Maybe you went into the law because you aren’t much of a tech person. You’re more of a big-picture thinker, hired as in-house counsel for your legal acumen, ability to strategize, and broad-based experience. You figure you don’t need to worry about the ins and outs of the eDiscovery process. You can delegate eDiscovery to outside counsel and third-party vendors, trust them to exercise good judgment, then lean on them to slash the bill. That’s how it goes, right?
Today’s corporate counsel must have an in-depth understanding of eDiscovery approaches and technologies, including technology-assisted review (TAR). First, there’s your ethical duty to “stay abreast of changes in the law and its practice, including the benefits and risks associated with relevant technology.” Beyond that, though, technological know-how enables in-house counsel to deliver legal advice with greater value for the business. If you have an active knowledge of eDiscovery technologies such as TAR, you’ll be able to competently and confidently wield control over the company’s data and its budget.
Here’s what you need to know about TAR, one of the most critical eDiscovery technologies.
What is Technology-Assisted Review (TAR)?
How does TAR work?
TAR 1.0: Predictive coding
TAR 2.0: Continuous active learning
What are the benefits of in-house TAR?
How in-house counsel can implement TAR
Ensuring transparency and confidentiality when using TAR
In-house counsel’s understanding of TAR drives value for the business
Before we begin, be sure to grab your free eDiscovery resource below:
Download the whitepaper to learn how eDiscovery helps you easily sift through huge volumes of data and find all relevant documents for a case in a fast and cost-effective manner.
Technology-assisted review or TAR—also known as computer-assisted review or CAR—is a way of handling the review phase of eDiscovery by deploying algorithms that can identify potentially interesting documents. TAR is dramatically faster than human-only review and can be every bit as thorough and accurate.
Although the specific approaches to TAR vary, the general idea is that the technology “watches” as a human reviewer codes documents. It then uses natural language processing, a form of artificial intelligence, to discern which words and phrases are associated with tags for relevance, privilege, or other factors. As the technology learns, it continually evaluates the corpus of remaining documents and floats those that it expects to be important to the top of the pile for human review.
Lawyers routinely rely on TAR to classify large volumes of documents. It empowers legal teams to make decisions rapidly by prioritizing the most critical documents. TAR has been accepted by the U.S. courts since the seminal 2012 decision in Da Silva Moore v. Publicis Group & MSL Group and is now viewed as black-letter law.
TAR works much like a Spotify playlist that’s guided by the listener’s preferences or a streaming service like Amazon or Netflix that makes recommendations based on previous entertainment choices. In this case, though, TAR follows the judgment of human subject-matter experts to determine document responsiveness.
There are two general variations of TAR: TAR 1.0, also known as “predictive coding,” and TAR 2.0, also known as “continuous active learning.” With both versions, the subject-matter expert trains the algorithm, which then follows a defensible workflow to make relevancy decisions in a consistent, cost-effective manner. The difference lies in how that training begins.
Predictive coding is a method of document review in which a senior lawyer first analyzes a randomly selected “seed set” of documents. The lawyer’s decisions are then fed into the predictive coding software to train the algorithm to identify similar responsive documents.
As lawyers feed the algorithm more documents, the review team tests the system for quality and accuracy. The review team continues to provide more documents to the system to refine it until the algorithm is sufficiently accurate. Once this threshold is met, the algorithm ranks the remaining documents according to how responsive they are. Documents that the algorithm determines are nonresponsive are sampled by the human review team to ensure that they have been correctly classified.
With continuous active learning, there is no initial seed set of documents. Instead, the review team simply begins coding the documents while the computer continuously learns in the background. With each code entered, the algorithm updates its information and sorts the remaining documents according to its current understanding. The algorithm provides the review team with documents it calculates are most likely to be responsive. The review continues, with the computer learning from each human decision, until the system reaches a point of diminishing returns where the remaining documents are not relevant to the review.
So, that’s how TAR works—now let’s turn to why in-house counsel should consider using it.
TAR delivers much-needed eDiscovery cost and efficiency savings. First, TAR is far quicker and less costly than traditional manual review. Instead of paying lawyers who bill by the hour, now the tedious grunt work of document review can be completed in a fraction of the time using machine learning and automation. This allows lawyers to focus their brain power on higher-level, more valuable strategic work.
In addition, the newest generation of legal eDiscovery technology is more affordable than ever, meaning it is no longer solely within the reach of the biggest and most prestigious (and expensive!) law departments and their outside counsel.
In-house counsel can enjoy the benefits that TAR confers, thanks to:
With the reduced costs and accelerated timelines enabled by TAR, companies can make the most favorable litigation decisions more rapidly than ever. This aids in early case assessment and helps companies identify witnesses to depose, decide on defenses to assert, and enter meet-and-confer hearings with data-based insights in hand.
TAR also allows in-house counsel to maintain greater control over the company’s data, without any of the unnecessary complications of outsourcing review to outside counsel or third-party vendors.
Sold on the benefits of TAR but not sure how to start? We’ve got a few suggestions for you.
Whether your organization chooses to manage eDiscovery in-house should be decided following a cost-benefit analysis.
Although TAR presents the potential for significant time and cost savings without sacrificing quality, in-house counsel shouldn’t expect it to be perfect—but then, neither is human review. Fortunately, perfection isn’t the legal standard; reasonableness is. For more guidance on the reasonable and defensible use of TAR, check out the Technology-Assisted Review (TAR) Guidelines issued by the Electronic Discovery Reference Model (EDRM) in 2019.
Bear in mind that TAR can only determine the responsiveness of text-rich documents, such as emails, Word documents, PowerPoint presentations, and the like. TAR cannot evaluate data from spreadsheets, images, blueprints, or videos, meaning that any review of these materials will require additional human intervention. Additionally, you may need to provide more training or hire additional staff for the legal department if you choose to implement TAR in-house.
Finally, you’ll want to ensure that your IT support, executives, and other employees—as well as your internal and external service providers—are aware of and follow your information governance and data retention policies.
Transparency is critical to a defensible TAR process. As the court stated in Da Silva Moore, “Transparency allows the opposing counsel (and the Court) to be more comfortable with computer-assisted review, reducing fears about the so-called ‘black box’ of the technology.” In-house counsel should therefore document all steps in the TAR process to ensure that they maintain transparency and defensibility.
As part of that transparency, parties using the TAR 1.0 approach may be required to produce the “seed set” used to train the TAR algorithm. Because TAR is not immune to errors, attorneys should be aware that confidential or privileged documents could be inadvertently disclosed along with responsive documents. To avoid waiving the attorney-client privilege or work-product protections, it is advisable to enter into protective orders, clawback agreements, and/or Rule 502(d) orders that mandate the return of inadvertently disclosed privileged documents during discovery.
In-house counsel must be well-versed in eDiscovery approaches and technology in order to serve the business well. Using TAR can significantly reduce costs, both those incurred inside the business as well as those currently billed by outside counsel and third-party vendors. As in-house counsel are increasingly asked to stretch their budgets, today’s affordable TAR tools offer an accessible solution.