Inman AI Survey: Which Office Jobs Go Before the Next Hire

On October 19, 2026, Inman published its latest AI survey, and one number stands out for anyone who runs a brokerage. According to Inman's write-up of the survey, 17% of brokerage leaders said AI let them reduce headcount or avoid adding administrative or transaction coordination staff. A transaction coordinator, or TC, is the person who keeps a deal's paperwork and deadlines on track from accepted offer to closing.

Here's why that matters to an independent broker. On the agent side of the same survey, 62% of agents pay for a general AI assistant, yet only 4% say they get more appointments because of AI. Your agents may already be paying for AI while the office still does the same paperwork by hand. So the useful question isn't whether to buy AI. It's which office jobs, such as admin or transaction coordination, AI should carry before you make your next hire.

Who answered, and who paid for it

Before leaning on any of these numbers, it helps to know where they come from. The survey is the Inman Intel Index, with 460 responses collected from September 16 to October 2, 2026. Lofty, a vendor, sponsored it. Inman says its journalists wrote the report without sponsor review.

The people who answered chose to take part, and they come from Inman's engaged reader community. That's a useful group to hear from, but it isn't the whole industry, so we read these figures as a signal from engaged brokers and agents rather than as industry-wide fact.

The survey also mixes two groups, agents and brokerage leaders. We'll tag every figure with the group it describes, since an agent's answer and a leader's answer aren't the same measurement.

Better copy, few new appointments

Start with agents. The 62% who pay for a general AI assistant are paying for a chat tool that handles many kinds of tasks, such as ChatGPT, Claude, or Gemini. Among agents, 52% say AI made their content better, while 33% say they spend less time on admin.

Then there's the 4% of agents who say they get more appointments because of AI. Inman's survey doesn't say why that number is so much lower than the others, and we won't invent a reason.

Credit where it's due, too. Better content is a real gain, and a third of agents spending less time on admin is worth something to them. These tools are doing work.

Our read, and it's ours rather than the survey's, is that the gap is in how the work is set up. A general assistant helps the person typing into it. It doesn't, on its own, change how a file moves through your office or who handles the next step.

The office side of the survey

Now the leaders. The 17% of brokerage leaders who said AI let them reduce headcount or avoid adding administrative or TC staff is the figure an owner should study. It's a minority of leaders, and the survey wording covers either outcome, so it doesn't say how many did one or the other.

What it does suggest is that some brokerages are already using AI on office work, not only on agent marketing. For an independent broker deciding whether to post a job for an admin or a TC, that's worth knowing before the ad goes up.

It also reframes the agent numbers. If AI is mostly helping agents write, the office hasn't changed yet. The jobs that might let you hold off on a hire live in the office, in the paperwork and the deadlines.

Owners and agents remember adoption differently

One of the more telling gaps in the survey is about tools nobody learned. Among brokerage leaders, 45% said agents signed up for an AI tool and never learned to use it. Among agents, only 15% said that happened to them.

Those are two different groups answering about the same kind of event, so we won't treat the difference as a measurement of anything. Still, the distance between them should make any owner pause before buying the next tool. If leaders and agents see adoption that differently, it's worth checking what's actually being used before adding more.

What leaders who shop for AI look for

The survey also asked leaders how they choose outside AI tools. About 15% of leaders said their firm doesn't shop for outside AI tools. Among leaders at brokerages that do, 38% said connecting to systems they already use was the single deciding factor, 30% pointed to adoption without heavy training, and only 7% said price.

Each of the three systems named does a different job. A CRM, short for customer relationship management, is the system that tracks clients and leads. Transaction management software holds a deal's documents, tasks, and deadlines in one place. The MLS, or multiple listing service, is the shared database brokers use to list and find properties.

That 38% points to a practical place to start. An AI that can't work with your CRM or your transaction management software leaves someone copying information between them, which is exactly the office work you were hoping to reduce.

Handing transaction coordination to an AI employee

Here's how we'd approach it. We'd give one recurring office job to an AI employee. As we use the term, an AI employee is one office duty assigned to AI, built on your brokerage's own checklist and working with the systems where that job already lives, with a named staff member who reviews its output.

Transaction coordination is a strong candidate, because many of its steps repeat from deal to deal. Tracking contract deadlines, requesting missing documents, and keeping the transaction file complete follow the same steps each time. We'd start with one of those tasks rather than all of them.

Write down how your best TC or office manager does it now, step by step. Then decide which steps AI drafts and which ones a person approves before anything reaches a client, a lender, or another agent. That review matters. In the same survey, 41% of agents said they caught a factual error in AI-generated material before it reached a client.

Here's the short version we'd give any independent broker. Before your next admin or TC hire, pick one repeating office task, connect AI to the systems where that task lives, and keep a person reviewing its output. If that task runs well for a month, you'll know whether the hire can wait.

To know whether it ran well, count something before you start. That might be how many files missed a contract deadline last month, or how often someone had to chase the same document twice. Check the same count a month later, with the person who reviews the AI's work keeping the tally.

We've covered similar starting points before. One post looked at free AI agents that each handle one document chore, while another covered AI coworkers sold with a named role and permissions. The same idea carries over to a brokerage office.

Common broker questions about AI in the office

What is the Inman AI survey?

It's the Inman Intel Index, a survey of 460 self-selected Inman readers drawn from agents plus brokerage leaders. Responses were collected September 16 to October 2, 2026, with results published October 19, 2026. Lofty sponsored it, and Inman says its journalists wrote the report without sponsor review.

How to decide which office job AI should take first?

Pick one repeating task, such as tracking contract deadlines or requesting missing documents, and write down how your best TC does it today. Choose AI that works with the CRM or transaction management software you already run, and keep a person reviewing its output.

How much office staffing has AI affected at brokerages?

In the Inman survey, 17% of brokerage leaders said AI let them reduce headcount or avoid adding administrative or transaction coordination staff. That's a self-selected sample of engaged Inman readers, so it's a signal rather than an industry-wide figure.

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