ServiceTitan Max Is Open. Which Job Turns Hours Into Profit?

On October 6, 2026, at its Pantheon conference, ServiceTitan made Max, the AI package it calls an agentic operating system, available to every residential in-home contractor, according to the company's Pantheon 2026 announcement. Agentic is industry shorthand for AI that's built to carry out tasks rather than only answer questions. Commercial and roofing contractors got a limited Max pilot instead, so the full opening is for residential in-home shops like HVAC, plumbing, and electrical.

That's a real step, and ServiceTitan deserves credit for putting AI in front of every residential shop instead of a select few. The honest catch sits in ServiceTitan's own research: 66% of contractors using AI save at least three hours a week, yet only 47% of them report improved profitability. If you run a residential shop, getting access to AI isn't the hard part anymore, and neither is finding hours. The open question is which job you hand the AI first so those hours actually show up in profit.

Max reaches every residential shop

What makes Max matter is who it comes from. It's ServiceTitan's own AI package, so a residential shop already running ServiceTitan doesn't have to go find a separate AI vendor just to get started.

That placement lines up with what contractors say they want. In ServiceTitan's survey, 61% of AI users said they prefer AI built directly into their core platform. On the question of where the AI should sit, ServiceTitan is building toward what contractors in its own research asked for, and the release adds a mobile app so owners can ask from their phone why a number moved.

What the release doesn't give you is a number for what Max does to revenue or profit. It talks about those outcomes without attaching a figure, so we won't put one on it either. Where Max earns its keep in your shop is still your call.

Three hours a week that never reach the P&L

The survey behind those figures is ServiceTitan's 2026 State of AI in the Trades. ServiceTitan commissioned it, Thrive Analytics ran it, and it surveyed 1,017 U.S. contractors between May 26 and June 15, 2026, with a margin of error of plus or minus 3.7 points. ServiceTitan released the findings on September 29, 2026, a week before Max opened up, so none of these numbers measure Max.

Among contractors already using AI, 66% save at least three hours a week. That part is working. But only 47% of AI users report improved profitability, and 43% report increased revenue. Those are separate answers to separate questions, so we won't stack them or turn the hours into a dollar figure. Read side by side, though, more AI users report getting time back than report seeing better profit.

Picture a residential shop where AI trims the time it takes to write up job notes. Nobody decided where those freed hours should go, so they dissolve into the week. Estimates still sit unsent, the owner still reviews invoices at the kitchen table, and the saved time never turns up anywhere the accountant can see it.

The survey doesn't say why the gap exists, and we won't pretend it does. Our read is that hours turn into profit when they come out of work tied to money, like an estimate that gets followed up or an invoice that goes out right the first time. Saving time on work that sits far from the money gives you a calmer week, which is worth having, but it won't move the P&L.

Price stopped being the wall

If you assumed price was what kept contractors from using AI, this survey says otherwise. High purchase cost now ranks last among the barriers at 17%, down 12 points from the prior wave, though the question wording was lightly revised between waves, so the drop is a direction more than a precise measure.

What sits at the top is a knowledge problem. In the survey, 44% of respondents name understanding how to use AI in their business as a barrier, 43% point to a lack of clear use cases, and 40% cite integration with the software they already run. A use case is one specific job the AI does for you, start to finish, with a result you can check.

On what would help, 40% say clearer ROI would drive adoption, meaning a plain view of what the AI returns for what it costs, and 38% want AI trained on trade data. The hard question in the trades isn't whether AI is affordable anymore. It's where AI fits in a given shop and what it pays back once it's there.

Why trust sets the pace

The survey also asked why contractors try AI in the first place. The top reason is hiring difficulty and the workforce gap, at 37% of respondents, followed by off-hours support at 34% and lower operating costs at 33%. That top answer tells you what's really being asked for: more capacity from the techs and office staff you already have, because adding people isn't happening fast enough.

Trust is where things slow down. Only 11% of respondents place high or complete trust in their AI tools, and 58% report little or no trust. Among AI users, 35% name accuracy and reliability as their leading concern. That's a sensible instinct, not a fear of technology. An AI that gets an estimate or an invoice wrong in front of a homeowner puts your name on the mistake, and in a residential shop the owner's name is the brand.

So the hiring gap and the trust gap point to the same starting move. Give the AI one job, close to the money, where a person on your team still signs off before anything reaches a customer.

Start with one job and a person who signs off

An AI employee, the way we use the term, is software that owns one recurring task in your shop, does it the way your best office person would, and answers to someone on your team. Its stop rule is the moment it has to pause and wait for that person's okay, for example before an estimate or an invoice reaches a homeowner.

Here's how we'd set it up, and you don't have to become technical. First, name the job. Pick one task where saved time turns into money, such as following up on estimates that went out and never got an answer, or checking each invoice against the finished work before it's sent. One job, not five.

Second, point it at the truth. The AI should work from your price book, your past jobs, and the way your office already writes to customers. When respondents ask for AI trained on trade data, the closest trade data you have is your own.

Third, put the stop rule on the send. The AI drafts, checks, and flags, and a person approves anything a customer will see. That sign-off is what lets your office actually use the thing instead of quietly working around it.

If you're on ServiceTitan, Max may turn out to be the tool that runs that job. ServiceTitan's Pantheon release doesn't put a number on Max's results, so we can't tell you how it will perform in your shop. What we can tell you is that the choice of which job comes first has to be made before any tool can pay it back.

Turning saved hours into jobs and margin

Here's the short answer we'd give any residential owner asking where AI pays. Saved hours turn into profit only when they come out of a job tied to money, like an estimate or an invoice. Pick that one job first, keep a person signing off, and measure it before you add a second.

Measuring is the part that answers the ROI question the survey keeps circling. Before the AI touches the job, write down how it runs today: how many estimates go unanswered, or how often an invoice goes back for a fix. A month later, look at the same thing. If more estimates turn into jobs or fewer invoices need a second pass, the hours went somewhere useful.

That also answers the fear of buying an AI package and ending up with a dashboard nobody opens. A named job with a person who signs off gives the crew something concrete to trust, and it gives you something to point to when someone asks whether the AI was worth it.

We've written before about a ServiceTitan feature that drafts marketing campaigns, in our August post on reaching the right audience, and about how a powerful tool still needs a place to start. The pattern holds here. ServiceTitan did its part by opening Max to every residential shop. Deciding which job goes first is the part only you can do.

FAQ: ServiceTitan Max and your first AI job

What is ServiceTitan Max?

ServiceTitan Max is the AI package ServiceTitan calls an agentic operating system, and on October 6, 2026 it became available to every residential in-home contractor. Commercial and roofing contractors got a limited pilot instead.

How do you turn hours saved by AI into profit?

Pick one job tied to money, like following up on estimates or checking invoices, and keep a person signing off before anything reaches a customer. Measure that one job against how it ran before, then decide on the next one.

How much does AI pay off for contractors?

In ServiceTitan's commissioned survey, 66% of contractors using AI save at least three hours a week, but only 47% of AI users report improved profitability. ServiceTitan's Pantheon release doesn't put a number on Max's results, so the payoff in your shop depends on which job you point it at first.

If you want the long version of how we think about readiness, Start with a FREE quiz and Get Your AI Score.