I Started an Online Business in 56 Minutes (Beginner Friendly) ⏐ Ep. #204

Guys, you can't see this. I know because you're looking at my face, but I literally have tingles on the back of my neck because of what just happened. I hit record. I showed my clock. No audience plans, no fakery, no trickery. And I launched a business from idea to customer that said yes in under an hour. Under an hour. Now you can see this video is way shorter than an hour. Because there was some technical difficulties, we edited out some crap. But I shared my clock at the beginning and I'm sharing it right now. And. And it took about 56 minutes to go from huh, here's a business idea. To actual real business owner in my local area that doesn't know me. He's never heard of me. I didn't find him because of my audience to say yes, I will pay for that service. One of the biggest opportunities in the world right now is AI voice agents. Specifically AI after hours answering agents for home service businesses. There are thousands of home service business niches. There are 30 million small business owners in America. Most, most of them are local, sweaty startup home service business owners. Most of them are not answering their phone after hours. They're missing out on jobs, multiple hundreds to multiple thousands of dollar jobs because they're watching Netflix with their wife now. They don't have to choose anymore. They can have an AI robot answer their calls for them. It sounds real. There's not a long awkward pause. It can take down their name and number. It can transfer them directly to the owner in bed if he wants. Only if the AI discerns that it's important enough, that it's high ticket enough. But these business owners don't know how to do it themselves. That's where you come in. And I know you don't know how to do it either, but I'm about to show you. I'm going to show you how to get the customers and how to fulfill the service in a matter of minutes. And basically with three or four software tools that almost all offer free trials, you can validate, test and launch this business and know if it'll work before quitting your job, before going all in. So I share my screen. I sent 50 texts to business owners. I picked a niche. I created an AI voice agent all on camera. My heart's pounding. You're gonna love this. Enjoy. It's 9:41 in the morning right now. I'm about to show you. It's actually very, very simple. You can do it in just a few clicks, in a few minutes. This is something that you could charge 20 to $200 per lead for as in every time they get an after hours call, you swipe their card. Or you can charge hundreds of dollars per month for this and it costs you close to nothing per month. So without further ado, let's get started. I had to start by brainstorming some high ticket home service businesses that are likely to get after hours calls. Why high ticket? Because they can pay more for leads. Why after hours? Because those guys are more likely to need my service. Like if your toilet starts spewing water at 10pm you're going to call a plumber. But plumbers are actually a lot more sophisticated than you might think because private equity has already gotten a hold of the big three plumbers, age, H Vac and electrical. So I'm not going to go there. I'm going to go for something high ticket, something that gets after hours calls but that hasn't already been attacked by private equity. And I chose garage doors. It costs between $420 $500 to get your garage door fixed. And if you're trying to go somewhere at 10pm and that thing won't open and you can't leave, you're going to call a guy and most likely that guy's sitting at home with his phone on silent watching Netflix with his wife and he's losing out on hundreds and thousands of dollars until you come in. Until I come in. So I'm going to start by scraping a database, removing all the landlines because I want to text these business owners and then only after I have an interested fish on the hook am I going to build an AI voice agent. And it's literally going to take me seconds. We're not going to file an llc, we're not going to build a website, we're not going to make a business plan, a business name, a logo, none of that. We need to start by getting our first customer. So let's get to scraping. I'm going to start by going to useoutscraper.com also if you want to get a discount, this is not an outscraper sponsored video, but you can use my affiliate link to get a discount in the show notes below. All right, now remember, I started this at 9:41 in the morning. I'm going to edit out the tech difficulties and all that, but I want you to know how long this actually takes me in real life. So once we're in outscraper, we're going to go to Brand Google, Google Maps Data Scraper, and I'm going to say garage door, not supplier. I'm going to go to Dallas. So I'm in DFW and I want to choose DFW cities because it's going to be more likely for me to close customers here even if I never have to meet them in person because they people want to buy from people that are local. So I'm just going to type in a handful of DFW cities for Worth, Kinney and Allen. These are like some of the biggest suburbs in Dallas, Fort Worth area. All right, maximum results. I'm going to do a thousand and you can choose one of these packs. I just want their phone numbers. Honestly, phone is not blank. That's all I want. All right, we're going to click get data now. It'll give me an estimate of how much this costs. $9.28. Boom. Eight to 30 minutes. Now, it was interesting that Google Maps doesn't have a garage door repair category. I was kind of surprised by that because they have thousands of categories. But if you at least search the keyword, it's going to show me some garage door suppliers and some garage door repair companies. So I'll just have to filter out the irrelevant results. Now I'm going to go to tasks, says it's in progress. While we wait for this, let's go set up our AI voice agent. All right, so I am going to log into my high level account and I'm going to use their AI employee feature and they do have like a 30 day free trial for AI employee, which means you can test and launch and validate this business without spending hardly any money. You're going to spend a little bit on the scrape, a little bit on the texting software, but we're talking under 100 bucks and you can know if you're in business or not. All right, so I'm going to go down and do explore Voice AI. Okay, look at this. This is brand new. I've not set up anything beforehand. I'm going to click Create Agent and I'm going to create custom agent. I'm going to say Chris's YouTube video business name. Chris's Garage Door Repair Voice Jessica. Jessica. Sounds good to me. My time zone. We're only setting this up for inbound calls and we're going to change the welcome message to just say, hey. You have reached Chris's Garage Door Repair. How can I help you today? Okay, knowledge base. Now, in time, after we actually launched this business, we do want to import a knowledge base which you can use ChatGPT to generate a knowledge base is just going to be something like, here's our price for a double garage door repair, a single garage door repair. Here's our surcharge for after hour calls. Or maybe the whole knowledge base is just like, let me get your info, we'll call you first thing in the morning. Or it responds with let me pat you forward to the owner real quick. Which you can also do with this. Most of these people will know that they're speaking to an AI, but they'd rather speak to an AI than nobody. And when they call your competitors, they're not speaking to anyone. It's just going to voicemail. So in a perfect world, you use the AI to get their generalized data. Name, phone, email address, and then you patch them through to the owner, especially if the AI can discern that they are highly motivated to get a job done asap. All right, names. To collect name, we don't need their email address and we're already going to have their phone number. So let's just do name and address. I want to receive an email when the call is over and I'm not going to mess about triggering a workflow with this because that gets a little complex. But it's not even necessary for this. Okay, let's go to my phone number pool. I have a number saved in my high level account. That's a local area code number, which is another reason why I want to scrape people here at local to dfw. And my working hours for this agent is going to be 24 7. All right, now let's go ahead and test this thing out. Let's see if it works. I'm going to put in my number. This is my personal cell phone number attached to this iPhone right here. And I saved the number because my phone normally ignores unknown callers. So I saved it. Let's see how it goes. All right, it's calling. That didn't take long.

Hey, you have reached Chris's Garage Door Repair. How can I help you today?

Yes, my garage door is not opening and I need someone out here as soon as possible.

Got it. That sounds frustrating. Let's get this sorted out quickly. Can I get your name please, Chris? Make sure everything's set up for a prompt callback.

My name is Chris.

Thanks, Chris. Just to confirm that. Chris, right?

Yep.

All right, thanks for confirming, Chris. Could you please give me your address so we know where to send our technician?

123 Maple Lane.

Thanks. Just to confirm, that's 123 Maple Lane, correct? Yes, thanks for confirming. Your address. A team member will reach out to you as soon as possible to get this taken care of. We always aim for a quick response. Is there anything else you need help with right now?

Nope, that's it.

Thanks again, Chris. Someone from our team will be in touch shortly to help with your garage door. Have a great day.

All right, one minute, 14 seconds. It got my information. Let's see what got captured here in High Level. There's the exact transcript. The recording is available, it just hasn't been uploaded yet. So it will be there. There's a summary, then you click save and then I will get an email summary of the call as both the agency owner, the business owner and the garage door owner would get an email of that as well. So that's literally it. We talk about these AI voice agents. I just created one. It took a couple minutes. What's crazy is like six months ago I emailed this guy from this company that was Silicon Valley funded and it's an AI voice agent for small businesses like these. And he raised a ton of money, he's got employees and he's doing great. He's crushing it. And I'm sure his calling service has a ton of other features that we can't create in just a couple minutes. But what's crazy is now just a few months later, we can recreate that same experience, almost the same experience, with just a few clicks, under a 30 days free trial. But let's get back to my scrape and see if I have this list of all these garage door business owners. All right. It says it's going to finish in about 10 minutes. So when you actually set up this AI voice agent for a business owner, you're going to want him to have his own dedicated phone number or you can port in his example existing phone number. Use High Level to create a workflow after that call is made to forward the homeowner directly to the business owner, but only if the homeowner asks a few questions in a certain way. So maybe they say, you know, it's not urgent. I just wanted him to know that I have this problem. Cool. I'm not going to forward you, but if they say my pipe is burst, my house is flooding, please let me talk to the owner, I need him here now. Then you would just tell the AI in this very simple workflow builder to forward the call onto the owner. And if he doesn't answer, then you can set up secondary and tertiary people for the AI to forward on the number two. Now if you click switch to advanced Mode, you can go into Chat GPT and say, hey, I need you to make a knowledge base or script for my AI voice agent. You are an after hours AI voice answering service for a garage door repair company called Chris's Garage Door Repairs. And you can plug in this exact prompt that you see here with the background info, instructions for handling callers queries, general rules for handling callers queries, and script instructions. You don't need to think of all these things yourself. Just have ChatGPT do it for you. That's your knowledge base. It doesn't need to be complicated. It's basically one prompt and then every time a call is made, it's going to refer to that prompt and answer accordingly. All right, let's see if our scrape is done. Boom. Okay, here's what we're going to do. We're going to click the Google Sheets button and transfer all this data into Google Sheets. Boom. Look at that. How beautiful. How many results do I have? Highlight the column. Look down here. 513 results. Awesome. What did it cost me to run this scrape? A dollar and eleven cents. A dollar and eleven cents for over 500 results? Can't beat that with a stick. Now this is a very important part of the process. We need to clean up this data because there's too much going on here. 90% of these columns we don't need. So I'm going to show you exactly how to clean these up in the most efficient way. These are all the columns I'm going to delete. A C. I mean, really, I'm going to delete almost all of these columns. I want the name of the business, I want the website, I want the address. But most importantly, I want the phone number. There's the phone number in column Q and there's another phone number in column aa. That one's going to be more accurate because this one came from Google Maps, which is more exhaustive, and this one came from Company Insights, which is probably some database like Manta. And those are a lot more hit and miss. I mean, you can see that only about 60% of the numbers are present. So we're going to delete column A, B and then we're going to delete all of these other columns. Boom. Let's put filters on here. They're all categorized as garage door suppliers because there is no category for garage door repairs. But here's what you can do. Highlight column C and column B. Go to format conditional formatting format cells. If text contains repair. Boom. Now we can see in green all of the businesses that have the word repair in their name, and we're going to start there. Now, granted, there will be a lot of other ones that don't have the word repair in their name that still repair garage doors. But we're just trying to keep this simple for now. Then you can sort by color, fill color green k. So now we're only left with 144, which is okay. And some of these companies just have multiple locations, but it's the same company. To keep this simple, I want to identify the repair companies more easily, But I'm still going to enrich all of the data because I only have 542 leads. So after I remove all the landlines and bad data, I don't want to be left with too few leads. So the most important thing I'm going to do here is I'm going to go to column aa, where the phone number is, and I'm going to go find control f click these three dots, and I want to find plus one space. I want to remove all of the plus ones and all of these phone numbers because I want to standardize them. And then I'm going to find all the dashes and replace those with nothing because I want to standardize them. So now on all the phone numbers, you can see they're standardized. But some of These businesses have one phone number but 10 locations. So I want to remove all the duplicates. So now I'm going to highlight all the cells. I'm going to click data. Data cleanup. Remove duplicates. Data had as header row, unselect all go to column aa phone. And I'm going to remove duplicates based on everything found in column AA. Boom. There were only 10 duplicate rows. That's good. Now, also, I'm going to do something very important here. I'm going to sort them all by phone number, and I'm going to make sure that there are no blanks. And this will be important for a reason. I'll show you here in a minute. I'm going to delete all of these columns, even that phone number. I don't need any of this stuff. Now, remember, this data cleanup process will be the same no matter what industry you're in. This is not just for garage doors, so don't skip over this part. This is important. All right, Now I'm going to sort these numbers by fixed line, landline, or mobile. And I'm going to delete everything else. See a lot of these. It doesn't know if it's fixed line Landline or mobile. So I'm going to delete them because I don't want to take the risk. I'm only going to keep these cell phone numbers. Now, we started at 512 or 540. We're left with 153 cell phone numbers. So I'm going to delete these columns. I don't need them now. And I'm left with all these. Okay, cool, cool, cool. So now we're ready to text these guys. So we're going to go over to straight text. And this is my favorite app for doing bulk texting. I love it because it sends your text from your MacBook or from your iPhone. You don't have to have an iPhone or MacBook to use it. You can use Android, you can use PC. But it is statistically proven that texts have a higher response rate and higher open rate if they are in blue. Because about two thirds of Americans use iPhones and they trust text more if it's in blue. Don't kill the messenger. So we're going to click new broadcast and we can put in custom variables like first name or last name, but we didn't scrape those. We're not worried about this for this. So I'm going to duplicate this sheet and I'm going to randomize. If you still don't believe that there's no audience plant after this, then I just can't help you. If you use this formula equals rand open and then close parentheses and then double click it, that will add a perfectly randomized number into each cell. And then if you sort it smallest to greatest, greatest to smallest, it will perfectly randomize this list. Okay. And so I did that here. And then I'm only going to text 50 people. Now, remember, actually, let me do this. I want to sort by the ones that have repair in the name and then everything else is random because I want to text people that are more likely to be my customers. So I'm going to delete everything after this row because I'm only going to text 50 people. And then if we want to go back and text the other 80 or so, we can go to this tab right here. So about half of these have exactly half of them have the word repair in their name and the other half don't. So we're going to go file, download CSV back to straight text. We're going to do this. Upload CSV, choose File Outscraper, 50 valid phone numbers found. And then I'm going to type my message so you don't Want the message to be over 160 characters because then it's going to be two SMSs and then if it does go out via Android and not imess message, it's twice as many. It just breaks up the text. It doesn't look as good. Also, the best reason to not make your first text too long is shorter texts have a higher response rate. That's just a fact. Right. You're going to want to tell them everything all at once because you're excited. Don't. I'm just trying to validate that they own the business that I think they own first. We'll get to the sales pitch later. So here's my pitch. Hey, I'm here in Dallas. Do you fix garage doors? I found you on Google Maps, Chris. I'm going to say do you still fix? Because that's going to like. Cause them to be like, yeah, I still do. Yeah, I'm still. Why wouldn't I still be in business? I'm putting my first name in there so they can see it's coming from a real person. I'm putting the word Dallas in there so they can see that it's someone local. So, yes, they are thinking that I need my garage door fixed. But that's just how the pitch has to be if we want them to respond. So this is all going to be in real time. Okay, it's 10:23. We've been at this for about 42 minutes so far, and we're going to send a text to 50 people. It does say undifferentiated messages. Your imessage could get blocked. It is good to put like a city name or a first name in there. I highly recommend that I'm not going to because I didn't take the time to scrape the first names and only some of these had city names. But you know what? I take that back. I'm going to add city names. I'm going to do something a little tricky. I'm going to add a handful of different city names just to make the text more personalized. All right, we're going to download CSV. Boom. All right, now that we have 50 phone numbers, they're all cell phones. Half of them have the word repair in their name, half of them don't. We've added city names so we can personalize the text message. We want them to be somewhat unique. They won't all be unique because there's not 50 different cities, but they'll be more unique than not having any custom variable in the text. So we're going to click upload CSV and I'm going to say, hey, I'm here in Dallas. I'm here in first name because remember I put the city name under first name and then I'm going to click save broadcast and then I need to see that they look okay. So you can see that it successfully imported the first names or the city names as first names. Now let's do it. Let's see. All right, look, you can see here as they send, is it blue, is it green? Android versus iPhone and is it putting in the right custom variables? This says it's putting in plano in this case. So what's cool is if I go over to my imessage, boom. There's the text right there. They're sending green, green, that one was red. Boom. First response. So I am surprised to see so many androids. Maybe because they're Google voice numbers or something. But look at these responses I'm Getting so far. 1, 2, 3, 4. 4 Responses out of 12 sent. Go over here. Now. I'm not going to respond at all until all 50 are sent out. All right, let's hit the fast forward button on this. We're done. Look at this, 50 Cent, all done. Here we go. Boom. Look at our analytics. Boom. Let's see what our response rate is in the first five minutes. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 16 out of 50, that's like 30% over 30% in the first 10 minutes, which means we'll probably end up around 60 or 70%, which is insane. I mean, look at all this. Okay, let's get to responding. All right, so I just kind of made this up on the spot. Awesome. Do you answer your phone after hours? The reason I ask is because I created an AI voice agent that will, I'm going to say four garage door repair companies that will answer your calls 24, 7. And you only pay me 50 bucks when you close a lead from one of those calls. No contracts, no money up front, no risk free setup. You have nothing to lose. And the AI agent sounds like a real person. All right, that's my pitch. I haven't optimized it. I haven't a b tested it yet. I just made it up on the spot. I want them to know it's for garage door repair. It's for them. I want them to know that I'm local like them. They have nothing to risk. They don't have to pay me anything up front. They don't have Any contracts. Now, if this business grows and it's successful, then I'm probably going to change this pricing. I might require a contract. I might require money up front. I might require a monthly retainer in addition to per lead, or maybe only a monthly retainer. I don't know. What, I don't know yet. But anytime I'm launching a business, I don't want the price to get in the way of the business being successful or not. If this business doesn't work out, which most don't, I don't want it to be because of the price. I want to have a killer offer, even if it doesn't make me a lot of money, so then I can adjust and raise my prices later on. So I'm going to copy this. And don't get me wrong, if there's demand for this, I will start this business today. You don't need to twist my arm to get me to start a business. I'm going to put my full name in there. Boom. All right, I'm going to just count how many times I send it. 2, 3, 4, 5. They said no. 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17. That guy said he doesn't fix them. 18. All right, I've sent my pitch to 18 people. That's 18 out of 50. About 20, 21 have responded so far. We're closing in on a 50% response rate. Only a couple of those said they don't repair. So as suspected, we can rest assured that most of those garage door supply companies are actually garage door repair companies, and maybe about 10 to 15% of them are actual supply companies. All right, let's count our no thankses here. No, thanks. Oh, that's another pitch. That's 19. 19. Yes, I repair. Oh, Any interest? Yes. Okay, that's a good sign, guys. I don't know these guys. I don't know anyone that fixes garage doors. Go Google this guy's phone number. It's an actual garage door repair business owner here in dfw. I'm like, I have tingles. I literally. I don't know if you can see that. I literally have tingles on the back of my neck right now because I just got a yes. Like, no, I don't have his money. Yes. He probably won't close. Most deals never close, but I got a yes in hour. Guys, look. 10:39. You see my clock? We started at 9:41. It's been under an hour. Okay, that's 20. Not interested. Okay, so these are really good numbers. Let's extrapolate real quick. 550 garage door companies, like 5 cities in DFW, but that was probably half the population. Okay, so let's say there's 1,000 garage door companies in DFW. DFW is about 2%. It's really closer to three. Let's say 2% of the U.S. population. Okay, so 1,000 times 50. Because to get 50 is 2%. Just take my word for it. 1,000 times 50 is 50,000. So we're talking about 50,000 garage door repair companies in the U.S. let's be conservative. Let's say there are 30,000, 20,000. It doesn't matter. Okay? And already I have one of 50 to say, yes, I'm interested. I want that. I don't know that he'll close, but one of 50, okay, that's a 2% yes rate. Not response rate, but yes rate. That's massive. If I can close one out of every two of those, which I really think is conservative, then that means I can close 1% of everyone I text. Okay, so if there's 20,000 in the country, I text all of them. That's 200 customers. They each pay me $500 a month for this. That's $100,000 a month. $100,000 a month, $1.2 million a year. Your costs are almost nothing. You don't need employees. If interested, we'll reach out at a later date. For now, we handle everything in house. It's not a no. And here's the thing. All those other people that didn't even text back at all, they don't know what I offer. I could easily convert them at the same rate, right? So I'm getting a 2% yes rate after only 40% of people have responded. So to think that I could convert 1% of these is very, very conservative. So I'm like, I'm super amped. I mean, look at my watch. 10:42. I've been going at this since I hit record. An hour and one minute. And it took about 56 minutes to get my first yes. This was not staged. This was not faked. There are no audience plans. This is a high ticket. Home service charge per lead charge, a flat retainer per month, garage door foundation repair, sprinkler repair, landscaping, roofing, gutter cleaning. It doesn't matter. There are thousands of home service niches. Every single one of them could use an AI voice agent. I would love you forever. If you use my high level affiliate link in the show notes, you'll get 30 days free instead of 14 days free. Almost every other sign up link out there for High level is 14 days free. Mine is 30 days free. I'll get a little kickback if you use it. No sweat if you don't. It's honestly the best AI voice agent that I've found when it comes to ease and speed to launch. There are other ones that are more customizable, that have more features, but they're more complicated as well. My heart is pounding right now. Someone needs to start this business. Please do it for me. Do it for yourself, for your family. So you can quit your job. Thanks for tuning into the Kerner office. Please share with a friend and we'll see you next time.