AI Killed the 6% Commission. Here's What's Next.

Clinton Botway spent 15 years as a real estate broker across Manhattan, Brooklyn, Queens, and five states. He closed four hundred million dollars in sales and leases. He managed three thousand apartment units. He was good at his job.

Then a Microsoft engineer told him everything he did every single day, AI could do better.

That was Labor Day last year. Clinton walked through his daily routine: responding to inquiries, scheduling tours, answering questions about floor plans, managing leads, updating clients on progress. The engineer listened and said the same thing after each task. AI handles this. AI handles this. AI handles this.

Clinton looked at his own last name, Botway, and made a decision. If he didn't build the replacement for what he was doing, someone else would. So he quit his manager position at a boutique brokerage and started from zero.

Four to six months later, he had built what Zillow spent thousands of man hours and unknown dollars creating. He spent a few thousand dollars and tens of hours. And he figured out how to charge 1% instead of 6%.

The real estate industry doesn't move fast. But AI moves like nothing the industry has ever seen. And that's the story everyone needs to hear.

What Happened to Salesmanship

The traditional real estate pitch relies on a simple belief: the agent matters. The salesperson is charismatic. They know the neighborhood. They can close.

Clinton doesn't buy that anymore. "I don't think I'm ever going to buy a car because the salesperson is amazing. I don't think I'm ever going to buy a house or rent an apartment because the salesperson is amazing. I think the salesman is a net neutral."

He's not being cynical. He's being honest about how modern buyers actually move. They want to shop online. They want to see what they want fast. They want to move fast. The salesperson who opens the door and answers questions is not adding value through charisma. They're adding friction by being late, or requiring an appointment, or controlling access to information that should be instant.

The real lever in real estate transactions is not the agent. It's time. Clinton calls it "the most underrated, most important factor of a transaction that people don't think about." Non-contingent offers win not because they're legally cleaner. They win because they save time. Buyers move faster. Uncertainty drops. Deals close.

That insight changed everything about how he built Botway. The app doesn't need a charming broker. It needs to show you the apartment before you visit. It needs to verify you instantly. It needs to give you a two-hour window to see the space whenever you want, not when the broker is available. And then it needs to follow up until you say yes or no.

"You open the door, give people information, answer any questions that they have, and then follow up with them until they say yes or no." That's the entire job. Everything else is theater.

The 1% Bet

The standard real estate commission sits at 6%. Clinton charges 1%.

That's not a pricing error. That's a business model reset. When labor is automated, margin can collapse and volume can explode. When you don't need a human to schedule the tour, verify the tenant, answer questions about the lease, and close the deal, the cost structure of your company changes completely.

Botway runs on top of a network that has 1.5 million apartments available with self-showing capabilities. Only 100 of those units are in New York City right now. But the infrastructure exists. The technology exists. The unit economics work at 1%.

The old model charged 6% because it required human time at every stage. The new model charges 1% because AI does the work instead. This isn't clever pricing. It's what happens when the cost of labor approaches zero.

But here's what Clinton learned that most founders miss: the hardest part of scaling at 1% is not the technology. It's finding customers. "If I had VC money right now, I might hire some people and I would probably spend most of it on marketing. Like the tech part of it, again, not to say that it's free, not to say that it's easy. Certainly not. But the tech portion is no longer the most expensive part."

The game has shifted. Five years ago, real estate tech founders raised millions to build the product. Today, the product is almost free to build. What costs is acquiring customers at scale. That's the bottleneck now. That's where the money goes.

Why He Rejected VC

This is where Clinton diverges from almost every other founder at his stage. He has funding offers. He has investor interest. And he has chosen to bootstrap.

The reason is simple. He had a mentor who told him something that rewired his entire strategy: "Taking other people's money won't help you figure out what your company does."

That stuck. Most founders hear that advice and nod politely. Clinton actually believed it and acted on it. He decided that before he raised capital, he needed to answer one core question: what is he actually selling? Is it 1% commissions? Is it speed? Is it convenience? Is it the app itself? Is it the landlord tool? Is it the tenant experience?

VC money would have forced him to pick an answer fast and scale it before he knew if it was right. Bootstrapping lets him run experiments, talk to thousands of users, build different feature sets, and learn what actually drives adoption. Only then does capital make sense.

He's not anti-capital. He's anti-premature capital. "If I don't have income or a safety net or a wife that's got an amazing job like I do and amazing supportive family, you maybe you shouldn't do it." That's the honest version of the advice nobody else gives. If you don't have runway, don't quit. Build while employed. Prove the model works first.

The real estate industry doesn't talk this way. Tech founders don't often either. But he's right. The failure rate for real estate brokers is about 70% within two years. Small business failure generally sits around 90% in the first year or two. Taking money you don't understand how to spend doesn't move those odds in your favor. It just moves the finish line.

What AI Actually Changed

When people talk about AI replacing jobs, they usually mean AI is smarter. That's not what Clinton experienced. AI isn't smarter than a great real estate broker at negotiating or reading a room. AI is just faster and more consistent at the repetitive parts.

"Everything that you can do, AI can do." That's not hyperbole in Clinton's world. It means responding to the 47th inquiry about whether the apartment comes with parking. It means scheduling the 200th tour. It means explaining lease terms for the thousandth time. AI doesn't get tired. It doesn't resent the repetition. It doesn't take a sick day.

And that changes the math. If 70% of your day as a broker is repetition, and AI removes 70% of the friction from that work, then what is a broker's job actually worth? Not 6%. Maybe 1%. Maybe less.

The hard part for Clinton was that he didn't come from tech. "The obvious difficulty for me, I don't have a tech or technical background. I'm a real estate broker." But he learned something equally hard to admit: "You don't need to know how to code to code." There are tools now. You can build products with LLMs, with GPT integration, with pre-built components. The barrier to entry is no longer technical knowledge. It's clarity about what problem you're solving.

And he learned to use every model available. "The best LLM is not one LLM. It's all of them." He doesn't commit to GPT alone or Claude alone. He uses what works. Different models are better at different tasks. A founder who locks into one tool too early is making a bet that might be wrong in six months.

The Real Cost of Moving Fast

There's a honesty in how Clinton talks about this. "Some days you wake up, or I wake up and I might feel sick, and not a lot gets done that day, which is a little bit sad." That's not inspirational. That's real. Building a company alone while learning to code while bootstrapping without safety nets is not a grind. It's unstable.

He gets it. That's why he doesn't pretend that everyone should do it. "If you don't have income or a safety net or a wife that's got an amazing job like I do and amazing supportive family, you maybe you shouldn't do it." It's the advice nobody in venture capital would ever say out loud. It's also the most important advice he could give.

What matters is this: if you do have runway, if you do have a safety net, if you do see a gap in the market that is clear enough that a Microsoft engineer told you to build it, then you move. "I made the leap and I believe that when you commit to something, you got to do it with both feet." No half measures. No keeping the day job and tinkering on weekends. Either you're building the company or you're not.

And when you're building, you accept that the first version is never complete. "The MVP is, which is a minimum viable product, is never going to be a hundred percent done." It's going to be rough. Users are going to find bugs. You're going to iterate fast. That's not failure. That's how products actually get built.

The best advice Clinton could give any founder looking at this moment is not tactical. It's human. "The best advice that I could give anybody is find a mentor." Someone who has done this before. Someone who doesn't have a stake in your decision. Someone who will tell you the truth. And then listen to that person. Really listen.

AI didn't kill the 6% commission because AI is smart. It killed the 6% commission because the commission was payment for speed and access, and AI made speed and access free. The next version of real estate gets built by people who understand that shift first and act on it second.

Clinton built it. Someone else would have eventually. But he moved first. And that changes everything.