All In on AI

At ServiceTitan’s Pantheon 2026 conference, co-founder and CEO Ara Mahdessian outlined the company’s vision for AI-powered contracting businesses that can run and improve themselves. Building on a framework introduced at last year’s Pantheon, Mahdessian described five levels of business automation, culminating in a “self-running, self-optimizing” operation.

Mahdessian says contractors have long understood the importance of generating leads, improving booking rates, and increasing average tickets, but time and human capacity have limited their ability to optimize every process. “The only way to be great at everything is to automate it,” he says. Mahdessian emphasizes that individual AI agents are not enough; they must work together to optimize the entire business rather than competing to improve isolated metrics. For example, an advertising agent might continue generating leads when the schedule is already full, or a booking agent might schedule a high-value job without alerting dispatch to assign the right technician.

ServiceTitan developed a coordination Agentic Operating System, Max, that gives agents shared context, common decision-making capabilities, and the ability to coordinate actions while weighing competing priorities. The system also connects business outcomes to the decisions that produced them, allowing agents to refine their assumptions and improve over time. Mahdessian described this as the fifth level of automation: “A system that gets better over time, not just a business that runs, a business that counts out.”

Mahdessian says even modest improvements across key business metrics can produce substantial gains in profitability because revenue can grow faster than costs. In an illustrative example, increasing leads, booking rates, and average ticket by 10% would raise revenue from $4.8 million to approximately $6.4 million. Under his assumptions, profit would double from $480,000 to $960,000 as the profit margin increased from 10% to 15%. He also argues that contractors are well positioned to benefit from the AI transition because their businesses already operate across interconnected workflows on a single platform. “For the first time, the trades don’t have to catch up. We have the foundation.”

Trust Remains a Hurdle

AI experts have urged caution around the technology’s rapid development, emphasizing the need to consider reliability, human oversight, and the consequences of deploying systems that can make decisions and take action. That concern is particularly relevant for contractors weighing whether AI can improve profitability without introducing new operational risks.

Per the aforementioned State of the AI Trades report, 58% of contractors reported little or no trust in AI technology.

“The current tension lies in distinguishing between the use of AI and the use of unsupervised AI, as well as in the interactions among different AI solutions,” Payen says. “Ultimately, we need to ensure our outcomes, general safety, and security remain in control.”

While AI task automation is becoming standard, contractors need strong oversight through customization, visibility, and controls. Payen emphasizes the importance of data security, warning that third-party applications with broad API permissions may access sensitive business information, including customer records, payment histories, membership data, and payroll. He says ServiceTitan has built privacy policies, security measures, and technical safeguards into Titan Intelligence and Max to protect contractor and customer data.

Payen explains that contractors should introduce AI with human oversight and increase its autonomy only after validating its performance. Before fully deploying a virtual customer service agent, for example, contractors should spend several weeks monitoring conversations to ensure the technology behaves as expected. Tasks such as marketing emails, speed-to-lead responses, and job bookings are well suited to automation, he says, but initial human supervision can help establish a foundation for long-term success.

Payen emphasizes that AI should support contractors’ judgment rather than replace it. ServiceTitan focuses on developing transparent AI tools with controls that allow contractors to establish guardrails, evaluate performance, and intervene when necessary. While AI mistakes can have financial consequences or damage customer relationships, Payen says the goal is to remove repetitive work so contractors can spend more time in the field and building customer relationships. He also stresses that automation does not eliminate accountability. “Business owners and their teams remain responsible for the decisions made within their operations, even when AI is involved,” he says. Contractors need to understand what their AI systems are doing, where human oversight is necessary, and how the technology supports their business goals.

Benbrook notes that building trust in AI-generated insights starts with reliable data, requiring contractors to establish sound workflows, configure systems correctly, and ensure their teams use software consistently. However, psychological barriers also play a significant role, including concerns about losing control over business decisions or replacing employees. “The fear, in large part, stems from those two areas: from trusting the data that it’s feeding from, and from an awareness of how to use the technology in an intelligent way,” she says. Contractors should view AI as a tool to enhance employees’ capabilities, whether helping owners make better business decisions or enabling CSRs to improve their calls, rather than as a replacement for human expertise.

Benbrook advises contractors to take a deliberate, problem-first approach to AI adoption rather than investing in technology simply because it is available. “First, understand the problem that you're trying to solve. Be very specific with that,” Benbrook says. “Then you document, train, and you continuously review. Is the tool giving you what you need?”

Payen again notes the importance of ensuring AI tools work together rather than operating independently. “Think about a sports team. Even with the most talented players, if they can't see or communicate with one another, the team will not perform,” he says. “The same is true for AI agents operating in isolation. When agents share the same context and understand what's happening across the business, they can coordinate much more effectively. Contractors who invest in that foundation now will be better positioned to take advantage of what AI makes possible in the years ahead.”