HomeSmart’s AI Message Is Simple: Agents Still Need People

HomeSmart’s chief operating officer, Katie Cooper, used a recent RISMedia op-ed to make a familiar argument: artificial intelligence should help real estate professionals do better work, not eliminate the human relationships behind the business.

The notable part is not the position itself. Most brokerages now say some version of it. The useful question for agents is what happens after the statement.

Should you buy another AI tool? Change brokerages? Rebuild your workflow? Or simply keep working as usual?

For most agents, the answer is more practical: measure where technology is wasting your time, then decide whether human support or automation is the missing piece.

What HomeSmart is actually saying

Cooper’s argument is that technology is valuable when it removes administrative friction and gives agents more time for advising clients, building relationships and generating business.

HomeSmart points to several parts of its operating model:

• Dedicated support for onboarding, transactions and closings.
• Broker access by phone, email and in person.
• Professional development through its KickStart program and Ninja Selling partnership.
• Support for franchise owners through a franchise services team and success coach.
• A company-reported figure that more than 94% of incoming support calls are answered without reaching voicemail.

HomeSmart says more than 94% of incoming support calls are answered without reaching voicemail; the op-ed does not explain the methodology used to calculate that figure.

The broader development is less about a new product launch than about positioning. As AI becomes part of CRM systems, transaction platforms and marketing software, brokerages are competing on the layer around the technology: training, implementation, troubleshooting and accountability.

That matters because many agents do not need more software. They need someone to help them use the software they already pay for.

The cost question: support can be cheaper than another app

AI tools often look inexpensive in isolation. ChatGPT Plus is listed at $20 per month. ChatGPT Business standard seats are listed at $25 per user per month on monthly billing, or $20 per user per month when billed annually.

A real estate CRM can cost substantially more. Follow Up Boss lists its Grow plan at $69 per user per month on monthly billing, with higher pricing for teams and additional features.

Transaction-management platforms, lead-routing systems and brokerage technology fees may add another monthly charge or be bundled into a brokerage’s agent package. HomeSmart’s op-ed does not disclose agent fees, franchise costs or the price of its support programs.

That makes the financial comparison less straightforward:

| Option | Typical cost to investigate | Best use | Main risk | |---|---:|---|---| | General AI assistant | $20/month for ChatGPT Plus | Drafting, summarizing and brainstorming | Confidentiality and inaccurate output | | Real estate CRM | $69/user/month for Follow Up Boss Grow on monthly billing | Follow-up, pipeline management and automation | Paying for features nobody adopts | | Transaction platform | Pricing varies | Documents, deadlines and coordination | Workflow disruption during migration | | Brokerage-provided support | Not disclosed by HomeSmart | Onboarding, transactions and business guidance | Support quality may vary by market |

The right comparison is not “AI versus people.” It is the cost of an hour of an agent’s time versus the cost of removing a repetitive task—and the cost of fixing an error when automation fails.

What this means for your business

If you are spending evenings writing routine follow-ups, organizing notes or searching for transaction information, an AI-enabled workflow may be worth testing.

Start with low-risk tasks:

• Turn meeting notes into a follow-up checklist.
• Draft a first version of a client email.
• Summarize inspection or disclosure documents for your own review.
• Create social-post variations from information you have already verified.
• Identify missing fields or upcoming deadlines in a transaction file.

Do not allow an AI system to make unreviewed claims about a property, interpret legal obligations or send client communications automatically. Real estate errors are expensive, and “the software wrote it” is not a defense.

The support question is just as important. If your brokerage offers responsive transaction help, broker access and structured training, that may be more valuable than a marginally better AI feature. A tool can automate a checklist. It cannot tell a new agent which issue deserves escalation, how to handle a nervous seller or when a clean-looking transaction is actually drifting off course.

Who should pay attention

This development matters most to three groups.

First, newer agents should evaluate support before being dazzled by technology. Ask who answers transaction questions, how quickly they respond, whether training is live or recorded, and what happens after business hours.

Second, teams should audit duplicated software. A team may be paying for a CRM, an AI writing tool, a transaction platform and separate communication software while still relying on spreadsheets. Consolidating workflows can matter more than adding another artificial-intelligence feature.

Third, brokerage owners should treat support as an operating investment. Adoption is unlikely to improve simply because a company purchases an AI platform. Agents need clear rules, examples, training and a person who can resolve problems.

Who can mostly ignore it

Established solo agents with a reliable referral business, disciplined follow-up system and strong broker access do not need to make a major change because of this op-ed.

If your current tools are inexpensive, your processes are working and your clients are receiving timely advice, there is no business case for adding technology merely to appear current.

Likewise, agents who rarely use their existing CRM should not add an AI layer yet. First determine why the current system is being ignored. The problem may be poor setup, unnecessary complexity or a workflow that does not match how the agent actually sells.

Tools worth testing—and their limits

For agents who want to experiment, a general AI assistant can be a low-cost starting point. ChatGPT

Skip this if you routinely paste confidential client information into software without understanding its data controls, or if you are unwilling to fact-check every client-facing result.

A dedicated CRM such as Follow Up Boss may be a better investment for agents whose main problem is inconsistent follow-up rather than writing. Follow Up Boss

Skip this if you do not have a defined lead-response process or will not maintain contact records. Automation cannot repair an empty or inaccurate database.

The practical lesson from HomeSmart’s argument is not that every agent needs more technology. It is that technology only creates value when it is connected to competent people, clear processes and judgment.

For agents, the next move is simple: identify one repetitive task, price the available solutions, and test whether the time saved produces more conversations with clients. If it does not, the problem may not be a missing tool. It may be a missing workflow—or missing support.