- Calm Agent
- Posts
- 🟢 You, Your Client, and the Zillow Ultimatum: What's Your Move?
🟢 You, Your Client, and the Zillow Ultimatum: What's Your Move?
And, How OpenAI's o3 Can Help Real Estate Agents With Their Property Walkthroughs
Maaret Forst
April 21, 2025
Today's Insights
FROM THE CALM AGENT DESK
Beyond Clear Cooperation: Whose Interests Are Really Being Served?
An Industry Debate Most Clients Never Hear About
The Clear Cooperation Policy just got more complicated. In March, NAR introduced "Multiple Listing Options for Sellers," sparking a fierce debate among brokerages. Meanwhile, Zillow announced that it won't display properties not listed in the MLS within 24 hrs, creating industry-wide ripples.
While most homebuyers and sellers remain unaware of this industry debate, it has significant implications for how listings are marketed and ultimately, how client interests are served.
Zillow's Transparency Revolution
I recently listened to a Tim Ferriss podcast featuring Zillow co-founder Rich Barton (episode #806) that illuminated how Zillow originally transformed real estate. When Zillow launched, the industry kept home prices and addresses largely hidden from consumers. Barton and his team built Zillow around a "power to the people" philosophy, democratizing access to information that had previously been guarded by gatekeepers.
The "Zestimate" feature, providing algorithmic home valuations, was revolutionary. It brought unprecedented transparency to an opaque industry, even as it faced resistance from industry associations uncomfortable with this new openness.
Today's Position Taking
Shortly after Zillow's announcement, eXp announced its alignment with Zillow's position, with CEO Leo Pareja emphasizing, "preservation of cooperation is a win for consumers."
Then Guy Gal from Side Inc. published an open letter with a different perspective: agents have a fiduciary duty to serve clients in their best interest, regardless of whether listings are listed on the MLS or not. The decision belongs to the seller.
It is always hard to know exactly what is right when there are strong opposing views amongst reputable people on contentious issues.
If there is one thing you can count on when it comes to the CCP debate: all of the real estate companies driving it will always do what is best for their bottom line first, and make arguments for why it's best for clients and agents second.
But nobody knows what's best for their clients better than you, the experienced agents who take the time to actually understand their needs and their circumstances.
My unsolicited advice to you is not to succumb to the pressure or temptation of choosing one side or another in this debate or any other. Rather, seek to understand what motivates these companies, what their incentives are, and how they benefit from what they are advocating for.
Explain that to your clients. Enlighten them. Talk it through together. And then do what's right for your clients and them alone.
Worrying about things that are outside your control is a terrible waste of energy and creativity and imagination. The magic you seek is in redirecting that energy to all of the things that you can impact directly.
We live in a time of great distraction. There is an incredible power in ignoring the noise. Don't get sidetracked by the latest hype cycle or shiny object. Focus on that which is most enduring and fundamental. Orient your attention to serving your clients. Everything else will sort itself out.
You have a fiduciary duty to do what is best for your client. Nothing else matters.
Beyond Simple Divisions
This isn't a black-and-white issue with clear teams. Different companies have different needs:
Some need all listings in one place to make their websites valuable
Others want flexibility for their high-end or privacy-focused clients
Many are just trying to figure out what works in their local markets
What works in LA's luxury market might not work in middle America's suburbs. What makes sense for a celebrity seller may not be the same for a typical homeowner.
When companies take strong positions, ask yourself:
Who makes money from this stance?
How does it protect their business?
What threats to their business are they trying to avoid?
Your Judgment Matters More Than Ever
Your Value Proposition
Your value as an agent stems from your ability to assess what your specific client needs in their unique situation, rather than adhering to one-size-fits-all approaches.
Your independence allows you to make decisions based on client needs rather than corporate directives.
Different Clients, Different Needs
This debate highlights a simple observation: different clients need different approaches.
Some sellers want maximum exposure
Others need privacy and targeted marketing
Some listings benefit from strategic timing
Most importantly, clients need approaches tailored to them, not one-size-fits-all rules
The Fiduciary Duty
When you sit across from clients, you have something these corporations will never have: a duty to that specific person. Not to shareholders, not to a business model, not to an industry rule.
The next time you hear a company frame a business policy in strong terms, ask yourself: "Who benefits most from this?" Then make your decision based on your client's unique needs, not someone else's business agenda.
AI UPDATES
OpenAI's New "Thinking" Models: What Agents Need to Know
Meet O3 and O4-Mini: Agentic AI That Works For You
OpenAI released its latest AI models on April 16, 2025, namely O3 and O4-mini. Unlike previous models that simply respond to prompts, these new "reasoning models" are truly agentic—they can decide which tools to use, take initiative, and work through problems independently, much like a trusted assistant.
Why should real estate professionals care? These agentic models can analyze property photos, research neighborhoods, draft detailed market analyses, and solve problems far more accurately than previous AIs - all with minimal direction from you.
The Reasoning Revolution
These thinking models represent a significant shift in AI capabilities. While OpenAI has O3 and O4-mini, several other companies have entered the race with their own reasoning models.
Model | Typical real estate task | Notable capability | Choose when |
|---|---|---|---|
O3 | Detailed market studies, image‑based property reviews | Strong reasoning and image understanding | A project needs careful multi‑step analysis and visuals |
O4 mini | Live Q & A with clients, quick comp checks | Low delay, lower cost | Response time is more important than depth |
Claude 3.7 Sonnet | Client letters, contract summaries | Clear language and an optional longer “thinking” mode | Tone and compliance matter |
Gemini 2.5 Pro | Research that depends on current web data | Built‑in Google Search context | You need the latest public information |
DeepSeek R1 | Private data work, multilingual chat | Runs on modest in‑house hardware | Data must stay on your server or cost is a concern |
Llama 3.1 405B | Custom internal tools, fine‑tuning | Highest open‑source reasoning scores | You want to build and update a model yourself |
What Makes These Models Different?
These aren't just faster versions of the same AI – they represent a fundamental shift in how we use AI:
They operate agentically – making independent decisions about what tools to use and steps to take
They think step-by-step instead of guessing based on patterns
They use tools automatically (web search, image analysis, coding) without being told
They work with images intelligently, zooming in on details and analyzing visual information
They can handle larger amounts of information (like entire market reports or MLS data)
AI PRODUCTIVITY
How can O3 analyze listing photos?
Below is a result of a conversation with o3, which is why it references “I”. That’s o3, by far my favorite model so far, talking.
I asked it to be truthful and fact-check everything.
What I can do reliably | Example outputs | Why it helps an agent |
|---|---|---|
Detect visual features (materials, roof type, siding, window style, lighting, furniture scale) | “Standing‑seam metal roof, cedar shingles on upper gables, composite decking” | Quick material checklist for disclosures and marketing copy |
Spot condition cues (mildew on deck boards, dated fixtures, fresh paint seams) | “Deck stain uneven at stair treads—budget a light sanding and re‑coat” | Prioritize prep dollars before photography or inspection |
Infer the likely room function and scale | “Room appears 12 × 15 ft; staging as a queen bedroom leaves 28 in walkways” | Guides furniture rental orders and virtual‑staging briefs |
Suggest low‑cost, high‑impact tweaks | “Swap chrome knob for matte black lever to modernize the entry shot” | Saves time during walkthroughs—instant punch list |
Flag privacy risks | “Street number visible on mailbox—crop or blur before public post” | Protects client confidentiality and fair‑housing compliance |
Generate copy hooks from visuals | “Sun‑drenched reading nook framed by clerestory windows” | Fresh language without repetitive “light‑filled” clichés |
How does the image reasoning work?
Region focus – I first isolate distinct zones (roof, façade, landscape, furniture).
Attribute matching – Each zone is compared against a library of known textures, shapes, and architectural patterns.
Cross‑check – I weigh multiple cues together: a divided‑lite window, stone veneer, and steep gable often mean Craftsman; if one element conflicts, confidence drops.
Practical mapping – The raw detection converts into agent‑friendly notes: materials list, staging fix, and potential buyer objection.
Privacy guardrail – If a prompt tries to reveal an address or owner identity not already disclosed, I refuse.
What I can’t do (yet)
Guarantee exact square footage from a single image.
Identify the brand or model year of appliances beyond obvious labels.
Confirm structural integrity—surface cracks don’t tell the full story.
Provide a street address unless you’ve stated it in the chat.
Limitations to keep in mind from O3:
I can’t pull live MLS data, quote private sales, or identify exact property locations from new photos, due to privacy issues.
Any legal, tax, or code‑compliance advice must be reviewed by a licensed professional.
Market statistics come from publicly available sources; always verify numbers before publishing.
Calm Agent ChatGPT Playbook for Real Estate Agents is here!
Access it in the Calm Agent AI Suite on the website. Here is a link to the download page:
What did you think of today's email? Your feedback helps us create better emails for you! |
Stay Calm!