Analyzing the vast – and ever-increasing – volumes of data needed to process a litigation, investigation and / or regulatory request can be a challenge for legal teams.
Technologies such as artificial intelligence (AI) and machine learning can provide lawyers invaluable support when it comes to analyzing data. Together, the technology solutions help clients and their advisors find the relevant data needed in documents to get their jobs done faster and more cheaply than ever before. Technological tools are continually developing, increasing their analytical power and their ability to reflect the ever-changing way we generate, transfer and store data.
But how do you realize long-term value when data analytics is provided to support litigation? And – most importantly – where will processes and technologies go in the future?
The current state of affairs
Currently, the data analysis and management process is focused on eliminating and analyzing the collection of data after it has been shared with legal teams. Here, a careful balance must be found. Lawyers or investigators must be thorough. They often scan vast amounts of information and need to be sure that they have taken all reasonable steps and are considering relevant data sources; while managing time constraints and limited resources.
Individuals sifting through conversation threads, across multiple platforms that can go back years to find a document (albeit vital) will exacerbate existing time constraints. The challenge, therefore, is to find effective ways of bringing together all relevant data sources; structure and standardize it so that a simple and consistent analysis can be applied; and in turn intelligently identify what is and is not likely to be relevant.
There are a plethora of solutions that currently help do this through powerful data analysis software platforms. For example, “email threading” works by identifying relationships between emails in a large data pool and then pulling them together into a single combined conversation, often allowing for quick later review.
“Conceptual search,” which refers to an automated search that uses an analytical index to derive a meaning, allows investigators to use an existing sample of text to find other documents with similar conceptual content, rather than simply search for fixed words or phrases.
And ‘continuous active learning’, which refers to models that learn and evolve based on input from human reviewers, can be applied to new data as additional sources are added, while retaining information previously learned. This approach ensures that attorneys’ / subject matter expertise and human acumen are applied in the most cost effective and timely manner to potentially huge document volumes.
While this requires a certain investment of time on the part of the case team, the overall time savings can be significant. Once operational, the technology can operate entirely independently – extrapolating from human behavior and continuous inputs to generate its own recommendations on what is likely to be interesting and what is not.
The road ahead
These data discovery tools and processes are delivering exceptional results today. But there are areas where we believe we will see future changes and developments. One is data governance.
As mentioned, today’s discovery processes are overwhelmingly focused on managing the data once it has been collected. It’s common for businesses to not always fully understand where information is kept in their operations, what it refers to, and who it belongs to. Therefore, delay the start of the data analysis element of the question, as the data must first be identified and mapped before the start of any collection exercise.
A key step in providing more efficient processes will be to encourage more stringent data management at the pre-collection stage. This may require an initial investment of time and money on the part of businesses, but makes it much easier to find information when the time comes.
More broadly, the challenges of data management are increasing, driven by changes in the way people communicate. People no longer perform reliably on a single device or platform. Increasingly, we are seeing discussions starting on a single channel, like email, and then moving on to instant messaging or the phone.
This has been exacerbated by the COVID-19 pandemic. Along with an accelerated transition to doing business on digital devices, people are now using a new range of data generation tools to do their jobs – instant messaging (IM) platforms like WhatsApp, Signal or Telegram. to name a few, as well as video calls.
Capturing and combining all of this information is a challenge that legal teams will have to deal with. Technology solutions can help with this, and the process might become more manageable as more information is stored digitally in the cloud – work can begin on data stored in the cloud before it even happens. reach the analytics platform, which saves even more time – and money.
Extension of technical capacities
Right now, there are real limits to the ability of data review platforms to handle key information like numbers. Manual analysis may be required when the data contains many figures, such as a company’s general ledger, which increases the cost of a project. The development of this capacity will be essential for more efficient analysis processes.
Advances in emotional analysis – the ability to develop an individual’s emotional ‘profile’ in their communications, and then identify emotions in their messages – is another exciting area that could dramatically improve the analytical power of the machine. . And, it will be important to see the development of audio analysis capacity and to consider the growing role of video.
There are the two sources that have seen an increase in data volumes over the past year or so as the pandemic has forced business online. While the volume of audio and video recordings from platforms like Microsoft Teams or Zoom currently in play in litigation is low, it will only be a matter of time before it increases. reflecting the way business was conducted during the pandemic.
Examining audio quickly – whether it’s just over the phone or in a video call – is difficult, for the simple reason that there is a limit to how quickly you can listen to sounds. There are automated voice-to-text solutions that are flawed and can misinterpret a nuanced conversation that a human reviewer would pick up on.
Using computers to reliably and consistently review multiple audio or video elements simultaneously, while being able to accommodate natural vocal variations, factors such as accents and dialects, would save legal teams time. considerable. Ultimately, this would allow them to spend more time focusing on the parts of the analysis that computers just can’t handle.
Competence in practice
Solutions are now in place to help legal teams tackle data challenges more effectively, and further development of technology will only improve results. However, it’s important to remember that humans are at the heart of these processes.
Ultimately, there is no substitute for a well-organized legal team, backed by trained analysts who know how to make the most of the tools available. An ever-increasing reliance on technology means that sampling and validation of machine-derived results will play a key role in any project. This not only gives the legal team confidence that the AI is doing its job, but also allows them to validate their approach from the other side.