NetDocuments releases benchmark report demonstrating how legal context drives more effective AI
NetDocuments®, the #1 trusted intelligent document management system for legal professionals, today published the Legal Context Engineering Benchmark (LCEB) report, a new benchmark that measures how structured legal context affects both the quality and cost of AI answers.
The benchmark asks a simple question: what changes when the model, agent, questions and documents stay the same, but the AI has structured context to help it understand and navigate those documents?
The results: the Legal Context Graph cut the cost of a correct AI answer by 48% while maintaining essentially the same answer quality. For a 2,000-person firm asking four million AI questions a year, the report estimates that could translate to nearly $1 million in annual savings.
“As AI shifts to consumption pricing, the industry is swinging between two extremes: spending without limits and cutting without strategy,” said Josh Baxter, Chief Executive Officer at NetDocuments. “We’ve believed from the beginning that context is the lever that changes that equation. This benchmark is the first proof. Get the context right, and firms spend less to get better answers. That’s not a feature. That’s a new foundation for how legal AI should work.”
The LCEB compares the same AI agent answering 300 questions across ten real legal matters, with and without structured legal context. The report also publishes the methodology, giving legal organizations a framework they can use to measure the quality and cost of their own AI answers over time.
“Over the past six months, the conversation in legal has shifted from AI capabilities, which are ubiquitous, to context, which is essential for quality AI outcomes,” stated Scott Kelly, Vice President of Product at NetDocuments. “This benchmark gives firms the confidence to ensure AI investment creates lasting value for their lawyers, their business and their clients. We’re moving the conversation from assumption to evidence by demonstrating how structured legal context can help firms get significantly more out of their connected AI tools at lower cost.”
Making an impact
NetDocuments has used the benchmark to evaluate the impact of its groundbreaking Legal Context Graph. Announced in May, the graph turns a firm’s documents into structured legal context that AI can use from the start.
Giving AI access to documents is not the same as giving it context. AI may be able to search a firm’s documents without understanding which matters are related, which precedent the firm relies on, which version of a document matters, or how information scattered across a matter connects. Without that understanding, the agent has to spend time and compute figuring out relationships the firm already knows. The Legal Context Graph gives AI persistent, permission-aware context so it can start with that understanding instead of rebuilding it with every question.
The benchmark revealed two ways firms can use the Legal Context Graph:
- Save. With context, the same AI delivered the same answer quality at 48% lower cost. For a 2,000-person firm asking four million AI questions a year, that could translate to nearly $1 million in annual savings.
- Improve. Firms can reinvest some of that efficiency and let the AI reason more deeply. In the benchmark, that improved answer quality by 7% while still costing 18% less overall.
Token use per answer fell 52%, and the benefit grew as model capability increased, suggesting that better models make investments in context even more valuable. The benchmark covered 300 questions across ten real matters: 874 documents and about 60 million characters drawn from public regulatory filings and court dockets.
Context is crucial
The benchmark reflects a broader shift taking place across enterprise AI. As organizations move beyond experimenting with models and focus on production deployments, the quality and efficiency of AI increasingly depend on how context is delivered, not simply on the model itself.
Gartner’s Insights Abstract reinforces this trend, identifying context accumulation as the primary lever for simultaneously improving AI quality and controlling costs as agentic AI adoption grows. At a time when AI is shifting to token-based pricing, Gartner notes that assembling context is one of the largest cost drivers for AI agents, making persistent, reusable context a strategic advantage rather than an implementation detail.
MCP Connects AI to Legal Context
NetDocuments’ Model Context Protocol (MCP) integrations share information in context, going far beyond simply creating a connection for documents to move back and forth. This improves accuracy by providing AI with the right context from the start, instead of forcing every AI tool or agent to reconstruct context with every interaction. The result is more reliable AI outputs while reducing the inefficiencies and costs associated with repeatedly rebuilding context. This growing connector ecosystem spans frontier AI platforms including Claude, ChatGPT, Microsoft Copilot and Perplexity, alongside legal technology providers including Harvey and Legora.
Availability
The Legal Context Engineering Benchmark report is now available here.
At ILTACON 2026, NetDocuments will present benchmark findings demonstrating the graph’s impact and showcase the growing ecosystem of AI tools that can access this context. Join the NetDocuments Company Update at ILTACON on Tuesday, August 25 at 11am in Tennessee Ballroom C and visit NetDocuments at Booth 603. Click here to see other ways you can find NetDocuments at ILTACON.
