Client Profile
  • Multi-billion dollar semiconductor corporation
  • 10,000+ employees across 50+ offices worldwide
​Pain Points
  • An internal HR investigation involving 60,000 documents, mostly in Japanese
  • A two-week deadline to complete a complex task
  • A busy in-house general counsel with little available time to shepherd the review process
  • A limited budget
Solution Profile
  • Level Legal’s data management expertise and Japanese language skills accelerated the technology assisted review process
  • Casepoint eDiscovery solution was used for data ingestion, processing, early case assessment (ECA), advanced analytics, review, and production.
  • Casepoint’s CAL feature enabled Level Legal to provide quick and effective multi-lingual review with a streamlined team.

Even without the time crunch, the project was intricate. It involved seven distinct subject categories, the review of more than 60,000 documents and over 70 gigabytes of data, and dual language review since the majority of documents were in Japanese. Without the appropriate technology and expert selection and administration of the tools, the project ran the risk of absorbing far more time and budget than English-only review.

​Key Outcomes
  • Budget and timeframe were accurately estimated on Day One
  • Potential review population was reduced by 97%
  • Reduced population enabled completion of the matter a full week ahead of schedule
  • Costs were substantially below the client’s top-end estimate
  • The project was completed with minimal disruption to in-house counsel

After identifying the client’s needs, the Level Legal team chose Casepoint’s Case Assist e-discovery tool, Continuous Active Learning, as the ideal technology to deliver results rapidly, drive down cost and improve efficiencies.

Level Legal leveraged their own team’s expertise in administering the Casepoint technology to shorten the review timeline and cull the data to manageable levels. By using team members with significant investigation experience in both the English and Japanese languages, Level Legal eliminated a time-consuming learning curve and boosted the effectiveness of the Casepoint tool.

This enabled the team to power through material at a faster and more accurate rate, quickly find key and relevant documents, and very rapidly identify the review population. This in turn enabled Level Legal to meet the client’s initial concerns – “How long and how much?” – by accurately predicting, on Day One, the volume and pace of review and the level of budget this would require.

Alleviating in-house counsel’s concerns about workload, Level Legal managed all aspects of review from data preparation to administration, review and monitoring of the AI tool, to key document handoff – all with minimal disruption to counsel. The team moved the matter forward swiftly and effectively by continually reviewing and monitoring the AI model, timeline and budget. Counsel received daily reports regarding the progress of the project.


By aligning their strategy with that of the client’s inside counsel, Level Legal’s eDiscovery experts and technologists acted as a seamless extension of the client’s team, enabling them to continue their focus on day-to-day priorities.

Level Legal’s deep knowledge of foreign language review and experience administering advanced technology allowed the team, within 24 hours of project start, to establish hard timelines and precise budget projections significantly below in-house counsel’s top-end estimates.

The initial cull reduced the review population by 97%, achieving an exceptionally small but highly relevant document set. Out of more than 60,000 documents, fewer than 2,000 were found to be relevant. This efficiency resulted in the matter being completed within one week, well ahead of schedule and substantially under budget, and at a fraction of the cost of traditional multilingual review. The fusion of Level Legal’s data management expertise and Casepoint’s advanced analytics capabilities was the determining factor in accelerating the technology-assisted review process.