license to cull irrelevant data before beginning expensive linear review. In fact, it is fair to say that they have a duty to do so to protect their clients and country from waste and abuse. I have been working to refine a culling method since 2006. At that time I limited my practice to e-discovery and put aside my commercial and employment litigation practice. Since then I have written over a million words on e-Discovery, including five books, but this article series is the first full description of document culling.

Culling before review is important because it allows attorneys to control the costs of document review in a legally defensible manner. All too often these costs spiral out of control, or the review is done so poorly that key documents are not found. Both scenarios are obviously bad for our system of justice. We need cases to be decided on the merits, on the facts.

That is, in my opinion, how our system of justice is supposed to operate. It is certainly the way our legal system functioned when I learned to practice law and had my first trials back in 1980. In those days we only had a few thousand files to cull through to find the key documents, perhaps tens of thousands in a big case. Now we have hundreds of thousands of documents to cull through, millions in a big case. Still, even though the data volumes are far greater today, with the two-filter method described here, the few key documents needed to decide a case can be found.

In fact, the larger volumes of writings found in today’s information age, the Big Data, presents an opportunity for lawyers. There are electronic writings everywhere and can be hard to destroy. The large amount of ESI floating in cyberspace means that the truth is almost always out there. You just have to find it.

There is so much data that it is much more likely for key documents to exist than ever before. The digital trails that people leave today are much bigger than the paper trails of old.

The fact that more truth is out there than ever before gives tech-savvy lawyers a great advantage. They have a much better chance than lawyers in the past ever did to find the documents needed to keep witnesses honest, or put more politely, to help refresh their memory. The flood of information can in this way improve the quality of justice. It all depends on our ability to find the truth from the massive quantities of irrelevant information available.

The more advanced culling methods described here, primarily the ones in the second filter that use predictive coding – AI-enhanced document ranking methods – are especially effective in culling the chaff from the wheat in big data cases. I expect this kind of predictive analytics software to keep on improving. For that reason I am confident that we will continue to be able to find the core kernels of truth needed to do justice, no matter how much data we generate and save.