Ride the Lightning

Craig Ball Has Drunk the Predictive Coding Kool-Aid

June 19, 2013

And so, in large part, have I.

An excellent post by Craig parallels my own thinking about predictive coding and how that thinking has evolved over time.

I won't ruin the fun of reading a vintage pithy Craig Ball piece, but the essence is that predictive coding takes far less time and money than full-blown human review. Humans screw up relentlessly – machines do not. It is easy for both humans and machines to distinguish the clearly relevant from the clearly irrelevant but the machines are so much faster – and they can ask for guidance on questionable documents and become smarter still.

This was my favorite paragraph:

So why isn’t everyone using predictive coding?  One reason is that the pricing has not yet shifted from exploitive to rational. It shouldn’t cost substantially more to expose a collection to a predictive coding tool than to expose it to a keyword search tool; yet, it does. That will change and the artificial economic barriers to realizing the benefits of predictive coding will soon play only a minor role in the decision to use the technology.

Craig says it so well that I am left with only "Amen, brother." For smaller cases, the exploitive pricing of predictive coding continues to make it out of reach. One day, smart vendors will change that,  but keyword search tools will remain in widespread use until then.