Predictive coding for eDiscovery

gives small firms a big advantage

Small firms need to do everything they can to compete with the sheer human capital available to larger organizations. Technology is one area in which this can be equalized. One technology that is particularly helpful for smaller firms is predictive coding, which comes into play during a technology assisted review. Using predictive coding for eDiscovery provides for a faster and more accurate review process. Here’s how predictive coding helps.

Keep Up With Prohibitively Large eDiscovery Data Sets

Even with the recent 26(f) changes, a single discovery process can still collect millions of documents that are potentially relevant. In the past, humans were required to review every single document. Then came eDiscovery software, which helped reduce the number of files that needed human eyes. Now eDiscovery platforms like Xera utilize predictive coding for eDiscovery, which takes things a step further.

Rather than review every document, humans are able to review a smaller sample set. These sample sets are then used to extrapolate the relevance of the other data. While several sample sets may need to be processed in order to refine the search enough to find all of the relevant documents, it’s a drop in the bucket when compared to reviewing the data set as a whole.

Get More Accurate eDiscovery Results

A manual review will generally result in approximately 60% of the relevant documents being discovered. While these aren’t fantastic numbers, it’s about what can be expected from employees having to sort through potentially millions of documents. Predictive coding can uncover up to 70% of the documents that are needed, while keyword assisted recovery will only uncover 20% to 25% of the documents. Predictive coding for eDiscovery therefore provides small firms with more accurate results.

Reduce the Potential for Human Error

People make mistakes — machines do not. When dealing with large volumes of data, it’s very easy for people to accidentally code documents the wrong way. These errors could potentially never be discovered. Predictive coding for eDiscovery sidesteps this possibility through the use of accuracy metrics. When the algorithm is uncertain of the relevancy of a document, it codes it as such. This means that items that are questionable can be reviewed again by human reviewers. If the overall accuracy of the eDiscovery solution is in question, human reviewers can add additional sample sets until it has improved.

Reduce the Billable Cost for Each Case

People are expensive. In smaller firms, it’s more than possible that everyone — including partners — may become looped into the discovery process. This creates a significant amount of billable hours. Predictive coding for eDiscovery, however, reduces the amount of people needed during the discovery process. Since the software does most of the heavy lifting, a small team can be used for discovery.

predictive coding for eDiscovery requires the upfront cost of the software solution itself. From then on, the eDiscovery platform will reduce billable hours, cutting overhead and making legal services more affordable.

Get Discovery Results Faster

Perhaps the most compelling reason for a firm to invest in predictive eDiscovery is to get the information that they need faster. Human reviewers can take months to pour over documents in a case — and they will take even longer if they are working multiple cases at once. Once the sample sets have been coded, a predictive coding for eDiscovery process is virtually instantaneous. This gives the firm everything that they need to start building their case at once. When small firms are competing directly with larger firms, this head start can be absolutely critical to the outcome of the case.

Predictive coding for eDiscovery is a cutting edge technology based on advanced algorithms, artificial intelligence, and machine learning. Through the use of this new technology, smaller firms will be able to vastly improve upon their services, reduce their costs, and provide superior results. For more information about using predictive coding eDiscovery, download our latest eBook: “Leveling the Playing Field With Predictive Coding.”

Predictive Coding E-Book