How Analytics Is Shaping the Current and Future Practice of Law

But the existence of all of that data — or, more specifically, the ability to mine knowledge and insight from vast quantities of relevant data — is precisely what is now giving some lawyers a significant competitive advantage. Aided by solutions leveraging the new analytics technologies, attorneys are driving efficiencies and vastly improving legal and business decision-making. It is not an overstatement to say the age of big data has given rise to a new kind of lawyering in which legal strategy and the business of law are fundamentally data-driven.

Analytics is the engine that powers this kind of data-driven legal practice, which relies on advanced tools that can take massive volumes of legal data, structure it and strip out irrelevant or redundant information, and make it readily searchable and comprehensible to users looking for very specifics kinds of information. These are tasks that would take humans weeks, months or more to effectively complete.

Here we will briefly consider three areas or “maturity levels” of analytics — descriptive, predictive and prescriptive — and look at their role in revolutionizing the practice of law today.

Descriptive Analytics

In legal analytics, the power resides not in the data itself, but the insights derived from it. Descriptive analytics uses advanced technologies such as natural language processing and machine learning to mine large volumes of legal data and turn it into actionable knowledge and insight. Descriptive analytics differs from predictive and prescriptive analytics in its exclusive focus on historical data, which it extracts and organizes to help users identify legal trends over time, analyze behaviors of participants in litigation and highlight other factual information that lawyers can use to better determine the likely outcomes of cases, develop winning legal strategies, estimate the value of a case, forecast litigation costs and timing, and make crucial decisions, including whether to settle or proceed to trial.

Semantic search is a key tool in the analytics arsenal, improving accuracy by determining the searcher's intent and the contextual meanings of search terms to generate the most relevant results in a matter of minutes or seconds. Search filters and recognition algorithms are used to disambiguate legal terms that have different meanings in various jurisdictional or practice settings. Lawyers also play a vital role in descriptive analytics, constantly providing input on context and metadata, and providing quality control to leverage the technology's ability to “learn.”

Advanced data visualization is a vital component in descriptive analytics. Visualization helps users quickly identify patterns, trends, relationships and insights behind the data that would be nearly impossible to aggregate with text-review alone, thereby increasing the accuracy and efficiency with which professionals comprehend relevant data and make key decisions. Studies have concluded that the human brain processes images at a rate 60,000 times faster than it processes text. Visualization presents information in a way that is immediately accessible, intuitive and interactive, turning complex data into easy-to-digest charts and graphs. From search term mapping to presentation of legislative bill progression to review of patent images, data visualization offers legal professionals the opportunity to draw insights more quickly and efficiently.

Predictive Analytics

If the primary focus of descriptive analytics is on what has happened in the past, predictive analytics turns its attention to what will happen in the future. This encompasses a variety of statistical techniques — including things such as predictive modeling, machine learning, regression analytics, time series models and graph analysis mining — to analyze current and historical data and make reasoned predictions about what is likely to happen in the future. Effective predictive analytics require cutting-edge technologies, intelligent algorithms and tremendous amounts of rich data from a variety of sources that can provide the broadest view of an issue or potential decision and the factors that are likely to influence it. The richer the data, the easier it is for machine algorithms to “learn,” which in turn ultimately leads to more accurate predictions.

Predictive analytics is a not a substitute for human judgment or experience. Its real value lies in the ability to determine what is likely to happen based on past behaviors and current data. This functionality may prove especially useful in certain practice areas where cost analysis is crucial. For example, in the area of medical malpractice, analytics can identify cases with similar fact patterns, and can help predict when a case will close and how much will likely be awarded. Based on this information, lawyers can make a data-driven determination on whether and how to move forward with a lawsuit, and they can typically arrive at that determination in a relatively short period of time.

Similarly, if you wanted to assess opposing counsel in litigation, you could view their win/loss ratio, clients, status as plaintiff/defendant/counterclaimant and case count, and then use that information to make fact-based predictions about your prospects in the case. If you were trying to move your case to another venue, you could analyze the judge's track record to determine whether he or she is likely to respond to a motion to transfer, assess the judge's grant/deny rates, and estimate how long a case might take to get to a grant of a permanent injunction, trial or termination.

Using predictive analytics in patent portfolio management, patent portfolio managers can estimate and budget the cost of patent submittal and patent prosecution — and do so with an 80% a level accuracy. The use of predictive analytics in forecasting passage rates for legislation can drive even higher rates of accuracy and allow companies to make decisions about deployment of lobbying resources.

Other industries are already using predictive analytics routinely to forecast a variety of outcomes related to cost, inventory, market variability and more. Use of these solutions in the legal market is starting to accelerate as the benefits of predictive analytics become more apparent and lawyers better understand its value in strategic decision-making.

Prescriptive Analytics

The still-emerging technology of prescriptive analytics extends beyond descriptive and predictive analytics to recommend specific courses of action and to illuminate the likely outcomes of those actions. With prescriptive analytics, legal professionals will be able compare and contrast multiple “futures” based on historical trends and current data.

In the legal context, imagine an application that proactively makes recommendations on what legal strategies to employ based on outcomes of similar cases. It might ask questions of the user if it lacks sufficient information to make a recommendation. Such a tool might also conduct legal research to find relevant case citations or information that can support or refute the user's legal arguments. It might also recommend other search topics or keywords to pursue based on past searches, using techniques similar to those used by Amazon's product recommendation engine. In theory, it could even draft case arguments based on user inputs and/or case concepts, using language taken from the outcomes of similar, successful cases.

While predictive analytics provides insights into the potential futures based on data, prescriptive analytics goes one step further and provides actual advice, and it will continually refine its recommendations by tracking outcomes of actual decisions and incorporating that information in future recommendations. As analytics, machine learning, natural language processing and other deep learning technologies advance, it is only a matter time and acceptance before such technology becomes a true and trusted partner to lawyers and legal teams. Undoubtedly, these advances in technology will raise new questions related to “authorized practice of law” and push the boundaries further regarding human versus machine-driven decision-making.

The Future of the Data-Driven Lawyer

Exploding data volumes demand that legal professionals interact with data in new ways. Analytics is a growing part of the solution to this problem. It enables the transformation of data into knowledge, detecting patterns and connections which no human, no matter how experienced or diligent, can perceive without the assistance of advanced tools.

We are already very close to a future in which attorneys developing a case strategy will routinely apply a combination of descriptive, predictive and prescriptive analytics to a body of rich litigation data to accurately estimate variables like time to trial, the potential value of a case and likely outcomes.

Legal teams will be able to generate a visual snapshot of a judge's past cases and previous rulings, quickly determine which arguments are likely to work in front of that judge in particular types of cases, and decide whether seeking a change of venue is appropriate.

They will run searches to assess the tendencies and success rates of opposing counsel, viewing graphics that map out other lawsuits pending for the plaintiff or defendant and calculating the likelihood of settlement in those lawsuits.

They will use analytics engines to produce a list of recent relevant cases, identify specific arguments that have the potential to support or refute a recommended legal strategy, and assemble the research and arguments required to draft a brief.

They will be able to determine how case outcome and value might change as different variables — such as fact patterns, case concepts and venues — change, or as their own understanding of the case evolves. These examples are just a glimpse of the future workflow scenarios that we see coming in this age of legal AI and analytics.

Whether lawyers are analyzing past information to make searching more intuitive and faster while ensuring search results are accurate and relevant, or detecting patterns of behavior in attorneys, judges and opposing parties to help determine an optimal case strategy, analytical tools have begun to fulfill the vision of data-driven legal practice. Analytics does not — and will not — replace lawyers, but it is already making the practice of law smarter and more efficient.

*****
Jeff Pfeifer is vice president of Product Management for LexisNexis. Over a 28-year career in legal technology, he has worked to introduce a series of cutting-edge solutions for lawyers and other legal professionals. He is responsible for the product development strategy for LexisNexis in North America. Follow him on Twitter @JeffPfeifer.

Anyone following the news headlines of late is aware that artificial intelligence (AI) is being heralded as the technology that will transform industries far and wide — including the legal profession. The potential for AI and other advanced technologies is vast. The evolution of technology in the practice of law today has already led to significant advances in data analytics and data visualization, each of which are having a significant impact on legal work. The nature of legal work today and the need to consume vast amounts of unstructured text make our profession a ripe target for the promise of machine learning and artificial intelligence.

The legal landscape is also evolving rapidly. Tighter budgets and higher client expectations are forcing law firms and in-house legal teams to constantly improve productivity and efficiency to maintain a competitive edge. Corporate clients understand the value of experienced, highly skilled legal minds, but are increasingly reluctant to pay top hourly rates for work routinely performed by junior law firm associates as they assemble and organize evidence, conduct research and begin drafting documents.

At the same time, lawyers and law firms continue to stagger beneath the ever-growing volumes of legal data being generated: More than 14 million legal case decisions, tens of millions of legislative bills and hundreds of millions of regulations recorded in the U.S. alone. This is in addition to the extensive volume of internal data accumulated by law firms and corporations. The amount of research and analysis that lawyers must perform in litigation has become much more time-consuming; for lawyers in data-intensive practice areas, standard review of documents is becoming unmanageable. As the amount of electronic data increases, legal research, analysis and discovery become increasingly challenging and time-consuming.

But the existence of all of that data — or, more specifically, the ability to mine knowledge and insight from vast quantities of relevant data — is precisely what is now giving some lawyers a significant competitive advantage. Aided by solutions leveraging the new analytics technologies, attorneys are driving efficiencies and vastly improving legal and business decision-making. It is not an overstatement to say the age of big data has given rise to a new kind of lawyering in which legal strategy and the business of law are fundamentally data-driven.

Analytics is the engine that powers this kind of data-driven legal practice, which relies on advanced tools that can take massive volumes of legal data, structure it and strip out irrelevant or redundant information, and make it readily searchable and comprehensible to users looking for very specifics kinds of information. These are tasks that would take humans weeks, months or more to effectively complete.

Here we will briefly consider three areas or “maturity levels” of analytics — descriptive, predictive and prescriptive — and look at their role in revolutionizing the practice of law today.

Descriptive Analytics

In legal analytics, the power resides not in the data itself, but the insights derived from it. Descriptive analytics uses advanced technologies such as natural language processing and machine learning to mine large volumes of legal data and turn it into actionable knowledge and insight. Descriptive analytics differs from predictive and prescriptive analytics in its exclusive focus on historical data, which it extracts and organizes to help users identify legal trends over time, analyze behaviors of participants in litigation and highlight other factual information that lawyers can use to better determine the likely outcomes of cases, develop winning legal strategies, estimate the value of a case, forecast litigation costs and timing, and make crucial decisions, including whether to settle or proceed to trial.

Semantic search is a key tool in the analytics arsenal, improving accuracy by determining the searcher's intent and the contextual meanings of search terms to generate the most relevant results in a matter of minutes or seconds. Search filters and recognition algorithms are used to disambiguate legal terms that have different meanings in various jurisdictional or practice settings. Lawyers also play a vital role in descriptive analytics, constantly providing input on context and metadata, and providing quality control to leverage the technology's ability to “learn.”

Advanced data visualization is a vital component in descriptive analytics. Visualization helps users quickly identify patterns, trends, relationships and insights behind the data that would be nearly impossible to aggregate with text-review alone, thereby increasing the accuracy and efficiency with which professionals comprehend relevant data and make key decisions. Studies have concluded that the human brain processes images at a rate 60,000 times faster than it processes text. Visualization presents information in a way that is immediately accessible, intuitive and interactive, turning complex data into easy-to-digest charts and graphs. From search term mapping to presentation of legislative bill progression to review of patent images, data visualization offers legal professionals the opportunity to draw insights more quickly and efficiently.

Predictive Analytics

If the primary focus of descriptive analytics is on what has happened in the past, predictive analytics turns its attention to what will happen in the future. This encompasses a variety of statistical techniques — including things such as predictive modeling, machine learning, regression analytics, time series models and graph analysis mining — to analyze current and historical data and make reasoned predictions about what is likely to happen in the future. Effective predictive analytics require cutting-edge technologies, intelligent algorithms and tremendous amounts of rich data from a variety of sources that can provide the broadest view of an issue or potential decision and the factors that are likely to influence it. The richer the data, the easier it is for machine algorithms to “learn,” which in turn ultimately leads to more accurate predictions.

Predictive analytics is a not a substitute for human judgment or experience. Its real value lies in the ability to determine what is likely to happen based on past behaviors and current data. This functionality may prove especially useful in certain practice areas where cost analysis is crucial. For example, in the area of medical malpractice, analytics can identify cases with similar fact patterns, and can help predict when a case will close and how much will likely be awarded. Based on this information, lawyers can make a data-driven determination on whether and how to move forward with a lawsuit, and they can typically arrive at that determination in a relatively short period of time.

Similarly, if you wanted to assess opposing counsel in litigation, you could view their win/loss ratio, clients, status as plaintiff/defendant/counterclaimant and case count, and then use that information to make fact-based predictions about your prospects in the case. If you were trying to move your case to another venue, you could analyze the judge's track record to determine whether he or she is likely to respond to a motion to transfer, assess the judge's grant/deny rates, and estimate how long a case might take to get to a grant of a permanent injunction, trial or termination.

Using predictive analytics in patent portfolio management, patent portfolio managers can estimate and budget the cost of patent submittal and patent prosecution — and do so with an 80% a level accuracy. The use of predictive analytics in forecasting passage rates for legislation can drive even higher rates of accuracy and allow companies to make decisions about deployment of lobbying resources.

Other industries are already using predictive analytics routinely to forecast a variety of outcomes related to cost, inventory, market variability and more. Use of these solutions in the legal market is starting to accelerate as the benefits of predictive analytics become more apparent and lawyers better understand its value in strategic decision-making.

Prescriptive Analytics

The still-emerging technology of prescriptive analytics extends beyond descriptive and predictive analytics to recommend specific courses of action and to illuminate the likely outcomes of those actions. With prescriptive analytics, legal professionals will be able compare and contrast multiple “futures” based on historical trends and current data.

In the legal context, imagine an application that proactively makes recommendations on what legal strategies to employ based on outcomes of similar cases. It might ask questions of the user if it lacks sufficient information to make a recommendation. Such a tool might also conduct legal research to find relevant case citations or information that can support or refute the user's legal arguments. It might also recommend other search topics or keywords to pursue based on past searches, using techniques similar to those used by Amazon's product recommendation engine. In theory, it could even draft case arguments based on user inputs and/or case concepts, using language taken from the outcomes of similar, successful cases.

While predictive analytics provides insights into the potential futures based on data, prescriptive analytics goes one step further and provides actual advice, and it will continually refine its recommendations by tracking outcomes of actual decisions and incorporating that information in future recommendations. As analytics, machine learning, natural language processing and other deep learning technologies advance, it is only a matter time and acceptance before such technology becomes a true and trusted partner to lawyers and legal teams. Undoubtedly, these advances in technology will raise new questions related to “authorized practice of law” and push the boundaries further regarding human versus machine-driven decision-making.

The Future of the Data-Driven Lawyer

Exploding data volumes demand that legal professionals interact with data in new ways. Analytics is a growing part of the solution to this problem. It enables the transformation of data into knowledge, detecting patterns and connections which no human, no matter how experienced or diligent, can perceive without the assistance of advanced tools.

We are already very close to a future in which attorneys developing a case strategy will routinely apply a combination of descriptive, predictive and prescriptive analytics to a body of rich litigation data to accurately estimate variables like time to trial, the potential value of a case and likely outcomes.

Legal teams will be able to generate a visual snapshot of a judge's past cases and previous rulings, quickly determine which arguments are likely to work in front of that judge in particular types of cases, and decide whether seeking a change of venue is appropriate.

They will run searches to assess the tendencies and success rates of opposing counsel, viewing graphics that map out other lawsuits pending for the plaintiff or defendant and calculating the likelihood of settlement in those lawsuits.

They will use analytics engines to produce a list of recent relevant cases, identify specific arguments that have the potential to support or refute a recommended legal strategy, and assemble the research and arguments required to draft a brief.

They will be able to determine how case outcome and value might change as different variables — such as fact patterns, case concepts and venues — change, or as their own understanding of the case evolves. These examples are just a glimpse of the future workflow scenarios that we see coming in this age of legal AI and analytics.

Whether lawyers are analyzing past information to make searching more intuitive and faster while ensuring search results are accurate and relevant, or detecting patterns of behavior in attorneys, judges and opposing parties to help determine an optimal case strategy, analytical tools have begun to fulfill the vision of data-driven legal practice. Analytics does not — and will not — replace lawyers, but it is already making the practice of law smarter and more efficient.

*****
Jeff Pfeifer is vice president of Product Management for LexisNexis. Over a 28-year career in legal technology, he has worked to introduce a series of cutting-edge solutions for lawyers and other legal professionals. He is responsible for the product development strategy for LexisNexis in North America. Follow him on Twitter @JeffPfeifer.