827 | This Solo Agency Owner Is Racing to $1M With AI Agents in 9 Months | Siloh Moses

I’m one guy with zero employees and I’m racing to make $1 million in nine months. 31 days in, 70k, no ads, all organic. One guy, a team of AI agents, a million-dollar clock. Is that possible? Honestly, I think I’ll hit it faster. Watch me. Would you believe that not long ago this man was homeless? Not down on his luck for a couple weeks. Homeless in Las Vegas for seven months. Same clothes, eight, three bites of whatever I could find to eat that day just to pacify the hunger. Instead of saying, how do I survive? I started to ask myself, well, what do I need to do to get out of this? Right now AI is creating a new class of business owners and it’s not about coders and it’s not Silicon Valley. It’s real people who figured out how to sell it. Meet Siloh Moses. Between then and now, this man built a charity from one pot of spaghetti into 800 volunteers serving 176,000 meals and he hated it. So when he started over, he made himself one rule, no employees ever. He built a company called Uno OS where every role in the business is filled by an AI agent and now he installs those same agents into small

businesses stuck in the exact chaos he came out of. And right now he’s 31 days into a public challenge to hit his first million dollars in nine months. I replaced a $1.2 million team with zero employees. I haven’t hired a single person and I never will. In this episode, you’ll learn the exact same system he’s using to build this. When I found AI agents, I looked for a way to be able to get the same impact to have the same vision, the same mission without the chaos. The local businesses he’s selling AI to. The average business takes 47 hours to call a lead back. So while you’re on a job, a plumber down the street is bleeding 12 grand a month. He never sees.

And the unique way he pitches to them. What are my agents found? A $66,000 hole in a guy’s company on a live call. Siloh, take me to day one, seven months homeless in Las Vegas, no addiction, no mental illness. How does that happen to a guy like you? I was working for another company at this time and his company was a call center. I was working in his company as a call center rep. Walked in one Monday. Big boss was there. Big boss said, I need to talk to you. Essentially, that conversation led to need to let you go.

Let me go, but told me that he would give me a severance paycheck that fall on Friday. Couldn’t work, but at least I get a check Friday. So by his time, life was already kind of tough. Already 30 days behind on the rent, 60 days behind in the car payment, showed up Friday as early as possible. And found out that the business was shut down, that the offices were empty, and the owners skipped town with everyone’s last paycheck. And that was essentially day one of homelessness. And so seven months, 210 days, same clothes, eight, three bites of whatever I could find to eat that day just to pacify the hunger, save the rest for later. So it wouldn’t starve. Instead of saying, how do I survive? I started to ask myself, well, what do I need to do to get out of this? And what is my first step?

And those two questions essentially allowed me to get out of the situation and create an organization that served 175,000 people. And so the charity was the same move. It was a single pot of spaghetti, which I went out on a street corner. I served 30 people. I just went home and told somebody what I did. And the following Monday when I went to serve, they came with me. And then we grew that into an organization of over 800 volunteers serving 176,000 meals. And AI agents are just like the cleanest version I’ve ever had of doing that so far. What flipped for me was the impact that I was making with a previous organization.

And the impact that we were making it, we were making it with a lot of complexity, a lot of chaos. We had the best intentions. We had purpose. We had drive, we had mission. And what I found was that even with all that purpose, drive and mission and the chaos wrapped in that it wasn’t sustainable. And so I found AI agents, I looked for a way to be able to get the same impact to have the same vision, the same mission without the chaos. So you find AI agents and the line that stops people on your profile is this, I replaced a $1.2 million team with AI agents. Take that apart for me. What does that $1.2 million actually pay for? I’m in with my charity organization, again, a group from just me to 100 plus people. And I wound it down again due to burnout and COVID just as an opportunity for me to kind of reset. It still runs today. It just runs smaller.

And what this taught me was that I never wanted to start my next organization or my next company with 100 personalities and 100 paychecks and all of that internal chaos that comes with it. And so the goal with where I am now with AI agents is how do I get the same impact of 100 people but with less than 10 or even fewer? And that’s what led me to AI agents as being a single founder with zero employees and a team of AI agents to act like 100 who get the same impact of 1,000. And I employed and built agents that would handle roles that essentially I’m not the best at. So I have an agent that does my CFO, that’s all the accounting agent. It does my CTO, he’s my chief technical officer. And so what we do today is we do the same thing. We actually build out the same agents I built for my organization. I then lease out to other companies who have the same kind of complexities and chaos within their businesses. I’ll be straight.

Like we’ve only been selling on, you know, as for 28 days because that’s when I sparked that I started my challenge. And the challenge is to do 1 million in nine months with a single founder and employees. And now that’s the window in that window we’ve done 56. So we’ve done 56k. Actually today the lease numbers were 70. And so that’s a soft launch. That’s no paid ads. It’s all organic marketing and referrals from existing customers. Everyone in AI is chasing the flashy tech startups. You went the other way plumbers, HVAC, roofers. Why is the unsexy niche the smart bet? In home services, speed is the product. It’s not necessarily what they offer as far as applying a customer. It is what how quickly can we get to that potential customer.

Now the average shop doesn’t matter if they’re 1 million or 5 million or 6 million or more. Average shop takes around 47 hours to call a leadback. And that’s 70% of buyers already go with whoever answers first. So these businesses lose 20 to 30% of their booked jobs due to no shows. And that’s a tech driving out to an empty house. And that’s fuel and that’s time and that’s insurance and that’s maintenance on that truck just for that tech to show up to an empty house. And so no one’s serving them. And what I mean by them is everyone off building AI tech for tech startups. Meanwhile, the plumber is losing 9 to 10,000 a month to jobs that he never gets because you respond to too slowly. For me, it’s where I find the most complexity. So as a founder of a new company, I also just started this week actually cold calling. And so I started cold calling businesses in a specific niche that I wanted to dive into. It was very interesting to data that I got back in real time. And so I’m calling this one industry and I found the business owners, I found their mobile phones and my agent went to work, sent out a message,

but I replied to everyone in the back because I wanted this real time data. And so I replied, I sent them a message that was like, hey, a very simple message. I said, hey, are you still doing X? Let’s call it X like roofing. Hey, are you still doing roofing? That was a simple one line text. And I sent 100 to test this market 100 text messages to qualified business owners. I knew they’re business owners and it was their mobile phone numbers. So can you guess how many of those text messages got back to me within the first 90 seconds out of 100?

Three.

Close to, can you guess how many got back to me that day out of 100? None.

Yeah, like one more. Can you guess how many got back to me out of 100 in the first 72 hours?

Yeah, three, none. And think about that for a second. I sent a text four days ago. And yesterday, I got one of my latest texts. Three days later, the owner gets back to me and says, yes, we’re still doing roofing. That’s three days. If I’m a business owner, if I’m a homeowner, and there’s a leak in my roof and it’s coming through the ceiling and my floor has been flooded, I’m not waiting three days for you to get back to me. Usually the first place that we can do that with is going to be their top of funnel. It’s going to be like lead response time because everything else depends on the leak converting. A tighter calendar doesn’t necessarily mean or matter when the booking is already gone because the competitor took it because they responded faster. What we have is what’s just called a lead response agent. The lead response agent actually does that. So lead comes in, the agent doesn’t sleep, doesn’t call out, doesn’t get stuck in traffic, right? So the agent’s working 24 seven. So that lead comes in, that lead is now responded back in the first 90 seconds. And that lead is responded back to in a very personal way as if it was a human being, hey, John, we saw you just called us or

submitted a quote. We looked in to have someone come out and fix your AC unit. We understand it’s very, it’s 120 degrees today. And having your AC go out is probably not the best ideal situation for you. Tell us a little bit about your AC unit and let me see if I can get emergency tech out there today. And that’s the kind of conversation our AI agents are having with their leads. They also have the highest opportunity to be able to convert that lead into a paying customer. Because once you’re on the phone, the first company to answer is most likely they go with that company. Okay, so that’s how you get new money in the door fast. But you’ve told me the part that really makes owners jaw drop. Isn’t the money coming in? It’s the money already inside the business bleeding out where nobody’s looking. So put a number on the table.

Tell me about a time one of your agents caught something that a human completely missed. Okay, so we have property management company owner has 60 plus employees. They’re doing a 6 to 7 million and came to me simply because out of referral. And so when they came to me, I said, okay, what do you need? They said, well, what we really need is we need marketing. I go, okay. Let’s kind of look other places because I have a feeling that you might need something else. He found out that January where it was slow, he has numbers. So we fed his financials on the back end to our AI agent, we call Chris Castellini. Chris backstory on him is actually a CFO, one of the world’s best CFOs. And so we designed our CFO for him to be model one of the world’s best CFOs. And Chris Castellini is still alive today.

And so he did this, Chris did this on a live call. And within minutes, Chris found that the company in January alone did 507,000 revenue. What he also found is that 304,000 of that went payroll, which meant 60% of his revenue was going to admin. Now, Chris also in real time ran benchmarks against other companies his size in his industry. And when he did this, he found that this was a service business, well, the average service business was doing 30 to 40% revenue towards payroll. So this meant that this owner was in an unhealthy place. He was making $500,000 in a single month. So he had run $66,000 loss and didn’t even know it.

Now, Chris didn’t just open it and flag it, he compared it to payroll ratios and benchmarks again to other companies in his industry of his size and found that this was outside the healthy range. And then he built a 90-day game plan. And a 90-day game plan was number one, we need to raise your prices 7% to 10%. Number two, we need to put you on a hiring freeze. And then we consolidated his admin roles, which Chris found out a lot of them were overlapping. When we ran the numbers, just by this efficiency, he’ll do an extra $1.2 million. So when we go into a business, what we look for is, we look for, again, the complexity of the chaos and how we get there is by asking a few questions. And so the first thing we do is, okay, how many employees do you have? And we’ll go, okay, well, we’ve got, say, 100 employees, we’ll call it 101. Then we go, what’s your average revenue per employee? The average revenue per employee is the money we expect back when we hire someone. When we hire someone, we put them in a seat for 50K. We’ll just call it that $300,000. Tell me about your employees, tell me about how many meetings they have per week, supervisors and managers, the VP’s, the C-level

executives, they’ll do maybe two to three meetings a day. Okay, two to three meetings a day. And that’s across the entire business. We’re going to go to three meetings per day. Oh, three per day is going to be 15 a week. We go, okay, so that’s 15 meetings per week per employee. Now, a general rule of thumb is if there are how many 15 meetings per week, then we need to double that as far as the hours that they’re spending in admin, because they’re going to take some time to get ready for that meeting. Then they’re going to take time to actually host the meeting. Then they’re going to take time to do the post-meeting stuff. So it is sending the emails, who is responsible for what, is the time spent in a Slack and organizing the next meeting. And so we need to double this to 30.

And so that will go to 30 hours per week in admin. So now what we see here on the right hand side is they’re revenue per hour is actually negative. And this is what we did for that contractor. We came in here and go, you’re negative. You’ve lost money not only this month, but if you stay this way, you’re going to continue to lose money for the rest of the year or for infinity until you have decided to do something new and change, make a change. And so what we found is that in this example, admin hours per year per employee is 1500 hours a year per employee spending in emails, meetings, post-meeting, pre-meeting, and the meeting itself.

But we also found that there’s revenue is locked in these admin hours. And that revenue is $237,000 per employee. And when we come in with our, we essentially say, we know we can cut 50% of this just off the top, but we don’t say that. We say, we like to come in under a promise and over deliver. So we’ll say, we’ll come in at like 20%. Again, the employees still work there. Employees still have a seat. They still have a desk. They still have an office. They’re still in meetings. We’re just reducing the number of meetings that they’re taking per day. And we’re reducing the amount of hours it takes for them to get ready for that meeting to actually host that meeting and to post me notes in the emails. That’s all being, that’s all redundant, all being replaced by AI. And so what we found is we’re able to recover 312 hours per year per employee in admin. And additional revenue now has gone up $47,000 per employee.

That’s gone up, right? Up $47,000 per employee. Now the amounts of hours saved for the team is 31,000 hours a year. Following this example here, we can provide them with a $4.79 million lift in the first 12 months. So the agents win the lead faster and they find money already hiding inside the business. But all that assumes the phone rings in the first place. Let’s talk about you. You’re a one man shop. You publish online every single day. So where does the pipeline actually come from? And how do you produce that volume alone? You could build the best system in the world and it’ll starve if nobody calls. So we publish every day. And I say we, me, myself and my agents, we publish every day.

Every piece of content is written in my voice. It’s written in my style. And my agents are specifically trained on who is silo. What is silo believe? What’s silo’s mission? What is silo’s vision? What is their 10 year goal or five year goal or 30 day goal? Who is silo’s customers? Now the content engine is Gary Vaynerchuk. He does the research, he does the drafting, he does the posting. When someone comments, Gary replies in seconds. So you ask him for a free download. He’s going to reply to that comment. He’s then going to move the conversation into the DMs. And then he’s going to qualify by asking a few questions very conversationally. And again, you don’t know that you’re interacting with an AI agent.

So very conversationally, he’ll qualify them by asking them a few prompted questions. And then if they do qualify, he’ll push them into our CRM. He’ll call with them, send out the confirmation, reminder notes. And I show up on the Zoom call with notes about and transcripts about their entire conversation, everything Gary can find out about their business that’s publicly available. And I’m 100% ready to go for that phone call, even though I never spoke to that person. So not only are we looking to create content just for sake of content. No, we’re creating content that speaks to a very specific audience that drives some type of CTA at the end. And the CTA is usually going to be book an AI ready assessment call. AI assessment leads to a 90 minute call.

We provide the business owner with a written report that essentially gives them an idea of where they can install their first agent in their business, whether they work with us or not. Everything you’ve described is automated. So be honest. Why do you still physically drive out and sit inside a client’s business before you build anything? Why does the human still matter? Yeah, it’s a great question. For me, the soul of the business is going to be getting into the business and sitting down with the owner and sitting down with all the employees and the people who make decisions there. You don’t get that from socials. You don’t get that from a cold email blast. You don’t get that from scraping Reddit. You get that when you actually go visit the business and you tour the facility and you touch their trucks and you talk to their customer service reps and you talk to the sales guys. You talk to the people in marketing.

You talk to accounting. That’s where you start to find what the soul of the business is. And yes, everything is automation. But automation is essentially to speed up, again, what is the vision of the company, the mission of the company? What is the vision and mission of the founder? So an owner feels that and they want you in their corner and they sign on. Walk me through the first 30 days and where’s the step everybody else skips that you don’t. I would argue against that and almost say that they’re already breaking in real time in front of customers with their existing system. Because your existing system is taking 48 hours, sometimes 72 hours to get back to a lead. But what we do is we build, we don’t build anything. What we do is we map out. We look for data. So what we’re looking for is we’re looking for leads that come in. Again, their responses and we’re looking at their sources. Are you responding back in 90 seconds? Are you responding back the same day?

Are you waiting four days before you get back to someone? And then we look at the response time, the no show rates. And then we start looking at their other tools that they’re using. Week two we do is then we go ahead with all this data and we start to build. And then we start to build out the lead response agent to get back to them in 90 seconds, not four days. And then we connect all types of forms like on their website, etc. We connect our calendar, their CRMs, our other tools, and we do it in their written voice. Week three what we do is that now this is the part where most operators skip is we actually start testing before we release the model. And so we’ll test it. We’ll say a lead will come in from my agents. My agents will go onto their website, submit a lead at like 9 p.m. on a Friday night. And then what we’ll do is we’ll test the leads and the leads that the lead who doesn’t reply or the lead that wants to reschedule will have the AI agents trained to continue to follow up with them until we get an answer. Now this is where we find our best use cases because the testing part is usually not done by any other operators.

We test before we go live.