Deposition Summaries
Solving a critical, ubiquitous pain point in deposition analysis & extraction
Problem Statement
Litigation attorneys conduct multiple 6+ hour long depositions and often need to find key facts and insights from hundreds of pages of transcripts to use for motions & legal arguments. Often times, key information uncovered in depositions have a significant impact on the strength of arguments of a case
Current Workflow:
(a) Create Summary: Review each line of the deposition, manually write out key facts and include exact line of the deposition. Refer back to the facts when writing motions, during trial, prepping for other depositions
(b) Search The Transcript : Ctrl F the long deposition transcript (200+ pages) to find key information
Learn More
Current workflow of lawyers: review deposition & manually extract facts
Top Needs
Needs stack ranked by importance vs satisfaction
1. Litigation attorneys need to efficiently find information from transcripts in order to construct winning arguments in motions & at trial
2. Lawyers need to make sure the information extracted from transcripts is accurate for a judge to accept their argument (citations)
3. Lawyers need to efficiently take information from the transcript into their own legal documents to present to opposing counsel
Interface
Process
50+ user interviews to identify the JTBD’s and needs across the litigation workflow & current tooling and processes. Map out broad journey and overlay needs, empathy map.
Secondary discussion to stack tack ranking based on importance vs satisfaction within the focus area (depositions) to identify pain points to focus on.
Overlay competitive analysis on other offerings in the space for context.
Discussion on the problem space with engineering team & stack ranking of problems and potential solutions.
Build some mockups & early solutions to get further validation from users -> RICE to determine prioritied solution space.
Develop phased rollout plan -> Assess product-level metrics & continuously meet with customers to determine bugs fixes, prioritization of feature roadmap
Interface
Solution
-> Launch AI-generated deposition summaries + tooling to quickly validate facts (clickable citations)
-> [Added in post-early user feedback]: Launch export feature & clipping feature given lawyers reliance on word
-> Launch AI-enabled tooling around depositions to assist in the extraction of (e.g. AI search, testimony by topic)
Worked with CTO, BE Eng, FE eng, ML eng to launch
Phased Rollout:
Phase 0: Offering deposition summary CSV output to early attorneys
Phase 1: Building UI around deposition summaries — with clickable citations, slide-over document window, and export functionality (key for motion drafting)
Phase 2: Fact Extraction Tooling (AI-Enabled Search, testimony by topic)
Interface
Phase 0: Document Output For Summary Validation
Phase 1a: AI Enabled Deposition Summaries in UI
Phase 1b: Slide Over Document Viewer (scroll to right page)
Phase 2: AI Search: Natural Language Search Across Depositions
Key Metrics
-> 80% decrease in time to review depositions
-> 50%+ cheaper vs cost for typical workflow
-> XX law firm customers signed, 60% M3 retention
Interface