Professional growth          Court news           Productivity           Technology          Wellness          Just for fun

Predictive coding 101: a guide for legal professionals

Predictive coding is one of the best things to happen to the legal industry. New to the idea? No problem. We’ll get you up to speed.

If any of our readers are – how shall I put this delicately – more than a couple decades into their legal career, they’ll remember when document review meant sniffing out the smoking gun in a room stacked floor to ceiling with bankers boxes.

It also meant 17-hour workdays, gallons of coffee, and an unstoppable will to win.

As with most industries, the legal industry continues to evolve and embrace technology. One of the most significant advances in recent years is the use of predictive coding, a technology that is undeniably revolutionizing legal document review.

In fact, over a relatively short period of time, it has become a game-changer for associates, paralegals, and anyone else involved in intensive document review projects.

In this blog post, we’ll explore what predictive coding is, the benefits it offers for document reviewers, and why it’s so important for the future of the legal industry.

What is Predictive Coding?

Predictive coding, also known as technology-assisted review, is a type of artificial intelligence.

In the initial phases of a predictive coding process, a human reviewer teaches the technology about what is (and is not) relevant to a particular case. Then, sophisticated algorithms begin to categorize and prioritize documents.

In the context of legal discovery, predictive coding allows computers to sort through large volumes of documents and to quickly and reliably identify which ones are most relevant to the case at hand.

It does this with statistical analyses and natural language processing to identify patterns in the text, then to assign a relevance score to each document.

You still look at the documents to make the final call, but the computer system helps surface items that are most likely to be relevant so that you don’t waste hours poring over stacks of useless data.

Benefits of predictive coding to document review teams

The benefits of predictive coding are almost immeasurable.

For one thing, it alleviates the need for the aforementioned 17-hour days and gallons of coffee. That alone is something to celebrate.

Here are a few of the other significant benefits, too:

#1: Reduced time and cost of document review

Predictive coding significantly reduces the time and cost of document review by automating much of the process.

Once you eliminate the need for manual review of every single document, predictive coding can save legal professionals countless hours that used to be spent either sifting through boxes or scrolling through seemingly unending pages on a screen.

That time can be used for other, more productive (and profitable) tasks. You can take on more clients, spend more time on marketing, or just work fewer late nights.

Of course, when document review time is significantly decreased, the cost to clients also goes down dramatically. It’s a win for everyone.

#2: Increased accuracy and consistency

Predictive coding can also improve the accuracy and consistency of document review.

Frankly, once you take out the human element in any kind of tedious task, you dramatically reduce errors.

That’s not a knock on legal professionals. I remember how hard it was to pay close attention to every line of tens of thousands of pages of documents day after day after day. It’s a wonder meaningful evidence was ever extracted from document review before now.

Putting human error and fatigue to the side for one moment, predictive coding also ensures that all documents are reviewed using the same search criteria.

This is quite a change from the good old days when you couldn’t help but have certain issues “top of mind” while reviewing documents. That was all well and good until you let significant evidence on another issue escape your attention.

The consistency of predictive coding helps to eliminate these fluctuating biases and oversights.

#3: Better management of large data sets

As humans generate more and more data with every waking moment, the amount of data involved in legal cases continues to grow.

Consider, for example, how much more data is now relevant based solely on things like a litigant’s social media activity.

Predictive coding helps document reviewers manage these large data sets more efficiently and effectively. By automatically categorizing and prioritizing documents, predictive coding makes it easier for document reviewers to get to the heart of discovery without too much pain.

Predictive coding pioneers

Predictive coding really began to catch on after the judge in Da Silva Moore v. Publicis Groupe issued a ruling approving the use of this technology in discovery. It is considered a landmark case mainly because it was the first federal case to explicitly approve the process.

In Da Silva Moore, predictive coding was used to review over three million emails.

The software was able to identify the most relevant documents with an accuracy rate of 86%, which significantly reduced the time and cost of document review.

And, believe it or not, that case took place over a decade ago. The technology has only gotten more precise and more efficient in the interim.

The future of predictive coding in legal industry

Without a doubt, predictive coding is here to stay.

As artificial intelligence continues to advance, predictive coding will become even more sophisticated and accurate. This will enable legal professionals to continually improve on efficiencies and client outcomes.

Predictive coding has the potential for even more important changes to the legal system, however.

In the past, one of the ways wealthy litigants “beat up on” less affluent litigants was to bury them in document productions. Predictive coding changes all that by making large document productions cheaper to handle.

For example, let’s say a receptionist for an international product manufacturer sued her employer for racial harassment. During discovery, the receptionist’s lawyer would inevitably ask for all documents pertaining to complaints made by other employees concerning race.

Rather than producing strictly relevant documents, wealthy corporate litigants would often produce hundreds of thousands of documents knowing the plaintiff could never afford to pay her lawyer to go through them all.

With predictive coding, however, lawyers for less affluent litigants can move just as quickly as everyone else.

This is a massive game-changer when it comes to justice in America.

If you take away the advantages of the wealthy in litigation, it’s quite possible that everyone can have a meaningful day in court.

If you’ve not yet used predictive coding during the discovery process, it may be time to try an eDiscovery solution that incorporates this technology.

If you do, just make sure you find something else to do with the gallons of coffee you’re no longer drinking.

Our recommendations

Follow InfoTrack