43,000 Listings Vanished. Real Estate Has a Trust Problem.

In May, a strange thing happened in Chicago. Tens of thousands of homes appeared to disappear from Zillow, not because they had been sold, withdrawn, or corrected, but because a data feed was switched off.

The public explanation sounded almost absurd: roughly 43,000 MRED listings were caught in a dispute that, according to MRED, centred on nine listings Zillow refused to display. Those nine listings were Compass listings outside MRED's traditional Chicagoland footprint. A local listings database had suddenly become the stage for a national argument about who gets to decide how homes are marketed.

That should make everyone in real estate uncomfortable.

The fight has quickly been framed in familiar language. Zillow says this is about transparency and fair access for consumers. MRED says this is about enforcing its licensing agreements and protecting lawful seller marketing choices. Compass says sellers should be able to decide how their homes are exposed to the market. Each side can make a coherent argument. Each side also has a commercial interest.

That is why I do not think the industry should treat this as a sports match. The more useful question is not whether you are on Team Zillow, Team MRED, or Team Compass. The question is why the system is fragile enough that tens of thousands of consumers can suddenly lose access to listings because the market has no shared way to prove what happened to a listing, who authorised it, and what rules applied to it.

The easy story is not sufficient

It is tempting to reduce this to a clean fight between transparency and seller choice. That story is emotionally satisfying, but it is not sufficient.

Transparency matters. A buyer cannot compete for a home they never see. A seller may not get the full benefit of demand if exposure is limited to a small circle of agents or buyers. Hidden inventory also creates the conditions for unequal access, especially when access depends on which brokerage a buyer happens to work with.

Seller choice matters too. Not every seller wants maximum public exposure on day one. Some sellers have privacy concerns. Some want to test price quietly. Some are managing tenants, family circumstances, or timing constraints. The industry should not pretend that one marketing path fits every seller.

The problem is that both arguments collapse when they are not backed by evidence.

If a brokerage says a seller chose delayed marketing, the market should not have to take that on faith. If a portal says a listing violates its access standards, the broker and seller should be able to understand exactly why. If an MLS says a feed recipient breached its rules, the public consequence should not feel like a black box. At this scale, trust cannot live inside emails, private remarks, PDF forms, or screenshots passed around LinkedIn after the damage is done.

The court's temporary order did not settle the deeper question. It restored Zillow's access to the MRED feed, but it also limited Zillow's ability to exclude certain MRED listings while the dispute continues. Everyone got a piece of the story they wanted. No one got a final answer on the underlying structure.

A listing is no longer just a listing

This is the part of the debate that gets missed.

A residential listing used to look like a marketing asset: a price, an address, a few photos, some agent remarks, and a phone number. Today it is something much more important. It is a market data asset. It carries:

• •Consent and listing authorisation
• •Distribution rights and syndication permissions
• •Exposure history and marketing state (public, delayed, private, coming soon, office exclusive)
• •Pricing history and availability changes
• •Media authenticity (original, enhanced, staged, or AI-generated)
• •Fair housing implications
• •Lead-routing consequences
• •Platform policy risk

In other words, the listing is not only the thing being marketed. It is also the evidence of how the market is being run.

I have spent enough of my career inside real estate data systems to know how much of the industry still runs on assumptions. We assume the listing was properly authorised. We assume the seller understood the trade-off between privacy and exposure. We assume the listing was distributed consistently. We assume the photos represent the property. We assume the broker, MLS, portal, and consumer are all looking at compatible versions of the truth.

Those assumptions are becoming expensive.

The Zillow-MRED-Compass fight is one example. AI-generated listing content is another. Digitally altered images are another. Duplicate and stale listings in portal-led markets are another. They look like different problems, but they share the same root: the industry has spent years optimising the display layer while under-investing in the proof layer.

Visibility is the wrong argument

That sounds technical, but the idea is simple. Every listing should have a lifecycle record.

Who signed the listing agreement? What did the seller authorise? Was the listing public, delayed, private, coming soon, preview, office exclusive, or syndicated? Who could see it at each stage? When did its status change? Which portals received it? Which portals rejected it? Which policy was applied? What changed in the price, photos, description, floor plan, or availability? Was the media original, enhanced, staged, or generated?

These should not be forensic questions asked after a lawsuit begins. They should be normal properties of the listing object itself.

A trust layer would not decide every policy question. It would do something more useful: make the facts visible enough that the policy debate can become honest.

Seller choice needs receipts

The word choice does a lot of work in real estate. Too much, sometimes.

A seller should be allowed to choose a marketing strategy. But if that strategy limits exposure, the seller should know what they are giving up. They should understand that delaying IDX or portal exposure may reduce the pool of buyers who see the home in the first days of the campaign. They should understand how days on market will be treated. They should understand whether the listing will be visible only inside one brokerage, across the MLS, to all agents, or to registered consumers on specific portals.

And the industry should be able to prove that understanding later.

This is not about burying agents in compliance work. It is the opposite. Good infrastructure removes ambiguity. The seller's instruction should be captured once, in a structured way, then travel with the listing. The marketing state should be machine-readable. The permissions should be clear. The audit trail should exist without requiring a broker, portal, or MLS to reconstruct the story under pressure.

That protects consumers, but it also protects good agents. Most agents are not trying to manipulate access. They are trying to serve a client inside a rule environment that has become too complex for manual interpretation. The answer is not more slogans. It is better workflow.

Portals and MLSs both have to earn trust

There is also a trap in pretending one type of player is naturally virtuous.

Portals are not neutral simply because they face consumers. Their traffic, monetisation, ranking logic, display policies, and agent products all influence the market. A portal can defend transparency and still be a powerful gatekeeper.

MLSs are not neutral simply because they were built for cooperation. Their rules, feed agreements, participation models, and enforcement decisions can shape who gets access to market information and under what conditions. As MLSs become more ambitious, more national, and more platform-like, that responsibility only grows.

Brokerages are not neutral simply because they represent sellers. A brokerage can believe sincerely in seller choice and still benefit commercially from keeping inventory inside its own network for longer.

That is not an accusation. It is just reality. Everyone in the chain has incentives. The answer is not to ask consumers to decide which institution they trust most. The answer is to make the chain of custody clear enough that trust does not depend on institutional self-description.

This is bigger than the United States

The American MLS system has its own language and politics. Clear Cooperation, IDX, VOW, office exclusives, delayed marketing, private listing networks. It is easy for people outside the US to see this as a domestic fight.

It is not.

In the UAE, MENA, Europe, and many other markets, the structure is different. Portals often carry more of the consumer trust burden because there is no universal MLS-style cooperation layer. But the underlying problem is the same. Is the listing authorised? Is it current? Is it duplicated? Has the image been altered? Is the price real? Has the property already been rented or sold? Who is allowed to advertise it? Where did the data originate?

Markets without a Zillow or without a US-style MLS still need answers to those questions. In some ways, they need them even more urgently because the consumer-facing portal becomes the de facto source of truth.

That is why I think the future of listing infrastructure will not be defined by one country's rules. It will be defined by provenance. The winning platforms will be the ones that can show where a listing came from, how it changed, and whether the people touching it had the right to do so.

AI makes this unavoidable

AI is going to make this problem faster, not smaller.

Listings can now be written, translated, enhanced, staged, reformatted, distributed, and republished in seconds. That is good when the underlying data is clean. It is dangerous when the underlying permissions, media history, or availability status are unclear. Bad data used to move at human speed. Now it moves at software speed.

The same technology that creates the risk can also create the audit trail. A modern listing workflow should not rely on people remembering to document every step. It should capture the steps automatically. It should know when an image was altered. It should know which version of a description was published. It should know whether the seller authorised delayed exposure. It should know which portals received the feed and which ones rejected it.

This is where the industry should be putting its energy. Not into another round of public statements about who cares more about consumers, but into the infrastructure that makes consumer protection measurable.

The lesson from Chicago

The lesson is not that Zillow is right about everything. It is not that MRED is wrong about everything. It is not that Compass has no legitimate argument for seller flexibility.

The lesson is that the market cannot keep relying on trust without proof.

When 43,000 listings can disappear from a major consumer platform because of a dispute over a handful of records, the industry has a data governance problem. When every side can claim to be defending the consumer, the seller, the agent, and the market at the same time, the industry has a language problem. When no one can easily explain the full lifecycle of a listing from authorisation to display, the industry has an infrastructure problem.

The next real estate trust layer will not be another portal, another private network, or another policy memo. It will be the system that answers those five questions without drama.

The market does not need another argument about who owns visibility. It needs proof of the listing's life.