Most large legal departments handle hundreds to thousands (or even tens of thousands) of contracts each year, ranging from non-disclosure agreements and service contracts to employment agreements and leases. Each contract must be reviewed and edited before it can be executed, which is labor-intensive manual work—in other words, expensive and inefficient. Traditionally, lawyers go through the process of carefully reviewing relevant clauses and then passing documents back and forth electronically until both parties agree on the terms.

In addition to this historically inefficient process is the elephant in the room: In many cases, attorneys are essentially doing the same work—over and over. In fact, Global 2000 organizations spend $35 billion annually on high-volume contract review, $26 billion of which is spent on reviewing and marking up contract language that is semantically similar enough that most edits can be automated. Furthermore, an estimated $7 billion is spent on verbatim work on pre-execution contract negotiation—exact work that has already been done on similar contracts. In addition to this massive waste, the work is repetitive and tiring, which means the potential for error or oversight is high. While other phases of the contract lifecycle management (CLM) process have recently become more efficient with the application of advanced technologies, pre-execution contract review has largely been overlooked.

As law departments continue to look for ways to be stronger partners of the business, legal operations leaders are being more strategic. In fact, 72% of legal teams report having developed a roadmap for planning their technology investments over time, based on CLOC's 2019 State of the Industry Survey. The pre-execution stage of CLM is an area ripe for greater efficiency and a place where companies can quickly realize a significant return on investment.

With recent advances in artificial intelligence (AI) technologies like machine learning (ML) and natural language processing (NLP)—and the emergence of technology that has been developed with input from experts with substantial hands-on experience—pre-execution contract review is now an area where law departments can make dramatic efficiency improvements while boosting their bottom lines.

The pairing of AI with pre-execution contract review is a natural fit. Contract review is repetitive, high-volume work requiring high levels of concentration and accuracy. NLP can recognize semantic patterns in written language and suggest edits, not just flag instances where language may not correspond to a company's "playbook". While there are several dedicated CLM systems that take in-house lawyers through many of the necessary steps of the contract negotiation process, AI-assisted contract review takes the pre-execution phase of the contract lifecycle to a whole new level.

Even if you don't have a dedicated CLM system right now, you may be using a variant of it and not realize it. Like other technologies, CLM has a "maturity model" where organizations proceed through a series of stages toward holistic CLM functionality. In the first stage, there are some CLM processes, but their use is ad hoc. In the second stage, there are basic standards, but they aren't mandatory. In the next stage, those standards become mandatory. Eventually, structured CLM processes are integrated with other company functions. In the last stage, performance metrics are introduced, and processes undergo continual improvement over time. With a mature CLM system, law departments can manage contract specifics now and contribute knowledge and methods later to the next contract drafters.

How AI Addresses the Inefficiency Problem

What's really important, and if often overlooked, is what happens during that crucial pre-execution stage, between the drafting of a contract and its signing. This stage can be a black hole of inefficiency and wasted effort. Without a practical strategy to be more efficient, lawyers spend valuable hours emailing contracts back and forth as they tweak terms and language. Version control problems are a major source of inefficiency. Using Microsoft Word's "track changes" function may help, but versions can still get lost or misfiled, and it can be easy for contract reviewers to end up editing the wrong version of a document.

Apart from version control issues, companies dealing with a large volume of contracts face bewildering complexity in developing language for individual contracts. They typically use nearly identical language from a playbook of clauses that can be reused in many different contracts, according to a George Washington University Law School article co-authored by Robert Anderson and Jeffrey Manns, "Engineering Greater Efficiency in Mergers and Acquisitions." This practice is a potential timesaver, but workflows in which lawyers must do this manually—reading through every word of a contract to determine where standard language is called for, finding the right clause(s) and making sure the language is applied correctly in the specific context—is inherently inefficient and error-prone.

By leveraging a customized AI-based pre-execution tool, your lawyers and others involved in the contract creation or revision process can simultaneously review (without risk of version control problems) and make changes to a contract. Approved changes are immediately incorporated into the latest version. The system automatically suggests revisions based on previous edits to similar contracts. If the reviewers encounter contract clauses that don't seem quite right, they can quickly pull up earlier, similar contracts for reference and side-by-side comparison. AI-assisted technologies that leverage NLP and ML dramatically speed up the process and eliminate much of the tedium of this kind of work.

As mentioned above, there are less advanced, off-the-shelf applications based on publicly available contracts that may be able to "flag" some non-standard language, or tag various parts of a contract green (good), yellow (pay attention) and red (stop, don't use this). But these solutions are not client specific and don't "learn" as they are exposed to more contract data and user input. They are much less precise in recognizing different kinds of contract language, and they can't suggest or make edits like a dedicated contract review system with AI can. A business unit doesn't send a contract off to legal with a green light, yellow light or red-light request; they send it to be reviewed and marked up. A lawyer using such software would still have to find or create substitute language for red-flagged language, which doesn't adequately address the inefficiency problem.

Instead of thinking of AI as another application, it's more useful to look at it as a method or process that is embedded in existing workflows and constantly improves them. While AI has the power to speed up the process, it also eliminates much of the risk of error in the pre-execution stage of the CLM.

This all sounds terrific, but what's the payoff? Simple: You'll save in time and effort, and ultimately, the department will do its work more accurately and more efficiently. The new-breed contract review system will do much of the pre-execution work—and do it more quickly than lawyers shuffling contract copies around and manually comparing a redlined contract with a standard template. That means more contracts can be reviewed and executed in a given time period. Lawyers' time can be put to better, more strategic use. By leveraging your past work, an important business process is conducted more efficiently—both for the legal team and the company as a whole, in which you can put your work, to work.

In part two of this article, I will discuss how the deployment process can be improved and the results that can be achieved through AI-assisted contract review.

Dan Broderick is co-founder and CEO of BlackBoiler. As a former associate with Am Law 100 firm Kilpatrick Townsend & Stockton, Dan specialized in negotiation, related disputes, and developing more efficient processes for contract review.