Introduction: Building Zone and Using AI in Real Estate

I don't have a presentation to actually present, but I'll talk about Zone.

So, you know, we started Zone almost about four years ago. We scaled it to now 12 million plus in revenue. We've done a, we've raised money from VCs. You know, we've done all the, you know, things that pretty much every startup founder does.

But I think I'll talk a bit about like how we use AI and sort of how we build the product and a bit of, you know, backstory about Zone. own.

Origin Story: Experiencing Real Estate Friction Firsthand

So, almost four years ago, maybe a bit before that, so I went to University of Waterloo, was doing my PhD in nanotech, founded my first company, sold it, moved to Toronto, and bought my first home, and also went through the process of selling my first home in about a year into home ownership.

So I got to truly experience both sides of the real estate as a first -time home buyer and also as a first -time home seller, and realized, you know, this process seems old -school and it feels like it was built in the 1980s and it was actually

Why the Traditional Commission Model Feels Broken

built in the 1980s the Commission model how real estate works you have a mortgage broker you have a lawyer you have a real estate agent you know inspectors contractors and it feels like you know especially when you are a first -time

homebuyer you already deal with a ton of anxiety when you are purchasing the biggest transaction in your life and when you also have to coordinate with multiple people where they don't talk to each other as they should it also adds to the complexity of the transaction in itself and when I sold

my home I also realized how much in Commission's you know get lost especially you know especially in Canada when you are selling a home you pay five percent in Commission's you know the minimum down payment to purchase a home is actually five percent if it is under five hundred thousand so you know the

average Canadian it takes you know multiple years to save that down payment whereas you know two individuals walk away with that money for possibly few hours of work. Didn't make sense.

Going Viral and Validating the Problem

Posted my story on Reddit. The story goes viral.

A million plus views. Thousands of comments.

Now the problem is much bigger than my personal story. And we wanted to solve it.

Marketplace Strategy: Start With Buyers to Win

And how we wanted to solve it was, okay, it's a two -sided marketplace. We need to start from one side and work towards the other side.

And for those of you building a marketplace or involved in building a marketplace, you know, you always have to do the opposite opposite of what the current marketplace does for you to truly build a marketplace, which

means for us, the current marketplace, there are a lot more people wanting to sell their property and less people wanting to buy a home, which means we have to do the opposite, which means our marketplace should have more people who are looking to buy and less people looking to sell, and hence we can always compete with the existing marketplaces.

devices.

The Skepticism: Trust, Risk, and the “Skydiving Without an Instructor” Analogy

Now, when we had this first idea, I remember, you know, one of the Reddit posts, the comments said, dude, this is never going to work.

You know, the idea was to essentially automate everything using AI and tech and hence save the commissions and pass it back to the consumer.

And one of the comments was actually from one of the founders of Bolt, and he said, dude, this is not going to work.

You're asking someone who's going to skydive for the very first time to skydive without the instructor being attached to the person. So it's such a risky transaction.

No one would want to purchase a home without a real person. And there's some truth to it. It's the biggest transaction in someone's life. And hence, why would they trust AI to help them with that transaction?

So we said, you know, you have a point. point.

But if we could lower the risk of purchasing a home, then, you know, the risk of skydiving is, you know, I might die. But if you could make sure that the person doesn't die and today we get into an airplane and we don't ask who's the pilot, we just simply get in because that trust is established already, right?

So how do we build that trust? So that's

Standardizing the Homebuying Process to Build Trust

That's when we took a step back and said, you know, today, real estate works so different compared to the agent, based on the agent that you work with. You know, you work with some, you know, random agent from like XYZ Realty, your experience would be compared to, you know, someone else from a different agency. It would be very different.

So we said today there's no standard process of purchasing a home. home. So we said, we first need to standardize this. How do we standardize this?

So we looked at what a realtor does. They do a few things.

Number one, they essentially help you with the mortgage qualifications, tells you what you can afford and all of that stuff. And

then once you know that, then they start sending you essentially properties that fit your criteria, do analysis on these properties, set up showings, and then the showings happens, and then you

you negotiate, then you close on the property, and then all these other services that comes after you buy a home.

Product Architecture: Breaking Real Estate Into Workflows

So we said, okay, if we are to truly build an AI -native solution, we have to break this into components. And look at treat each thing as a separate workflow.

So we divided this into pre -purchase, during purchase, post -purchase, and there's one thing we cannot automate, that is the showing piece, which is where someone actually takes you to a home, opens the door, closes the door.

Decoupling Showings: An Uber-Like Network of Agents

So we said, okay, let's decentralize that, meaning let's take that out of our process, let's treat it as a separate project, and let's build like an Uber -like network of realtors all across the country so the showings can be facilitated.

So you get a showing like Uber drivers get, you accept the showing, you do the showing. showing. So everything else we mandated should happen on a chat from end to end.

End-to-End Chat as the Core User Experience

So we hired full -time real estate agent. We paid them a salary to come work for us. But they don't go on showings. Showings are facilitated by our showing network.

We gave them access to our own chat tool where they would converse with the client from the pre -approval until the offer and until they move into the property.

From Human-Run Ops to AI Automation

We documented every single part of this transaction from end to end and early this year we completed close to 700 million dollars in real estate transacted through our platform now that was enough data for us to go back and look at okay how do we now start automating this process using

AI so that is when we basically trained our models to replace portions of what these agents would do such as you know sending properties setting up showings for you, sending you PDFs of comparables, what your properties were, things like that.

Training on Transaction Data and Automating Agent Tasks

So we looked at, okay, today one of our real estate agents who work full time can handle about 50 chats simultaneously. By introducing AI, we should be able to scale that to 100. And we were able to scale that.

What’s Still Hard: Automating Negotiation

Now we're like, okay, the only thing that we haven't figured out yet is the negotiation piece, meaning everything else up until that point should be able to be facilitated through AI. AI. So you can see the process of us building up to this point.

It was, you know, break it down into workflows, then figure out what can be automated, what can be essentially using AI, and then whatever that cannot be, come up with a plan for that, and then slowly

increase the capacity of humans involved, then slowly take the human out of the loop, because if the capacity is increased, you're already replacing some of the workload that

Customer Value: Turning Commissions Into Down Payment Assistance

they've been doing and that has been the part for us today we've helped close a thousand Canadians by their first term we are across California Texas going to be in New York and Atlanta soon so by doing this the most exciting piece what

we were able to do is take these Commission's that usually sellers paid to these real estate agents we are able to pass it back to future buyers as as down payment assistance, right?

So this, let's say you are looking to buy a home for $500 ,000, the minimum down payment required is $25 ,000.

Today, Zone covers almost 30 % to 40 % of that for free. Like, you don't have to pay us back. It's not a loan or anything.

It's just, you know, we don't have a middle person to pay. So we give it back to those who are looking to purchase, and we've enabled almost close to 1 ,000 Canadians to buy their first home.

That's a bit about my story and how we use AI.

Q&A: Coverage, Selling With Zone, and Business Model

Thank you, Rishad. for giving us an intro to Zaun.

Do we have any questions? Anyone? Michael? Thank you.

That was very insightful. You mentioned markets in the US, but do you have any markets here in Canada?

Yeah. I mean, we are all across Canada. All across Canada?

Z -O -W -N. Yeah.

So, I'm trying to sell my house, and I'm with this problem of that 5 % commission, and I can't get it to sell because these real estate agents still want it. So how does Zone help the person who's trying to sell a home?

How Zone Helps Sellers: Flat Fees, Tools, and Speed

Yeah, so I think we currently work... We started from the sell side first, and what we realized was we built a ton of tools to automate the sell side using AI, home valuations, pretty much getting connected with the staging person, photography. photography, and instead of charging a percentage commission, we charge a factory of $5 ,000 to sell your home.

And what we realized was most sellers actually, you know, they care about saving commissions once you've sold the home. Everything a seller truly cares about is how soon can I sell my home, and how much more can I sell my home for? And which is truly a function of the macroeconomics at the end of the day. How many buyers are in the market to buy your home, right? Right. So no matter what kind of an agent you work with or what kind of commissions you pay, your home is going to be seen by the same number of people that are looking to purchase a home.

So we said, OK, if a seller comes to us, we should be able to tell them with full confidence that we can actually sell your home faster than anyone else does. How can we do that if only we had access to more buyers than the current marketplace? marketplace.

So we said there are people who are looking to buy in six months, 12 months in time. And what's stopping them from buying a home today is the down payment gap. So if we went to them and gave them $10 ,000, so cut that time to save by 10 more months, we would pull future demand into today's marketplace. And hence, Zone has access to more demand than anyone else does, which means if you were to sell your home with us, we would be able to sell your home much faster than anyone else.

Now, on the sell side, you're also paying AFlat for your $5 ,000, so you're saving $20 ,000 if your home is a million dollars, and you also are able to sell your home fast. So of course you can talk to me and we can help you sell your home.

So the thing is actually because we are also kind of an AI company and we are making towards Towards the realtors, their life is easy, right? Getting the leads, getting the lead nurturing and all that.

Why Zone Chose Consumers Over Selling to Realtors

But how did you come up with that idea that you have to go specific towards the customers and customers? Because as far as a company owner, I look into the realtors because they are earning money. That specific portion we will get from them. But you are trying to do something different from there. How do you come to that idea that you want to build something for specific end -to -end customers?

Yeah, I think it depends on what are you passionate about solving, right?

So when I first started the company, we applied to Y Combinator. We got into Y Combinator. And one of the things that they asked us to do was, hey, pivot to become a B2B company.

Sell whatever you are building to realtors and become a B2B company. And we actually said no, and we didn't take the offer. Why? Because that's not what I'm passionate about or my team's passionate about, right?

Because it's great to build a good business, But also, I want to be doing this for the next 10 years of my life, you know, forever potentially, right? And, you know, I don't want to be doing something that I don't truly enjoy doing, right?

Like, again, sure, I care about people. But, again, really, I don't care about real estate agents. At the end of the day, you know, they do well, they do well. I do care about people like me who are struggling to purchase a home.

You know, people like, you know, first -time homebuyers, immigrants who come to this country and, you know, who can't put a roof on top of their head. Those are the individuals that I care about. So I want to make sure the problem that I'm solving is something I'm passionate about.

So, again, is there a business to be built selling to real estate agents? Absolutely. Real estate agents would buy anything if you say that you're generating leads.

Q&A: The Future of AI, Trust, and Buyer Behavior

So how is the future of AI going to be in the real estate accounting as you're running this?

Yeah, I think more and more trust is going to be built. So, for example, even for Zone, right? At some point, we were questioning ourselves, you know, like, is this company going to be an AI -powered company or is it an AI -native company, right? Because it is such a big transaction at the end of the day, you know, are you going to be placing your full trust on the AI solution or do you always need a human in the loop?

So I think, you know, we are getting to the point, I think a lot more people are being like we see on our tests that, you know, one of the clear indications for us was when we We had a chat with an AI agent versus the consumer, and we compared that to a real person on the other side and a consumer.

The home buyer is actually asking a lot more questions when it is AI versus a real human, because real humans judge you for asking questions, right? You sometimes feel stupid, like, hey, you just placed an offer, you need to submit a bank draft of this much.

Hey, I've never done a bank draft. What the hell is a bank draft, right? But you don't ask that question mostly from a realtor. You Google that. But here, you're not getting judged.

And hence, you are able to, we see people asking a lot more questions. So I think a lot more trust is being built.

And as these AI models get really good at certain things, you are also borrowing that trust, right? Because we are using so many different tools today on day -to -day on AI.

As these models get better, you're also, you know, as zone, our trust also goes up because, hey, these models are really good. So you're already building on top of something where the trust is going to go up, I think.

Multiple Offers, Blind Bidding, and Avoiding Bidding Wars

How does zone see multiple offers and does it create bidding wars on purpose? Great question, right?

So one of the things that we actually advise right now is we don't ask our home buyers to go on offers. first, because as first -time homebuyers, you know, you want to be able to be in a place where there are no other offers, right?

And so one of the things that we advise right now is, hey, if there's an offer date, let's actually wait until after the offer date, because, you know, there's a lot of people who would just randomly place small bids, right? Let's say a property is listed for 600 ,000.

The seller truly wants, let's say, 650, but because it is listed for 600 ,000, you place, you know, a few offers. offers, but today in Canada, offers are blind. Means you can only see there are two offers placed, but you don't know what those prices

are, right? So as an agent, let's say I have a couple of friends, I can ask them, hey, place a couple of bids for 300 ,000 or 350 ,000, knowing the seller's obviously not going to accept it.

But then if I'm looking to purchase this home and then I see two offers registered, I'm a bit, you know, formal, right? Like I might, you know, I know it's listed for 600 ,000, two offers are there, I should probably go for 650, right?

So we try and avoid that by asking people not to place an offer on the offer date. Chances are, more than likely, in today's current marketplace, the property is not going to get sold on the offer date. It's not a hot real estate market today. So we actually ask them to wait until the offer date passes, and then we'll go back with an offer.

Good question, right? I think one of the things that we are trying to do is, like, today offers are a single number, right? One of the things that we've tested is what if we send a letter with the offer as to explaining why I really want to buy this property. Get sellers to have a bit more context than just a single number.

And we've seen that work. Honestly, some people actually, if the difference is about $10 ,000, you would rather, there are certain cases, obviously, you want the more money for your home. But how can we explain why I'm placing this offer and a reasoning layer for that?

Today, most offers are just 650, 650, right? But if I say 650, because this is what's happening in the market and this is what we think the market's going to do,

and hence you should accept this offer. So we are testing a bit of it, which is what I said,

the last 20 % we are trying to automate, that is the 20 % we are stuck at. So we are trying different things to see how we can automate it.

Q&A: Y Combinator, Team Evolution, and Growth Metrics

Hi, I'm just curious about your YC. How far along was the product when you got accepted to YC? And how many co -founders did you have? Or was it just, like, an idea?

Yeah, I think back then when we got, like, this was idea phase. We were barely making any money. I had three co -founders.

I had to let all of them go. So, you know, so right now I'm just the only founder.

I would love to hear the numbers. How much revenue have you generated year over year? What are your goals for the future? Given that this is like a very future -proof industry and business. What's coming next for you? And how big is your team currently out of curiosity? Are you investing?

Maybe, maybe. We did 1 .05 million last month, so annualize, multiply that by 12, so in terms of team, we are a team of about 45 employees in Canada, we have about three employees in in California.

What was the other question? Oh, what's our plan for, yeah.

So we went from $1 to $12 million in annualized revenue within the last 16 months. So our goal is, how do we do it again? How do we 10x again?

So that's the entire focus right now, which means bringing in people who have seen this scaling phase before, which is why we are building that leadership team

Early Product Evolution: From WhatsApp Chat to AI-First Support

to take us scale scale to that number so the origin story was you were just like a chatbot between the like the purchaser and the agent and then after x amount of months you realize that you can just remove the guys who actually do the selling of the houses like is it person talking with ai then a deal comes and then you send it over to the one that's selling like what was the

the origin story exactly of the yeah no i think you know um how we started was essentially it's just like you know the agent we hired the agents full -time got them to essentially do everything on a chat on a whatsapp pretty much right so you would come into our platform hey i'm looking for this property then they would go and look at properties send you listings um they would ask

questions about properties hey you know i like this property but then what's the square footage of the basement things like that they would answer these questions once you've essentially like the few properties, then they would set up the showings for you. You would go on the showings through our showing network.

You come back and tell, you know, these are the few properties I really like. What do you think the compatible markets look like? What offer should I place? And you know, that entire thing. So we truly had a human that does everything.

But we also had a listening bot that would listen to the entire conversation. Now once that part was done, we were like, okay, these are the ten things that we can easily automate using AI. You know, sending listings, PDFs of compatibles, whatnot.

So what we did was essentially we added an AI assistant that would essentially answer 70 % of the questions that the buyer is asking. And then now what we are doing is it's just the AI and the customer only in the chat as opposed to a human in the loop completely. So that's the evolution of sort of, you know, the product.

Okay. That clarifies it, I guess. Thanks. Of course.

Q&A: Creating More Buyers by Bridging the Down Payment Gap

Could you explain again how you increase the number of buyers that are attracted to any particular property? How you increase the number of buyers? Especially today's market.

Yeah, so let's assume, right? Let's say downtown Toronto, there's 1 ,000 people looking to buy a condo in, let's say, you know, in the next month.

Now, there are people who are also looking to buy a home in two months, three months, six months, and 12 months. And we spent time talking to these individuals asking, asking, hey, what's actually stopping you from buying a home today?

Obviously, people who already have the money to be able to purchase a home, they're waiting for market conditions to get better, interest rates to go down. But then the vast majority of people who are still not able to purchase a home because they don't have the down payment ready yet.

And we then spent time understanding how much do first -time homebuyers put towards their down payment on a monthly basis? How much do they allocate from their paycheck? check.

And we realized most first -time homebuyers spend roughly about $1 ,000. They put $1 ,000 set aside towards their home purchase on a monthly basis. So meaning, in 10 months, they would have $10 ,000 to be able to purchase a home.

Now, what we said was, OK, if seller's already paying 2 .5 % in commissions to the buying agent, we can also become the buying agent in this transaction. Take that 2 .5%. Because we've automated most of the process, we can give half of it back to the person who's buying, instead of a cashback, we built a fintech tool that allows us to advance this money to you.

So let's say you have $20 ,000 saved up so far, and you're saving $1 ,000 per month. So let's say you're buying a condo for $500 ,000. The minimum down payment that you need is $25 ,000, meaning you have $20 ,000, you're five months away from being able to afford this $500 ,000 condo.

but today you come to zone we will give you seven thousand dollars basically for free so now you have twenty seven thousand dollars meaning you don't have to wait for five more months you can buy a home today all right so that's what we've done no you don't have to pay that back so it's basically

free money right because the seller is giving to the agent but we've automated most of the work the agent does so which then allows us to like we can do whatever we want with their money but we We advance it to future buyers to bridge the gap between their savings and what's needed to purchase a home today, creating an advantage to the seller as well, because now you have more buyers that could possibly afford to purchase your home.

Conclusion

Awesome. Thank you so much. Thank you so much, Richard, for coming in.