X

Thank you for sharing!

Your article was successfully shared with the contacts you provided.
If a technical application existed that coded litigation documents at a fraction of the time and cost that you’re used to dealing with, would you use it? As new technologies emerge, Winston & Strawn, a 150-year-old firm with nearly 900 attorneys and offices worldwide, asks this question time and again. As litigation support project manager at the Washington, D.C., office, my latest conclusion is that autocoding is an important piece of litigation technology and its time has come. AUTOCODING DEFINED Autocoding is the process of using software instead of human coders to create coded fields of data for documents. It operates by analyzing the text files associated with different documents to determine what the “document type” should be. It extracts the various document descriptors and data associated with the document, such as date, author, recipient, subject, and more. The principal values of autocoding are the speed of creation — literally hundreds ofthousands of pages can be processed in a day — and the low cost, made possible by avoidingmanual keying. Autocoding can use text run through an optical character recognition (OCR) system or gleaned from electronic data discovery as source media. The result of autocoding is limited by the quality of the underlying text generated through OCR from the imaged document. In the early 1990s, some vendors offered autocoding services and software, but these systems and applications were not widely used. Perhaps these companies were ahead of their time, or maybe all of the technical components weren’t advanced enough to bring it all together. At that time, using autocoding for a document-intensive case seemed like an illusion. However, autocoding technology has come of age and we are clearly not in the same predicament today. BUDGETS, DEADLINES AND CASE MANAGEMENT When faced with managing a case that involves a good number of documents, including e-mails, you need the right tools and processes in place to help the litigation team. Litigation technology and autocoding are important because they offer reasonable project turnaround time and cost. Sometimes you don’t always plan in your budget for the costs to manually code litigation documents. Yet document coding can easily become the biggest cost of the document management project. Autocoding provides a cheaper alternative to quickly obtain fielded data. It costs a fraction of standard bibliographic coding, and, if you’re dealing with “clean” documents, the fielded data’s accuracy is as good, if not better, than manual coding. A recent case handled by my team involved construction litigation. Construction and project development matters, by their very nature, lead to the creation of complex contracts and are often document intensive. We received approximately 70,000 pages of printed e-mails from the opposing counsel that needed to be indexed for review. Autocoding was the perfect fit to help the litigation team manage and cull down the collection in order to find relevant documents. When we use autocoding vs. manual coding, we can typically review litigation documents in weeks vs. months. A TALE OF TWO SYSTEMS Previously, we tried ALCoder and Planet Data’s Auto Data Extraction service. Based on our experience, the vendor we picked for this project was Planet Data. Since the collection was all printed e-mails, Auto Data Extraction service was able to quickly and accurately extract and field data. When you query a database of thousands of documents to find the smoking gun or other supportive information, life is much easier with data that is close to pristine. Anyone who has worked with litigation databases knows that high accuracy of coded data means an easier job for all involved. Auto Data Extraction allows technology to take over for human processing and applies consistent rules in all situations, whereas the results of manual processing often varies between persons, and can even vary when processed by the same person. Therefore, I have found that bibliographic coding and electronic discovery using autocoding is accurate and consistent. There are a few things to consider when using autocoding. The quality of autocoding is directly relational to the quality of the incoming paper. If the documents are in poor quality, it affects OCR and the autocoded data. One way that Planet Data provides quality assurance to their processing is with “confidence rating,” which is applied to every document. It allows the litigation reviewers at Winston & Strawn to find the documents that have a confidence rating of 45 (out of 100, meaning 100 percent confident) or lower and choose whether to enhance the coding through manual processes. The confidence level report is part of the company’s standard operating procedure. Planet Data also provides manual coding services, so we don’t need to go to another vendor for enhanced coding of lower-rated documents. WHY NOT JUST OCR? Sometimes you can search the OCR’d documents without using the autocoding process, but when you perform searches, you get everything. With autocoding, we can pinpoint key documents faster and better focus our database queries. When deadlines are coming, time is scarce and the last thing we want is to have to sift through vast amounts of documents when autocoding can do the job faster and easier. This is what we found as the real benefit of autocoding vs. manual coding and OCR. Considering our particular construction litigation, the documents were a perfect fit forautocoding because they were printed e-mails. Planet Data gave us a searchable database in just over one week, whereas traditional coding would take a month. This allowed us to review and pinpoint key documents right away. Our clients then reviewed the documents via iCONECT, adding efficiency to the process. We used the autocoding technology to perform focused searches on our data set to create witness files for use during depositions. We may have achieved similar results with pure OCR, but the pre-fielded data let us start with a smaller universe to cull from to get our final exhibit set. This was all done during the discovery phase of the case, and we will use the same technology and process on subsequent document productions. IN CONCLUSION The construction litigation and our use of autocoding technology for it can serve as a guide for future cases at our firm. While I wouldn’t have done anything differently for this case, there are issues with respect to document unitization and extracted information from forwarded e-mails that can affect the outcome of a project. However, a talented project manager at your vendor of choice can help avoid these pitfalls. Law firms can also establish other in-house methods to complement the autocoding process. For example, one workflow we’ve become fond of is to enhance the coding using human beings on documents that do not autocode well. I’ve also thought about using this as a starting point for a regular manual coding project, where the people follow behind the autocoded data and clean up every record. Litigators tend to be an impatient lot, but for understandable reasons. For strategic, budgetary or similar reasons, litigation attorneys often delay the start of many litigation projects, including automated litigation support ones, until the last possible moment. Once that happens, the project — including collection and processing of electronic information or scanning and coding of documents — needs to be moved along at an extremely rapid pace. Fortunately, a company like Planet Data that offers scanning, OCR, autocoding and other artificial intelligence solutions helps make the process happen more quickly. For more information on Planet Data or its Auto Data Extraction service, contact Zoltan Horvath via e-mail or at (914) 333-0670.

This content has been archived. It is available through our partners, LexisNexis® and Bloomberg Law.

To view this content, please continue to their sites.

Not a Lexis Advance® Subscriber?
Subscribe Now

Not a Bloomberg Law Subscriber?
Subscribe Now

Why am I seeing this?

LexisNexis® and Bloomberg Law are third party online distributors of the broad collection of current and archived versions of ALM's legal news publications. LexisNexis® and Bloomberg Law customers are able to access and use ALM's content, including content from the National Law Journal, The American Lawyer, Legaltech News, The New York Law Journal, and Corporate Counsel, as well as other sources of legal information.

For questions call 1-877-256-2472 or contact us at [email protected]

 
 

ALM Legal Publication Newsletters

Sign Up Today and Never Miss Another Story.

As part of your digital membership, you can sign up for an unlimited number of a wide range of complimentary newsletters. Visit your My Account page to make your selections. Get the timely legal news and critical analysis you cannot afford to miss. Tailored just for you. In your inbox. Every day.

Copyright © 2021 ALM Media Properties, LLC. All Rights Reserved.