A cashier in Walmart. (Photo by: Jeffrey Greenberg/Universal Images Group via Getty Images)
Walmart: the US retailer has faced several class actions on wages and hours © Universal Images Group via Getty

From logging into a laptop, swiping a key card or turning on your mobile phone, your workplace data is constantly being harvested.

Moreover, as an increasingly sophisticated breed of employment litigation emerges, that information could be used against you, too.

Employment battles are growing ever more complex, and the volume of available data on employees is mushrooming by the day, catapulting to the fore a special kind of data-intensive employment law.

Law firms such as Littler Mendelson are using huge data sets to establish liability in disputes and to help companies work out if they are paying their workers fairly, or staring down the barrel of an impending claim.

The development has emerged as companies have faced a rash of fresh harassment and discrimination claims after the rise of the #MeToo movement and new gender pay gap reporting, as well as an outbreak of high-profile pay battles in the past two years.

Some of the world’s largest companies, including Walmart, Nike and Google have gone head-to-head with their workers in recent months. Walmart in particular has been stung by several so-called wage and hour class actions, in which employees join together to claim for missed breaks, unpaid wages and other violations governing hours.

In June, the US retailer was told by a California judge to pay $102m for giving workers illegal payslips, but was largely cleared of allegations that it had not been properly paying hourly employees for missed or shortened meal breaks.

Littler Mendelson says big companies are increasingly seeking its services to build arguments in wage and hour class actions using the data of employees. “We use client data to work out where liability sits and what people should have been paid,” says Aaron Crews, chief data analytics officer at Littler. “We can rebuild whole days to show what people were doing and when.”

In a world of structured data, the firm can harness information on badge swipes, system log-ins, GPS data and email log-ins to work out where the liability should lie, he says. “Did a certain group of employees get a meal break or not? Did they spend this much time before they started work putting on uniforms? We can build that [picture] with data.”

The firm can take disparate data sets — for example, the GPS data for a group of individual lorries, and the invoicing and routing data that launched them on the road — in order to build up a picture of a worker’s day, and assess whether their dispute holds water.

It has also developed a pay equity tool to help companies evaluate whether they are paying certain groups — for example, women — less than other groups.

The tool uses data harvested from clients’ HR systems and adds inputs such as performance reviews, bonus rules and length of tenure to reveal disparities within employee groups and establish whether workers are being paid unfairly, and why.

“With the rise of #MeToo and discussions around pay equity we decided to build a tool that made it easier for clients to understand the underpinnings of the advice we were giving,” Mr Crews says.

“We can tell where disparities exist, what’s driving them and whether they are statistically significant — meaning, is it the result of intentional decisions or the result of a pure roll-of-the-dice random chance,” he says.

With messy pay battles on the horizon in the US and internationally, including the BBC’s pay dispute with senior broadcasters Carrie Gracie and Samira Ahmed, companies are increasingly vigilant about such subjects.

This degree of high-volume data analysis has only recently become a possibility, however, as costs have rapidly reduced and cloud computing power has accelerated.

“It used to be very expensive to do terabytes of data,” says Mr Crews. “But we did a case a year ago where in three months we [harvested] 80bn data points and produced 33bn of those to the other side in essentially 90 days. “That would have been impossible five years ago.”

‘Sophisticated clients don’t want to pay 100 people to sit in a room and look at 2m pieces of paper’

high paper stack and a cup of tea

As well as hiring specialists in technology and IT systems to create specific data analysis divisions, law firms are also taking on mathematicians and computer programmers to deal with large volumes of data.

US law firm Paul Hastings set up a data science, analysis and investigation group, staffed with PhD data scientists, to develop tech tools to answer legal problems.

Tom Barnett, a lawyer and computer programmer who used to write and edit maths textbooks before taking his law degree, joined the firm five years ago.

He says: “Outside the legal industry — think of Google, for example — we are on a 45 degree curve [when it comes to learning about technology]. But law firms are on a 3 degree curve. There’s a huge gap there.”

The need for better efficiencies when it comes to legal billing and to show innovative ways of producing work are driving much of the push on recruiting new expertise, he says.

The team have devised an app to scan relevant precedents and terms for commitment letters in debt finance and capital markets transactions. It automatically creates a draft letter using similar provisions within minutes instead of hours.

“Economics drives a lot of activity in the business world and clients are demanding more efficiency and more value for the cost, and they’re using more tech,” he says. “Sophisticated clients don’t want to pay 100 people to sit in a room and look at 2m pieces of paper. They want things to be more efficient than that.”

Data modelling has also grown more sophisticated. The days when technology was used to scan documents for certain words are gone. Now systems identify patterns of words and intuit their use.

But Mr Barnett says his tools are not coming for associates’ jobs.

“We look at it as eliminating work that’s tedious, not eliminating jobs,” he says.

“The idea that AI would really replace serious complex analysis and decision making is science fiction right now, but it can do analytics and routine tasks much faster.”

The tables below rank law firms and in-house legal teams for the FT Innovative Lawyers North America awards.

Technology and Data
Rank Law firm Description Originality Leadership Impact Total
STANDOUT Littler Mendelson The firm developed a connected suite of products to draw on case and company data it collected over 20 years. Data-driven insights and recommendations for clients are offered through CaseSmart for benchmarking information. The system also deals with automated pay equity audits and employment law queries.   7 9 8 24
STANDOUT Ballard Spahr  Ran an initiative to review data on its top clients to improve relationships and ensure that projects are completed and targets hit. Access to more data about clients is driving better performance and profitability and improving efficiency. Commended: Melissa Prince 7 8 8 23
HIGHLY COMMENDED Bracewell Reviewed more than 4m items including documents, data, text and audio in the US government investigation into Michael Cohen, lawyer for US President Donald Trump for 12 years until May 2018. The firm created a special facility and set up technology and processes to ensure all information was completed quickly and protected from hackers.     7 7 8 22
HIGHLY COMMENDED Cleary Gottlieb Steen & Hamilton Created a tool to identify documents that contain privilege clauses. Using machine learning, it has an accuracy rate of 75 per cent, resulting in a cost reduction of 90 per cent for clients for this service. Commended: Christian Mahoney 7 7 8 22
HIGHLY COMMENDED Gowling WLG Built an online platform that connects applications for case management, document management and automation, and a library to allow lawyers at the firm as well as 20 other law firms to collaborate for their client, the Canadian Medical Protective Association.  7 8 7 22
HIGHLY COMMENDED Morgan, Lewis & Bockius Expanded Parallex, its proprietary software, to manage more matters related to defending asbestos litigation for its clients. The platform provides data and document storage and management. A real-time dashboard monitors deadlines, costs and billing information.  6 8 8 22
COMMENDED Ballard Spahr ClientInsight is a proprietary data warehouse accessible to Ballard Spahr clients that consolidates previously siloed databases. Using predictive analytics and machine learning, clients can accurately predict their legal expenditure.   7 7 7 21
COMMENDED Paul Hastings Cut the time to draft a letter from 4.5 hours to 1.5 minutes, thanks to a Paul Hastings-designed artificial intelligence tool that identifies relevant precedents and terms for commitment letters in debt finance and capital market transactions. 6 8 7 21
COMMENDED Holland & Hart Upgraded Respondent, a proprietary technology, to automate responses to initial rejections of patent applications, cutting time spent and human error. The software is being beta-tested by six other law firms.  7 7 6 20
COMMENDED Holland & Knight The firm uses HK-Adapt in its insurance fraud recovery team. The technology can extract data from dozens of formats to create reports or data visualisations. The processing of tens of thousands of documents was previously carried out manually.  6 7 7 20
COMMENDED Kirkland & Ellis The firm created risk models and dashboards to help clients assess anti-money laundering and workplace compliance risks. It also created a service to monitor a company’s risk over two years to ensure compliance.  6 7 7 20
COMMENDED McGuireWoods  Built a data analytics program to analyse thousands of documents in a client’s class-action lawsuit. The analyses helped determine different settlement amounts and challenge class certification.   6 7 7 20
Technology and data (in house)
Rank In-house legal team Description Originality Leadership Impact Total
STANDOUT Juniper Networks The network product provider’s legal operations team rolled out the Legal Analytics Management Platform (Lamp), which presents data from disparate sources in interactive dashboards allowing the legal team to use the data in various ways including grey market risk detection. This analyses the company’s data to identify which products are most susceptible to being sold through unauthorised distribution channels and where this most commonly happens. These insights have enabled the in-house legal team to pursue settlements. The tool is also used for other purposes by the wider business.  8f 9 8 25
STANDOUT Microsoft Implemented a compliance analytics programme that uses digital technologies such as machine learning and artificial intelligence to tackle corruption and other compliance issues. The system analyses sales transactions with the tech company's 250,000 third-party sellers, in order to detect any that are potentially high risk. This allows the company to identify and prioritise risk, and to allocate compliance resources more efficiently.  8 8 8 24
HIGHLY COMMENDED Bank of Montreal Started using AI and machine learning to automate compliance tasks. Initially used for one customer protection problem, it is expanding to other areas. The in-house team have also introduced real-time risk assessments, eliminating an annual six-week process, saving approximately 40,000 hours a year. 7 8 8 23
HIGHLY COMMENDED Lime Built a custom law enforcement portal that ensures prompt responses to global law enforcement requests. This has helped the micromobility company build better relationships with city governments. 7 7 8 22
COMMENDED Cisco Systems The networking equipment maker's Employee Digital Footprints system is a single interface that aggregates data from 12 sources, including training records, gift disclosures and expenses. The tool, which replaces a manual process, is used in internal employee investigations and to identify compliance risks. 7 8 6 21
HIGHLY COMMENDED Flex The electronics manufacturer implemented a legal-budget management system that aggregates financial data from various systems across the company to present it in interactive dashboards. The tool has enabled the team to measure and show how they have reduced spend. 7 7 7 21
COMMENDED Walmart The US retailer expanded use of AI to draft automatic responses to complaints, lawsuits and discovery requests. Deployed in California last year, the tool is now used in six states and the team are running a proof of concept analysing the requests to gain insight into where disputes arise and the underlying causes. 7 8 6 21
COMMENDED Bristol-Myers Squibb The team have developed two chatbots, one in records management, which answers questions from internal employees using robotic process automation, and one for docketing, which reads emails and dockets correspondence in a database. The chatbots replace manual processes. 6 7 7 20
COMMENDED eBay Developed a custom software tool that aggregates data from Slack Technologies, the workplace messaging service,  to improve ediscovery at the  ecommerce company. 7 7 6 20
COMMENDED General Motors Developed LexGauge, a budget management technology platform that also aggregates financial and legal data. The platform allows lawyers at the carmaker to access data quickly and easily in a one-stop-shop. 6 7 7 20
COMMENDED IBM The US technology group created a legal chatbot to help the wider business answer common legal questions. The tool uses IBM's Watson AI to score how certain the bot is of the answer and is accessible via the Slack workplace messaging system.  6 7 7 20

Explore the Innovative Lawyers North America rankings 2019


  • Most innovative law firms
  • Most innovative in-house legal teams
  • Rule of law and access to justice
  • Collaboration

Business of Law

  • Technology and data
  • New business and service delivery models
  • New products and services
  • Talent, strategy and changing behaviours
  • Diversity and inclusion

Legal Expertise

  • Accessing new markets and capital
  • Creating a new standard
  • Enabling business growth and transformation
  • Litigation and disputes
  • Managing complexity and scale
Copyright The Financial Times Limited 2023. All rights reserved.
Reuse this content (opens in new window) CommentsJump to comments section

Follow the topics in this article