Don’t Mistake AI for Automation: A Primer for Property Management

At October’s Rent Manager User Conference, a speaker asked a room of property management professionals a simple question: 

“How many of you track the late fees you’re paying on utility bills?” 

In a room packed with dozens and dozens of manufactured housing, multifamily, and single family property managers and owners, only a handful of hands went up.

It was a conference (like so many conferences this year) that was buzzing about the potential of artificial intelligence. But that moment? When so many hands remained solemnly unraised? It underscored a key takeaway from the conference as a whole:

Many of our industry’s most impactful opportunities are hiding in processes that need to be enhanced and supported—not replaced.

Before asking what AI can do for a portfolio, operators might benefit from a more immediate question: What should already be happening automatically, repeatably, and reliably?

Efficiency Has Become a Financial Imperative

The economics of rental housing have changed. During one of this year’s Rent Manager User Conference (RMUC), manufactured housing panels, one operator described how a higher cost of capital has made operational performance far less forgiving. When financing was cheap, they argued, inefficiency had more room to hide. Today, every avoidable expense places additional pressure on returns.

That concern is reflected in Deloitte’s 2027 Commercial Real Estate Outlook. Its survey of 950 real estate executives ranked capital costs, capital availability, and interest rates among their top concerns. The outlook emphasizes disciplined investment in an environment with less room for indiscriminate spending.

The math makes the stakes visceral. At an illustrative 5% capitalization rate, eliminating $1,000 in recurring monthly expenses could represent approximately $240,000 in property value, assuming those savings increase net operating income and the cap rate holds. Overlooked fees, redundant labor, and uncorrected billing errors can have consequences well beyond their individual line items.

Automation and AI Solve Different Problems

Across several sessions at RMUC, a broader theme emerged beneath the excitement surrounding AI: improving operations starts with understanding the work itself.

1. Some workflows call for intelligence and context 
2. Some workflows need a dependable, repeatable process.

Automation is valuable precisely because it’s predictable. Given a defined rule and reliable input, a well-designed workflow performs the same action consistently. It routes invoices, enforces approvals, tracks deadlines, and flags exceptions without requiring someone to remember the next step.

AI, on the other hand, adds another layer. It can identify patterns across complicated datasets, interpret unstructured information, and help operators investigate questions they might never have known to ask. But AI can also produce inconsistent or inaccurate results, which means its responsibilities and safeguards need to be deliberately designed.

It’s not an absolute distinction, of course. AI can absolutely be incorporated into automated workflows, and the two technologies are often more valuable together. 

What matters is understanding which tasks require consistency, which benefit from intelligent analysis, and which demand human judgment.

For property managers contending with the market forces discussed throughout RMUC, that’s a deeply practical distinction. A payment deadline shouldn’t depend on an AI model making a fresh decision every month. But an AI-powered analysis of years of payment and consumption data might uncover an emerging problem that conventional rules have missed.

McKinsey’s 2026 State of AI survey found that nearly nine in ten respondents reported regular AI use in at least one business function, yet only 37% reported positive enterprise-level EBIT impact. Nearly three-quarters of AI high performers reported fundamentally redesigning workflows, compared with just one-quarter of other respondents.

Those findings don’t establish that workflow redesign alone produces better financial results. But they DO reinforce an important idea we saw through the conference: there’s a difference between deploying technology and changing how work gets done.

Introducing AI into an inefficient process doesn’t automatically make the process better. It may simply make the inefficiency more technologically sophisticated.

Make Room for the Work That Matters

Again and again, across the five sessions we attended, speakers, panelists, and even audience discussion continued to circle the idea that emerging technologies’ chief priority should be giving property teams more time to do the work that requires people: solving unusual problems, exercising judgment, helping residents, and building relationships with owners.

Which tasks should happen the same way every time? Where would smarter analysis genuinely change a decision? And where does a human need to remain accountable? 

We’re not arguing that property owners and managers have to choose between automation and AI. We’re insisting, as pioneers in automation and AI-supported workflows ourselves, that they should understand their goals well enough to know where each can make a difference.

AI will undoubtedly continue to expand what’s possible in property management. But operators don’t need to wait for the next breakthrough to improve the fundamentals. There’s tremendous value in making the routine reliable, the exceptions visible, and the data useful.

The next time someone asks a room of property managers whether they’re tracking utility late fees, knowing the answer will be progress.

Having fewer late fees to count will be the payoff.

Ready to put these ideas to work in your utility operations? Connect with a Utilty Expert to learn how our AI-powered technology and dedicated utility expertise will help your team stay ahead of the daily workload.