This website uses cookies

Read our Privacy policy and Terms of use for more information.

SCALE WITH LIBERTAS & EXP REALTY

By Tim & Julie Harris

There's a principle we come back to again and again, because it has saved more real estate careers than any script, lead source, or marketing gimmick ever has: never build a castle on land you don't own.

Most agents nod along when they hear it. Then they go right back to building their entire reputation, their lead flow, and their digital identity on a platform they don't control and never will.

Today we want to talk about the most overlooked example of that mistake in our entire industry — the Zillow agent directory — and why it matters more right now than it ever has.

Everybody knows about the leads. Almost nobody understands the directory.

By now, most agents understand the obvious part of Zillow's business. Premier Agent is an advertising auction. You buy a "share of voice" in a ZIP code, and when a consumer clicks "Contact Agent" on a listing, they're routed to whoever paid for that ZIP — not the listing agent, and not necessarily the best agent. That's widely known inside the industry. It's been written about for years.

Here's what is not widely known, and what your clients almost certainly have no idea about: the agent directory itself — the "Find an Agent" search results consumers use to decide who to trust — is not an objective ranking of the best agents in a market.

Consumers believe it is. They type in a neighborhood, they see a list of agents with star ratings and sales numbers, and they assume the person at the top is the top producer. They assume it's earned. They assume it's neutral.

It isn't.

How the directory actually ranks agents

To Zillow's credit, the directory isn't a pure dollar auction the way the contact button is. According to Zillow's own published methodology, the directory surfaces agents based on star rating, number of reviews, and self-reported sales data pulled from the MLS profiles agents attach to their accounts.

Read that again, because the implications are everything.

The ranking is driven by how aggressively an agent works the Zillow platform — how many reviews they've collected on Zillow specifically, how completely they've optimized their profile, how diligently they've fed every MLS ID and every sale into the system. It is not driven by who actually sold the most homes in that neighborhood. It is not driven by verified, independent production data. It is not driven by who would genuinely serve that buyer or seller best.

So picture two agents:

Agent A is a quiet powerhouse. Top of the market in real-world production. Built a referral-and-repeat business so strong they never needed to chase online leads. They never bothered to claim and obsessively optimize a Zillow profile, never built a review-harvesting machine on Zillow's platform.

Agent B closed a fraction of Agent A's business — but they treat Zillow like a part-time job. Every closing turns into a Zillow review request. The profile is immaculate. The engagement is relentless.

In the directory consumers trust, Agent B sits on top and Agent A is nowhere to be found. The consumer walks away believing they just found the best agent in the neighborhood. They found the most active Zillow user.

That's not a ranking of excellence. That's a ranking of platform loyalty.

And now AI is laundering it

Here's the part that should get your full attention, because it's brand new and it's accelerating.

When a consumer asks an AI assistant — or even just Google — "who's the top listing agent in this neighborhood," the answer sounds authoritative. It sounds researched. It sounds like a machine went out, analyzed real sales data, and returned an objective verdict.

It did no such thing.

The AI is not drilling down into MLS production records to determine who actually sells the most homes. It is not auditing closings or verifying performance. It is scraping the sources that dominate search results and are easiest to read — and those sources are Zillow and Homes.com, the same pay-to-play platforms whose rankings were never objective in the first place.

So the bias doesn't get corrected by AI. It gets laundered by it. Advertising-and-engagement rankings go in one end, and out the other end comes a clean, confident, "objective" answer that consumers trust even more than the website it came from. The machine didn't do the homework. It just repeated the homework someone else paid to rig.

This is the single most important thing for you to understand about how buyers and sellers will choose agents over the next five years.

This is exactly why you never build on rented land

So let's connect it back to the principle.

Your Zillow profile is not yours. Your Zillow reviews are not yours. Your ranking in that directory is not yours. Every bit of it lives on land Zillow owns, governed by rules Zillow writes, that Zillow can change tomorrow morning without asking you. They've changed the model before. They'll change it again. And the day they decide your visibility is worth less, or that a new paid tier is required to keep it, you have no recourse — because you never owned the ground you built on.

The agent who built a business on Zillow's directory has a castle. It's a beautiful castle. And it sits entirely on someone else's land.

And no — it's not just Zillow

Before anyone dismisses this as Zillow-bashing, understand that the entire portal and agent-matching ecosystem runs on the same logic. The bias just changes flavor.

Realtor.com — the official NAR-affiliated site — runs nearly the identical model: it sells leads by ZIP code through its Connections Plus program and surfaces a directory built on reviews and self-reported sales. Homes.com sorts paying members above everyone else in search and on neighborhood pages.

Even the platforms that look more merit-based still gate behind a fee. HomeLight is invitation-only and actually matches using real MLS production data — closed volume, days on market, list-to-sale ratio — which is closer to what consumers think they're getting. But agents still pay HomeLight roughly a third of their commission at closing, so a genuine top producer who refuses to join simply isn't in the pool. Redfin's partner program is similar: vetted agents, but a 30–35% referral fee. And services like UpNest tilt the other way entirely — they reward the agent willing to cut their commission the most, which surfaces the cheapest agent, not the best.

Here's the punchline, and it's the whole point: there is no major consumer-facing platform that ranks the genuinely best agent regardless of whether they pay. Not one. So when an AI reaches out to "research" agent referrals, every single source it can touch is monetized in one way or another. There is no neutral library for it to pull from. The bias isn't a Zillow bug — it's the entire substrate the AI is drinking from.

Which is exactly why the only winning move is to control your own.

What to actually do about it

This is not a "complain about Zillow" article. It's a "go take the ground back" article. There are two jobs here, and you do both.

Job 1: Become the answer the AI gives — on your own terms

Since the AI is pulling from whatever it can read across the open web, your job is to make sure the verifiable, real, you-controlled version of your track record is everywhere it looks. This is the new SEO, and most agents are doing none of it. Do these in order:

  1. Claim and fully build out your Google Business Profile. This is the single most-referenced source that search and AI pull from for local professionals. Set the correct category (Real Estate Agent), define your service areas, add real photos, post weekly, and personally answer the Q&A section. An abandoned profile is a missed referral every single day.

  2. Own a real website on a domain you control — not just a brokerage subpage that disappears when you change brokerages. On it, put your actual production in plain, readable text: homes sold, neighborhoods served, price ranges, years in business, designations. AI reads text. It does not read your vibe.

  3. Add structured data (schema markup) to that site — RealEstateAgent, LocalBusiness, and Review schema. This is the machine-readable layer that tells AI exactly who you are, where you work, and what your numbers are. Have your web person implement it; it's a one-time job with a long tail.

  4. Build one authority page per farm area. A dedicated, factual page for each neighborhood you serve that answers the literal questions buyers and AI ask: who's active here, recent solds, price trends, what makes the area unique. Specific and updated beats broad and stale — AI rewards the page that directly answers the question.

  5. Get your name onto multiple independent, credible sites. AI triangulates: it trusts what's corroborated across many sources, not one. Pursue local press features, podcast guest spots, guest articles, legitimate production rankings (RealTrends and the like), your association profile, and reputable "best of" lists. Every credible mention tying your name to real results raises the odds the AI names you.

  1. Make your name, bio, credentials, and numbers identical everywhere. Google, LinkedIn, your site, your brokerage page — same spelling, same stats, same story. Inconsistency makes AI uncertain, and uncertain means it leaves you out.

  2. Diversify your reviews off Zillow. Google first, always. Build a real review-collection habit so your reputation lives on ground you control, not inside one advertising company's walled garden.

  3. Publish content AI can quote. A simple blog and YouTube videos with transcripts/captions (AI reads transcripts). Answer the ten questions every buyer and seller asks, in plain text, on pages you own.

  4. Fully build your LinkedIn. It's heavily indexed and very machine-readable. Complete profile, production highlights, real recommendations.

  1. Test it monthly. Ask ChatGPT, Perplexity, Gemini, and Google's AI: "Who's the best listing agent in [your area]?" See who it names and — more important — which sources it cites. Then go fill the gaps it's pulling from. What gets measured gets fixed.

Job 2: Own the ground underneath all of it

The tactics above help today. The principle protects you forever. The portals and the AI layer can change the rules overnight; these assets can't be taken:

  • Your database / CRM — owned, backed up, and worked with consistent touches. Relationship capital no platform can throttle.

  • Your email list — you own the list outright; no algorithm sits between you and your people.

  • Your brand and your domain — the home base everything else points back to.

  • Your referral engine — word of mouth is the one distribution channel no algorithm gates and no competitor can outbid you for.

The agents who do this don't have to wonder whether Zillow will surface them, whether the algorithm changed overnight, or whether an AI scraped the right page. Their business doesn't live or die on land they're only renting.

The directory will keep telling consumers a story. AI will keep repeating it with even more confidence — because the AI is using those very pay-to-play platforms as its source material. The agents who understood, early, that you never build a castle on land you don't own will be the ones still standing when the ground shifts.

Build on land you own. Always have. Always will.

— Tim & Julie Harris

Founders of Tim & Julie Harris Real Estate Coaching | Publishers of Harris Real Estate Daily | Hosts of PowerHouseTalk | eXp Realty Sponsors at Libertas

What did you think of today's newsletter?

We love all types of feedback!

Login or Subscribe to participate

📬 Thanks for reading Harris Real Estate Daily. Share this with a colleague who needs clarity about where the industry is headed.

Forwarded by a friend? Sign up with just one click here.

FOR REAL ESTATE PROFESSIONALS READY TO LEVEL UP

The All-New Premier Coaching Is Here!

You already know you can do more. Now you'll know exactly what to do.