Legal firms are looking to the tech industry to gain a competitive advantage in and out of the courtroom, but the age-old advice to “walk before you run” has come to characterize the partnership. Before law firms can employ artificial intelligence (AI) and machine learning (ML) to perform Legal Research, Litigation Strategy, Contract Review, eDiscovery, or even simpler tasks such as effectively employing Case Management or Customer Relationship Management (CRM) software, they must first develop and invest in a Data Quality Strategy. Without quality data, law firms cannot create meaningful data analytics or effectively conduct thorough eDiscovery to protect their clients.
So, what is Data Quality?
We can define data quality as the reliability of a particular dataset. Quality data is complete, consistent, accurate, timely, properly formatted, and valid. There are two major complications in the legal field that prevent the creation of quality data from being easily processed, researched, and analyzed:
- Unstructured Data. Current case management and analytic software, such as Relativity, NexLP, and Tableau all have difficulty transforming unstructured data into a format that case management and data analysis tools can easily use. For example, every patent application has a unique Customer Number located on the upper right corner of the document. While it is easy to visually see the that number, law firms either have someone scan images of the documents or they have to bring in IT to write code, script, and build a parser to extract this information. This comes at a considerable amount time and money.
- Data Uniqueness. Many law firms struggle with the issue of Data “Uniqueness,” or the ability to accurately and consistent apply a single set of terms in which there are no duplicates. To illustrate how this gets complicated quickly, let us look at case documents that are processed, uploaded, and saved with the date format MM-DD-YY. At the same time, patent applications are being processed, uploaded, and saved with the date format MM-DD-YYYY. While all the documents might share the same ‘date’ like ‘June 1, 2022’ – when the documents are shared between systems the dates can quickly become a mess.
Or maybe its 06:01am on the 22nd ?
While it may be obvious to you, machines do not know how to read these subtle changes and unless you provide the instructions, it makes filing dates, emails, phone calls and any other “time stamp” questionable. It is obvious the tremendous impact a computer will end up having on your case as you gather and analyze the data.
How does one obtain quality data?
Obtaining quality data is no easy feat. Most law firms have realized they cannot rely on lawyers and paralegals to take responsibility for data quality. In the legal field, where time is money, spending countless hours reviewing and inputting data is a heavy financial burden. Additionally, two lawyers simply defining the same term differently may result in incorrect conclusions or error messages, rendering visualization and data analytic software useless.
Many law firms have hired Data Quality Specialists or Data Engineers as part of the IT or Litigation Support Team. Often these specialists are part of an outsourced support package where a vendor is tasked with the responsibilities of data cleansing and preparation, which comes with its own challenges (consistency, time, and money). In 2013, Experian Data Quality found that 59% of data inaccuracy could be attributed to human error .
How can Inonde help?
Inonde software can produce AI-ready data and automate your data ingestion, cleansing, and integration workflows. A non-coding, non-scripting approach to data preparation and integration lets anyone clean data. We design software and solutions to specifically address your data challenges and make it easy to handle hundreds of categories of data.
Whether your law firm seeks to employ advanced AI and ML techniques, or simply to use CRM and case management software more effectively, Inonde has solutions for you.