How McCarthy Holdings, Inc. transformed dispersed company knowledge into AI-powered advantage with Glean
2 hrs/week saved per employee, company-wide
Introduction
McCarthy Holding Companies, Inc. (McCarthy), holding company of McCarthy Building Companies, Inc., has been solving complex construction challenges for more than 160 years. Operating in nearly every U.S. state, the firm builds hospitals, data centers, government facilities, and more—at scales ranging from less than $1 million to over $1billion. With approximately 4,000 salaried employees in offices and on active jobsites nationwide, the enterprise applies the same meticulous planning to technology that it brings to every build: each workflow and tool must measurably improve safety, efficiency and decision-making across the company.
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.
With 100 million files covering projects stretching back decades and detailed documentation for every role in the company, McCarthy employees identified a growing challenge accessing information across many distributed tools. Critical information — policies, procedures, project records, material specifications, historical data — was dispersed across SharePoint, Microsoft Teams, Procore and other enterprise systems. Employees were worried about losing time searching for answers that existed somewhere in the organization but weren’t readily reachable.
McCarthy’s digital team evaluated Microsoft Copilot as part of a broader push toward AI-powered knowledge access. Copilot handled note-taking and summarization well, but its search and retrieval experience was opaque and it lacked the LLM flexibility and user experience that the team envisioned.
After exploring alternatives and running a head-to-head POC, they chose Glean.
Launching Glean
Easy to learn, built to discover
When Glean went live, the contrast was striking. The interface was more intuitive, search worked across systems and Glean’s enterprise knowledge graph — built automatically from McCarthy’s own data rather than dependent on Microsoft’s graph — gave the team a level of control and transparency they hadn't had before.
“Everything just kept being a light lift,” recalled McKenzie Lewis, AI Platform Manager.
For a Microsoft-centric organization, moving to Glean wasn’t about replacing existing tools. It was about adding a secure layer that made all of them more useful. Glean’s LLM-agnostic architecture meant McCarthy wouldn’t be locked into a single AI provider as the technology evolved. And direct access to Glean’s development team meant responsive support when it mattered.
From the start, Glean was positioned as an AI experimentation layer internally — a governed environment where employees across the enterprise could explore what AI-powered work actually looks like in practice. Corporate team adoption quickly ran to 90%. Field and operations teams reached almost 60% in months, a meaningful number for a workforce spread across active jobsites nationwide.
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.
With Glean
Search delivers instant value
For McCarthy employees, the immediate impact was simple: finding information faster. Knowledge that once required digging across multiple systems now surfaces in seconds. In one case, a user located a file containing a 14-year-old 3D building model almost instantly — a search that previously demanded significant time from multiple people. Across the enterprise, McCarthy estimates a conservative two hours saved per employee per week, enabling every employee to make decisions faster.
Internal survey data found that Glean saves each employee 1.5 hours per week on information searching. With 80% of employees regularly using Glean, Zillow teams across HR, Engineering, UX Research, and more are able to prioritize higher value activities in their day to day workflows.
Agents across the business
With Glean's agent capabilities, McCarthy teams are now building and deploying purpose-built agents that automate complex, data-intensive workflows. Their teams have launched a number of core and advanced agents, including:
Their most-used agent logged more than 40,000 runs in its first two months, helping employees across every department comply with company guidelines.
These agents give employees instant, accurate answers to relevant training, benefit questions and procedural documentation, cutting reliance on HR support and reducing onboarding time.
An agent sends a proactive weekly digest to keep project leaders current on construction-related executive orders, regulatory changes, and proposed legislation, so a shifting policy landscape doesn't catch anyone off guard.
After reorganization or process changes, McCarthy teams now use Glean to compare new and existing procedural matrices. Glean recommends updates needed across thousands of affected documents — a task that previously would have landed on leadership to complete manually over several weeks.
This agent has automated some processes for verifying trade partner billing against contract terms and has already helped its human counterparts review more than $2.7 million in change orders.
Tracing a failed material across dozens of active projects is painstaking work at scale. Now an agent surfaces those connections in minutes — faster and more completely than any manual process could.
Lewis sees McCarthy’s rollout of AI agents as a competitive moat in the industry. “We’re now actually getting value from AI, while a lot of other teams are still trying to achieve that.”
Internal survey data found that Glean saves each employee 1.5 hours per week on information searching. With 80% of employees regularly using Glean, Zillow teams across HR, Engineering, UX Research, and more are able to prioritize higher value activities in their day to day workflows.
Driving Success with Glean
McCarthy’s agent rollout reflects a deliberate balance between openness and oversight. Any employee can use Glean to experiment and create in what Lewis calls "our playground to build agents," while a smaller, vetted group holds publishing rights for company-wide deployment — structured enough to be responsible, open enough to drive real innovation across the organization.
Unlike solutions tied to a single AI provider, Glean's system of context and LLM flexibility allows McCarthy teams to choose the best underlying model for each use case and quickly adopt new capabilities without rebuilding from scratch.
The adoption numbers reflect something harder to quantify: a workforce that has genuinely embraced AI as part of how it works. The team even created a “Glean Help” Agent that teaches employees Glean on their own terms.
For Lewis, the results speak for themselves. “We feel like we did something really great here.”
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.
