Case Studies

The ChallengeA national healthcare provider faced 14 related matters across 9 jurisdictions, with 11M documents dispersed across multiple vendors, databases, and case teams.Redundant Review Is a Data Problem, Not a Legal OneWith a traditional eDiscovery model, each matter would have required reprocessing, rehosting, and/or rereviewing large portions of the same data. Data insights and work product would be siloed inside individual matters and within disparate legal teams. This would severely escalate costs and drive inconsistent outcomes and operational drag.The SolutionLighthouse recognized that the problem wasn’t just data volume. It was the absence of a system that could learn across matters and apply that intelligence forward. With LighthouseIQ, counsel could take a fundamentally different approach—using a centralized, AI-backed data system guided by expert judgment, where decisions, insights, and work product flow seamlessly between matters and legal teams.AI-Backed ResultsReduced 11M documents to 90K requiring reviewReused 100K coding decisions across 14 related mattersAvoided duplicate hosting, processing, and review of 1.2M documentsEnabled instant productions from a national database with LighthouseIQ$650K in cost savings delivered with consistency and defensibility built in, not traded offBuilding Human-Guided AI at Multidistrict ScaleStep 1: An AI-Powered Data Repository, Expertly DesignedLighthouse migrated all 11M documents (from both Relativity and non-Relativity sources) into a single LighthouseIQ hosting environment. Lighthouse experts designed the repository architecture upfront to support cross-matter reuse and long-term litigation strategy.Lighthouse eliminated duplicate hosting, processing, and review of 1.2M documents.Step 2: AI Normalization and Cross-Matter MatchingWithin the repository, LighthouseIQ normalized documents and applied proprietary hashing to identify duplicates, near-duplicates, and previously reviewed content across matters. Lighthouse experts validated how matches and inherited decisions were applied, ensuring accuracy, consistency, and defensibility across jurisdictions.Lighthouse reused 100K coding decisions across matters.Step 3: AI-Guided Prioritization, Expert Review StrategyLighthouse review experts designed one strategic review plan for all 14 matters that lowered costs and maximized data reuse and cross-matter insights. Using cross-matter intelligence, IQ Review identified 150K documents (from within the 11M housed in the repository) that were most likely to be responsive across jurisdictions.This dataset was published to the national review database and fully reviewed by an experienced Lighthouse review team (trained by Lighthouse review managers) to categorize each document for both national and jurisdictional responsiveness. After review, Lighthouse copied this strategic production set to each jurisdictional database. This approach kept hosting costs drastically lower for each individual matter, while providing all local case teams with an immediate first production, well ahead of production deadlines.Out of 11M documents, just 90K required human review.Step 4: Continuous Learning Through a Human-in-the-Loop Feedback CycleAfter production, expert-approved coding decisions were fed back into the repository. LighthouseIQ automatically matched those decisions to corresponding documents across matters, creating immediate efficiencies while preserving expert intent. With every matter, the system became: more informed, more consistent, more cost-effective.‍The Results: A System That Gets Smarter Over TimeBy using LighthouseIQ, a sprawling, multidistrict litigation environment was transformed into a reusable intelligence system. The client achieved significant cost savings and faster productions, without sacrificing judgment, consistency, or defensibility.In the process, LighthouseIQ delivered $650K in cost savings.

The ClientThe client operates at the forefront of AI innovation while simultaneously navigating heightened regulatory oversight and increasingly complex civil litigation.The Legal ChallengeOver the past two years, this client has faced a sharp increase in high-profile, high-stakes litigation and regulatory investigations. Matters often involve novel technologies, modern collaboration and messaging platforms, and compressed response timelines.This was creating sustained pressure on traditional eDiscovery models and legacy discovery tools, which proved to be inefficient and difficult to scale at the speed required. Repeated data recollection, redundant review, and inconsistent issue identification also introduced unnecessary costs and risks. The client needed an approach that could apply intelligence across matters, learn from prior work, and deliver defensible results quickly.The Lighthouse SolutionWe implemented a LighthouseIQ-driven eDiscovery program capable of scaling across the client’s litigation and investigative portfolio, prioritizing the client’s need for speed, consistency, and defensibility. Through the rapid design and deployment of this framework, Lighthouse has helped the client:Meet aggressive discovery and regulatory obligationsReduce eDiscovery risk across multiple concurrent mattersSave hundreds of thousands of dollars in just a few monthsMaintain consistency, defensibility, and institutional knowledge across a growing litigation portfolioAs new matters arise, the program continues to scale, leveraging prior AI-driven insights rather than restarting the discovery process with each engagement.Pillars of the LighthouseIQ eDiscovery ProgramIn 2024, Lighthouse launched a programmatic eDiscovery initiative for this client that was grounded in what would become the LighthouseIQ platform and application suite. The objective was to move beyond point solutions and instead create an adaptive framework that continuously improves as new matters arise. The pillars of this framework and the results achieved in just the first year are below.Reusing Work Product at Scale with LLM-Backed TechnologyLighthouse built a centralized data repository designed specifically to support work product reuse across litigation. Each matter is maintained in its own siloed workspace, where LighthouseIQ is used to:Identify when prior work product is relevant to new mattersReuse review decisions, key documents, and privilege determinationsControl reuse across matters while maintaining strict, matter-level silos for privilege and confidentialityThe result:Reduced unnecessary recollection and reprocessing across litigation by over 10 terabytesSaved tens of thousands of dollars by minimizing duplicative attorney review while improving cross-matter consistencyAccelerating Fact Development Under Regulatory DeadlinesThe impact of LighthouseIQ has also been pronounced in matters requiring rapid issue and document identification under regulatory pressure. In a recent regulatory inquiry, outside counsel had only days to identify critical facts from hundreds of thousands of documents. This timeline would have been impossible to achieve using traditional search and review technology.Lighthouse deployed IQ Case Strategy to:Rapidly analyze hundreds of thousands of documentsSurface the key documents tied to three core legal issuesPrioritize results for attorney review within daysThe result:Reduced review costs by more than $100,000Completed the regulatory response within two weeksDelivered the documents attorneys needed within days (vs. the months it would have taken with traditional search tools), giving them more time to work on data-backed legal analysisBuilding a Defensible Forensics FoundationLighthouse also designed and implemented a centralized forensics collection program spanning all of the client’s major data sources, including:Google Vault and Google DriveSlackMobile devicesNon-standard messaging and social applicationsThe forensic program addressed nuanced challenges that arise in modern data environments, including the preservation, collection, and treatment of hyperlinked attachments, particularly where contemporaneous versions are unavailable. Lighthouse’s forensic team improved collection efficiency and defensibility by:Developing a core forensics playbook to standardize data retrieval across mattersDesigning targeted collection workflows that leverage usage and access patterns to prioritize files actually accessed by custodians, significantly reducing over-collectionThe result:Improved collection consistency across matters while minimizing unnecessary data processing and review via a repeatable forensic program

Background Regulators issued a sweeping investigation tied to a global company’s high-profile acquisition. The scope and timeline were demanding: more than 30TB of data required analysis, risk assessment, and production in less than 30 days. A defensible, scalable approach that met regulatory requirements while controlling cost, mitigating risk, and ensuring flawless execution was non-negotiable. The Lighthouse Approach LighthouseIQ eliminated unnecessary human review, decreased scope early, surfaced risk faster, and executed at scale without errors. Key elements included: IQ Review used AI to surface only what truly required human judgement. In parallel, a 300-person managed review team was rapidly ramped to handle the doc volume. IQ Priv accelerated privilege identification and used generative AI for privilege log drafting and names list creation. Key documents identified via modeling in parallel to review. Cross-matter analytics and work-product reuse across a parallel antitrust litigation Custom operational workflows, including M365 cloud attachment linking and secure reuse repositories All of it executed in parallel. No bottlenecks. No rework. Results 10TB, including 20M images were produced with M365 cloud attachments as required by regulators in under 60 days. This delivery boasted an 100% error-free production result and $20M in total cost savings. Cost avoided:

When an engineering partner suddenly pulled out of a major project, a global manufacturer needed answers fast. Was the termination allowed under the contract, or had the partner crossed a line that could lead to litigation? The company’s law firm had to move quickly. A deadline was approaching to file a termination claim, but that was only the first step. Once the partner responded, the firm expected tough follow-up discovery. To be ready, they needed to understand the full story before the dispute escalated. The firm identified and collected more than one million documents across fifteen custodians, most in the United States. While this is a large but not uncommon volume of data for such a complex investigation, the real challenge was determining how to interrogate it quickly without iterating dozens of times on keywords as is the case with traditional keyword search. As one attorney explained, “Most of the time, we don’t know the exact words people used and everyone uses different language anyway.” Every guess costs time, and every missed variation risks overlooking critical evidence. The Lighthouse ApproachLooking for a faster and more reliable approach, the firm used Lighthouse IQ Answers directly inside their Relativity environment, starting with Microsoft 365 data from the U.S. custodians. IQ Answers is not a general-purpose chatbot. It’s an enterprise AI tool that leverages large language models and other AI and ML models to answer questions but is grounded solely in the documents in your case. Instead of building complex keyword searches, attorneys simply asked questions and received clear, document-backed answers. Using this approach, the team conducted an early case assessment without relying on months of manual review. Once all documents were loaded, they used the AI to explore the data directly. Over the course of less than two months, the team asked 182 natural-language questions. That process captured 6,325 documents, of which the team flagged 835 as potentially important. To confirm the results, the firm conducted a second-level manual review of those documents. Attorneys validated 190 documents as key evidence and identified another 130 as potentially key. Notably, 832 of the 835 documents directly related to the 14 issues identified for the case. By combining AI-driven discovery with focused human review, the team turned an overwhelming volume of data into clear, actionable insight—delivering results in a fraction of the time required by traditional methods. Based on the intelligence IQ Answers delivered, the firm made a critical strategic decision: they opted against a full review. What would have been months of traditional document review and significant expense became a targeted, AI-driven investigation that gave them exactly what they needed in pre-litigation.

BackgroundThe client faced a high-stakes Hart Scott Rodino (HSR) Second Request with tight compliance deadlines under FTC oversight:7.5M documents (7.8TB) were collected in 2 phases from 28 custodians collected across email, collaboration platforms, mobile data, and hard copy sources.Differentiated responsiveness standards between groups of custodians, requiring tailored review strategies.The matter was re-opened months later and additional documents requested.The Lighthouse ApproachThe team implemented both IQ Review and IQ Priv, combining AI analysis with disciplined managed review execution. Key elements included: AI-supported relevance review and a team of 30 contract attorneys for documents that required eyes-on review Privilege review, privilege log and names legend automation via AIAI image analysis for visual and scanned contentTwo separate AI models were trained to address differing responsiveness criteria across custodial groups, ensuring precision without sacrificing defensibility.ResultsLighthouse successfully processed 7.5M total documents across both collections. Because our AI models remain largely stable even with new documents, analysis of the second phase of collection was able to start immediately. LighthouseIQ powered analysis meant that only 2,500 contract review hours were needed in total. Using AI insights, our contract review attorneys maintained high review velocity across responsiveness, privilege, and PII review streams. The approach delivered FTC-ready defensibility under close regulatory scrutiny while enabling rapid adaptation to an evolving regulatory scope. Through expert coordination across legal, technical, and review teams, the engagement delivered predictable, consistent performance even under compressed timelines and shifting requirements.

Key Events and OutcomesClient initiated an internal investigation into executive misconduct, requiring high-precision document discovery and behavioral analysis.Multiple workstreams delivered: thematic overviews, interview prep kits, and targeted behavioral evidence.Lighthouse Expert Search team employed across multiple time zones enabled seamless adaptation to shifting priorities.The Expert Search team delivered 160 total key documents over four days in three deliveries to accelerate the time to knowledge and minimize the risk of missing critical evidence.Enabled counsel to prepare targeted witness interviews by surfacing behavioral evidence and operational insights.What Was NeededA large retailer launched an internal investigation after receiving whistleblower allegations of misconduct. The project required rapid, high-precision document discovery and behavioral analysis across a substantial volume of internal communications. The client needed thematic overviews of key communications, curated document sets to support interview preparation, and targeted behavioral insights to inform legal and internal review. All work had to be completed within a single week to enable critical witness interviews and support preparation of a summary report for outside counsel.ProcessThe first step in triaging the needs related to the matter involved outside counsel conducting initial research using Lighthouse AI Search. This early facts assessment confirmed that the central concerns of the investigation were reflected in the data, and helped refine the goals and targets for a hand-off to the Lighthouse Expert Search team.Leveraging insights from counsel’s initial use of AI Search, the Expert Search team used advanced techniques developed within Lighthouse’s proprietary systems to support Key Document Identification workflows. These methods, combined with tagging and document filtering workflows for compliance and legal review, targeted queries of linguistic patterns, indicators of tone and behavior, and other expressions of language.In parallel, Lighthouse AI Search powered conceptual and semantic queries, surfacing nuanced patterns and sentiment indicators across a voluminous set of 300,000 documents. The combined technology and workflow approach allowed the Expert Search team to precisely identify the documents of highest importance and potential impact for the investigative team. Volume-reduction methodologies were applied to isolate the most likely relevant, non-duplicative content. Linguistic and behavioral searches focused on topics prioritized by counsel, with results delivered on a rolling basis to support interview preparation and legal review.Throughout the project, the Expert Search team worked in close coordination with the matter team, leaning on global coverage to provide seamless support and incorporating feedback into iterative search cycles.Expert Search ResultsThe Lighthouse Expert Search team delivered three waves of curated document sets totaling approximately 160 records, each tagged with Expert Search topics and annotation fields to support rapid review. This enabled highly targeted interview preparation by surfacing behavioral indicators, communication patterns, and operational insights relevant to the investigation. The outputs integrated into client workflows, including saved searches and coding layouts within the review platform.By combining multiple information retrieval, analysis, and synthesis technologies and augmenting with human expertise, the team surfaced unique documents responsive to similar lines of inquiry—providing broader and more comprehensive information coverage in a shorter time frame than any single approach could have achieved alone.

A global consumer products company with a distributed workforce needed to strengthen its information security posture. With sensitive intellectual property, regulatory obligations across multiple jurisdictions, and increasing use of Microsoft 365 collaboration tools, the security team sought a more resilient approach to protecting critical data against leakage, misuse, or unauthorized access. Challenge The existing environment lacked unified policies for sensitivity labeling, retention, and data loss prevention, making it difficult to enforce consistent governance across all business units. The client faced significant risks around: Data leakage from collaboration data in Microsoft Teams, SharePoint, and OneDrive. Lack of consistent data classification leading to overexposed sensitive content. Insufficient DLP controls for email and cloud-based sharing, creating regulatory and reputational risks. Growing compliance pressure across global operations, requiring alignment with GDPR, CCPA, and industry-specific regulations. Solution Lighthouse partnered with the client to design a comprehensive Microsoft Purview Information Protection and Data Loss Prevention (DLP) framework pilot that could scale globally. The solution included: This design provided the foundation for both proactive risk reduction and reactive incident handling. Results Through this engagement, the client achieved: Reduced risk of data exposure by applying consistent labeling and DLP rules across collaboration platforms. Improved regulatory compliance by aligning information protection policies with global privacy and industry frameworks. Enhanced incident visibility with reporting dashboards and adaptive policies that alerted security teams to high-risk events. Sustainable governance model enabling scalability as new collaboration tools and AI-driven workflows are adopted. Why It Matters As global enterprises accelerate digital collaboration, data security gaps in Microsoft 365 environments can create regulatory, financial, and reputational risk. By implementing a comprehensive governance and DLP framework, organizations can protect their most valuable assets: intellectual property, customer data, and regulated records, while enabling employees to work securely across borders. This project highlights how a well-designed information protection program, supported by Microsoft Purview, can simultaneously strengthen security and simplify compliance for multinational companies.

The most complex matters test outside counsel on every front with overwhelming data volumes, relentless deadlines, and multi-million-dollar outcomes on the line. Lighthouse is purpose-built for these moments. By uniting innovative AI with award-winning expertise, we give outside counsel the speed, precision, and confidence required for the most demanding matters, from high-stakes bankruptcies to sprawling multidistrict litigation and everything in between. So, when a massive second request hit with unforgiving timelines and zero margin for error, our team was ready. Here’s how we delivered speed, precision, and $20M in savings when it mattered most. Proven Performance Under Pressure The Challenge 30+ TB of data 2-month deadline A parallel private antitrust litigation The Lighthouse Advantage: AI + Expert Operations Lighthouse leveraged our proven approach for complex, high stakes matters: advanced AI integrated with skilled operations, review, and subject knowledge expertise: AI at the Core: $11.9M saved on review and production costs Lighthouse’s predictive large language models (LLMs) narrowed the responsive set by over 2M documents and helped reduce privilege review by 45%. Our generative AI then accurately drafted more than 100K privilege log entries, and identified and normalized 15K names and titles. $4.8M saved on key document identification Our experts used AI modeling to quickly surface documents that posed an antitrust risk, surfacing the 270 most important documents out of 4.4M documents in just three weeks. This work eliminated the need for traditional key document searches and issue tagging. No Rework, No Waste $3.5M saved on cross-matter work To support the parallel antitrust litigation, Lighthouse built a secure repository for cross-matter work product reuse and used AI to repurpose work across 680K documents. This work drove consistency and minimized duplicative review between the two matters. Operational Execution: 2 months Scaled and managed a 300-person review team Processed 30TB of data Produced 10TB+ and 20M+ images—error free Implemented a custom workflow to link M365 cloud attachments Why It Matters Second requests are a stress test, but they’re not the only time attorneys face massive data and unforgiving deadlines with millions of dollars at stake. When the pressure is highest, Lighthouse delivers. With AI, expertise, and operational discipline, we give outside counsel what they need to handle every complex matter.

Challenge Antitrust regulators issued a broad, high-stakes HSR Second Request to investigate a global company’s high-profile acquisition. To comply, the company and their outside counsel were faced with analyzing 30+ TB of data in less than a month.Solution  Lighthouse developed an AI driven approach powered by Lighthouse’s proprietary Large Language Models (LLMs) to eliminate relevance review, substantially reduce privilege review, perform privilege logging and assemble the names list, and identify key documents to mitigate risk without linear review.Simultaneously, Lighthouse operational teams executed multiple custom workflows to process and produce over 10TB+ of data and 20M+ images in 3 weeks—flawlessly.Their work included building a custom linking workflow for M365 cloud attachments and a secure data repository for work product reuse in a related antitrust litigation, while ramping a 300-person linear review team via Lighthouse’s Managed Review solution to meet the aggressive production deadline.Lighthouse AI Savings & ROI Breakdown Overall, using Lighthouse AI saved $20M+ with the following workflows: TAR powered by Lighthouses AI proprietary LLMsPrivilege identification powered by Lighthouse AI proprietary LLMsPrivilege log and names list generation via Lighthouse AI proprietary LLMsKey document identification powered by linguistic modeling and AIJunk file analysisCross matter analyticsHighlights of each of these workflows and the associated ROI are below.  Lighthouse’s Proprietary Predictive AI for Relevance: 40% Narrower Responsive Set Than Other TAR Tools Counsel only needed to review around 4,000 documents to stabilize and validate a TAR model backed by Lighthouse’s predictive LLMs that measured 85.91% precision at 76.49% recall. Lighthouse’s predictive AI model for relevance is shown to deliver a 30-40% smaller and more accurate R-set based upon comparative bake-offs. Had Lighthouse implemented a Second Request workflow using commercially available TAR tools, we would have likely seen a much broader responsive set, thereby requiring additional privilege review and production.Specifically, if Lighthouse had used traditional TAR tools, it would have resulted in an estimated additional 2M documents in scope for production, translating into approximately $5.8M in additional privilege review costs and $400,000 in added production expenses—costs avoided by using Lighthouse Responsive AI.Separately, Lighthouse’s Review Management team performed Junk File analysis and identified that around 95,000 documents out of the more than 860,000 non-TAR eligible documents were highly likely to be junk and could undergo a sampling workflow instead of a full linear review.Privilege Review: 45%+ Reduction  Lighthouse built an AI model for privilege that both removed documents from privilege review and accelerated the remainder. Using a tiered approach, counsel determined that around 780,000 documents (out of the more than 1.7M eligible for privilege review) could be removed from privilege review without linear review because they fell below the cutoff score and were unlikely to be privileged.Once the privileged documents were identified, Lighthouse’s generative AI was used to create first-pass privilege descriptions for the more than 100,000 documents on the privilege log. This removed the need for human drafting of log lines. As a result, the log required an investment of mere hours as opposed to days and the heavy expensive of a full contract attorney review and outside counsel QC.Lighthouse also used generative AI to build the names list for the privilege log. Lighthouse’s AI model quickly analyzed around 130,000 documents to identify and provide close to 15,000 normalized names with titles for the privilege log. Key Document Identification The Lighthouse AI team used modeling to quickly surface documents that could pose an antitrust risk to the company. This process eliminated the need for the more dated approach of search+linear review for key documents and issues tags. Out of an initial tranche of 4.4M documents, the Lighthouse team identified roughly 270 documents of the greatest interest and sensitivity and delivered them to counsel in 5 deliveries over the course of just 3 weeks.To provide additional support to counsel, the Lighthouse AI team also categorized the final Responsive document set based on risk profile—classifying a broad set of documents as Likely Risky and Likely Safe over the course of just 1.5 weeks.Work Product Reuse for Cross-Matter EfficiencyThe Second Request had a large set of overlapping data with a concurrent antitrust litigation. To ensure there was no duplicative review and to drive consistency, Lighthouse built a secure data repository that enabled work product reuse between this Second Request and that concurrent litigation and used Lighthouse AI to drive cross matter analytics. With Lighthouse AI, Lighthouse repurposed calls for around 680,000 documents, resulting in a savings of $3.5M between first pass review, 1L QC, and outside counsel QC.Speed, Quality, and Operational Delivery  The scale and complexity of this Second Request required an extraordinary cross-functional effort across several Lighthouse Client Services and Operational teams who worked tirelessly to deliver a seamless, high-quality result under immense time constraints. The core of this matter was completed in under 60 days, demonstrating an exceptional level of execution.Key highlights of the operational delivery include: High-Volume Processing and Custom Workflows: Lighthouse processed over 30TBs of data, ensuring rapid ingestion, indexing, and AI-powered classification. A custom workflow was implemented to link M365 cloud attachments, meeting regulatory requirements.Flawless Data Production at Scale: Within 3 weeks, the Lighthouse team produced over 10TBs of data and more than 20M images, achieving a 100% error-free production result. This ensured compliance without delay or rework.Scalable Review Team for Complex Work Streams: Leveraging our Managed Review capabilities, Lighthouse rapidly scaled a linear review team to 300 professionals. The team expertly navigated multiple complex work streams, dynamically segmenting data to mitigate risk while accelerating the review process to meet the production deadline. The bulk of review ramped and concluded in a mere 4 weeks. This work showcases the coordination, detail, operational excellence and sheer dedication of the Lighthouse team in delivering a timely, high-quality outcome in an intense regulatory investigation.

The ChallengeA major media company received a Letter of Inquiry (LOI) from the FCC, triggering a high-stakes regulatory investigation. The company was required to produce relevant communications within a month—but two weeks in, the legal team still needed to collect over 2 million documents from 16 custodians. Complicating matters further, the company’s software platform was mid-transition, raising serious concerns about data integrity and reporting reliability. The SolutionRecognizing the urgency and complexity of the matter, the media company and its outside counsel turned to Lighthouse. With an immense volume of documents, looming regulatory deadlines, and a technology transition in progress, they needed more than linear review—they needed a strategic partner with forensic, eDiscovery project management, and AI expertise. Within just four days, Lighthouse’s forensics experts collaborated with the company to collect and process all relevant custodian data, including associated family files. From there, our project management team worked with the company and its counsel to apply targeted filtering—focusing on communications between key senders and recipients. This reduced the original 2 million documents to a refined universe of 94,000. Using advanced email threading and junk file analysis, the team further reduced the review set to 59,000 documents. Given the aggressive timeline, volume of documents, and the dataset’s low privilege risk, Lighthouse consultants recommended deploying Relativity aiR for Review. Working closely with inhouse and outside counsel, Lighthouse developed a defensible AI review prompt using an iterative sampling workflow designed to meet stringent recall standards and maximize precision. Only 300 documents were reviewed during this iterative phase. Relativity aiR identified a predicted responsive universe of 28,000 documents. A first-level review was completed in just five days, followed by a quality control review conducted by outside counsel. Final validation confirmed 88% recall and 96% precision—exceeding regulatory and eDiscovery defensibility standards. The ResultsUltimately, 18,000 documents were successfully produced on time, along with an expert declaration on the defensibility of the process from a Lighthouse Strategic Consultant. When the FCC issued a supplemental request, the teams were able to use the aiR-powered workflow once again to quickly review 2,000 additional documents—resulting in the production of the 300 relevant files.

In a high-stakes matter, a Fortune 500 CPG company and its Am Law 200 counsel faced a common eDiscovery challenge with uncommon complexity: a constantly changing review population.The ChallengeAt the heart of this large-scale matter was a significant complication: scope volatility. The Technology-Assisted Review (TAR) universe had to be redefined and recalibrated—multiple times. For most providers, this would derail consistency, timing, and cost.The Solution Lighthouse leaned into the complexity. Using a modular AI framework, we built workflows that could flex in real time; adapting to evolving populations without losing precision. Key solutions included:TAR re-scoping at scale of a 4.6M document universe without destabilizing the model Automated tracking and updates to full-family inclusions and exclusion logic Integrated PII, privilege, and image AI analysis to ensure comprehensive review coverageThe Result Despite the volatility, Lighthouse delivered enterprise-grade performance: 76% precision at 75% recall and the required documents were produced on time. What could have been a fractured and inefficient review was a streamlined effort due to Lighthouse’s ability to manage dynamic review workflows using AI. In today’s litigation and investigation landscape, moving targets are the norm. Lighthouse delivers the consistency, agility, and accuracy needed to keep pace and produce with confidence.

Client: Global academic medical system Stakeholders: CISO, Information Governance, Legal Tech stack: Microsoft 365 + Microsoft Purview (SharePoint, OneDrive, Exchange, Teams) Objective: Identify, label, and protect high‑value IP across M365 Business Challenge Conventional pattern matching missed nuanced research content; labels were inconsistent. Emerging IP taxonomy lacked consistent, enforceable labels across repositories. Conventional pattern matching couldn’t reliably detect unstructured, nuanced IP. Teams needed clarity on when to use Sensitive Info Types (SITs), Exact Data Match (EDM), and Trainable Classifiers, and how to govern them. Wanted to compare outcomes with prior third‑party classifiers. What Lighthouse Did IP Taxonomy + Purview Labels Mapped proprietary IP categories to a label set Blended Classifier Strategy Combined SIT, EDM, and Trainable Classifiers Operationalize Piloted and tuned models aligned with retention/legal hold/eDiscovery, with auto‑labeling and user prompts. Controls Developed change‑management materials Outcomes Common IP Language: Agreed taxonomy mapped to enforceable labels. High‑Confidence Detection: Trainable classifiers surfaced custom IP Consistent Protection: High‑value content auto‑labeled with policy‑driven controls in M365. Governed Workflows: Clear guidance on SIT vs EDM vs Trainable; fewer false positives/negatives; faster to eDiscovery. Timeline Weeks 0–1 — Kickoff + Plan Weeks 2–4 — Design + Setup Weeks 5–9 — Run Pilots Weeks 10–11 — High Level Design + Training Why Microsoft Purview for Data Protection Enterprise-wide strategy - Unified data security, governance, compliance Integrated governance - DLP, retention, legal hold, eDiscovery Flexible detection models - Sensitive Info Types, Exact Data Match, Trainable Classifiers Persistent, label-based protection - Embedded permissions travel with data

When a Fortune 500 CPG company faced a second request tied to a business divestiture, its Am Law 200 counsel partnered with Lighthouse to manage the heavy data lift of 4.6M docs. With tight deadlines, shifting custodians, and complex privilege concerns, the stakes were high—and consistency was critical.The ChallengeVolatile TAR populations and changing custodian scopes Compressed timelines typical of second requestsHigh data volume across diverse file types including video and image The Solution Lighthouse deployed a suite of AI-powered tools to keep the matter on track: AI-driven TAR and privilege/PII detection to streamline review Generative AI + transcription tech enabled fast image/video analysis Modular workflows allowed rapid pivots as scope evolved The Result The AI-led approach let attorneys focus on strategy while Lighthouse handled operational complexity—proving the power of collaborative, AI-enabled review. The matter was completed on time and to the satisfaction of all parties involved. The legal team met their regulatory deadlines, and the CPG client was well-positioned for the next stages of its divestiture. Why It Matters This engagement shows how scalable AI, smart workflows, and strong partnerships can turn regulatory chaos into consistent, repeatable success.