Dan Martell - How to start a 1-person business with AI (ask these 3 questions)

00:00:00.080 I've built multiple successful companies starting with nothing but one person and AI.

00:00:04.760 Some of them are already making millions in revenue.

00:00:07.200 But did you know that over 600 businesses fail every single day in the United States?

00:00:11.720 Most don't fail because of bad ideas or bad marketing.

00:00:14.340 They fail because they didn't ask the right questions before they started.

00:00:18.240 So before I start any one person business, these are the three questions I ask AI.

00:00:22.480 So the first one is what to sell.

00:00:25.880 Everyone starts an AI business the same way.

00:00:27.880 They see AI do something super cool

00:00:29.920 and then they try to build a business around it.

00:00:31.800 That's completely backwards.

00:00:33.500 And I was doing the same thing way before AI was even a thing.

00:00:36.220 I started building companies when I was 17

00:00:38.260 and I went almost a decade completely failing.

00:00:41.560 My first two companies, I asked the wrong question.

00:00:44.040 First company was a vacation rental site.

00:00:45.760 My dad had a cottage

00:00:46.580 and I wanted to learn some new technology.

00:00:48.280 Instead of building a business, I built technology.

00:00:51.080 And then the guy who owned atthecottage.com, 0.99

00:00:53.360 he kicked my butt. 0.97

00:00:54.280 Then I decided to build another company, a hosting company. 0.99

00:00:56.920 Why?

00:00:57.220 I thought, well, I'm paying somebody else for hosting.

00:00:59.280 I should own the hosting.

00:01:00.440 Web hosting is like selling salt.

00:01:02.360 I'm sure somebody out there

00:01:03.520 is making a bazillion dollars selling salt,

00:01:05.420 but back in 1999, I wasn't that guy.

00:01:08.540 And both died because I started with the question,

00:01:10.880 what do I wanna build?

00:01:12.260 Then I changed the question.

00:01:13.900 What problem are people already paying someone else to solve?

00:01:17.120 That one question allowed me to build Spheric Technologies,

00:01:20.300 my first real company.

00:01:21.760 Because instead of starting with the technology,

00:01:23.900 I went to find out who was already paying for solutions.

00:01:27.360 And I discovered there was a whole bunch of companies

00:01:29.660 paying other companies to build

00:01:31.400 what was called enterprise portals,

00:01:33.120 these internal intranets.

00:01:34.860 And because I followed the money,

00:01:36.740 I found out where the pain was.

00:01:38.540 I asked who was already paying to solve this problem.

00:01:41.020 That allowed me to find the customers.

00:01:43.460 That allowed me to find my partners.

00:01:45.300 My partners brought me into deals.

00:01:46.860 As a 24, 25-year-old,

00:01:48.680 I was selling to Fortune 500 companies

00:01:50.760 all because I asked the right question.

00:01:52.900 And only four years later, not only did I sell the company, I became a multi-millionaire.

00:01:58.320 Same person, same knowledge, same skill, but a different question.

00:02:02.640 Now today, I never start another company without asking this question first.

00:02:07.000 Every single company starts with the market.

00:02:09.220 The market tells me where it hurts.

00:02:11.080 They show me what they're already paying for.

00:02:12.900 Then I get paid to build a solution.

00:02:15.040 Today with AI, it allows you to get even deeper quicker without trying to guess.

00:02:20.260 And I'm going to show you how to do that today.

00:02:21.720 because there's only two types of businesses you're either a vitamin nice to have or you're

00:02:26.280 a painkiller a must-have one of them can make you some money the other one will get you rich

00:02:31.480 when i think of vitamins i think of things that like could make my life better tools to help me

00:02:36.280 stretch or productivity apps or insurance for a problem i never experienced but maybe it could

00:02:41.720 happen i don't need them i don't have an acute pain but when i think of painkillers this is a

00:02:47.080 pill that solves an acute problem. These are tools that help people make more money, get more leads,

00:02:53.060 retain their team, hire quicker, because most people have like hair on fire problems and we

00:02:58.660 need to get them the painkiller to solve those problems. I call this a frustration list. When

00:03:03.220 budgets get tight at a company, the first things to go are the vitamins. The last thing is going

00:03:08.300 to go is anything that makes money, which is usually a painkiller problems. So a lot of people

00:03:12.880 ask me, okay, well, Dan, well, how do I find painkiller problems? Well, the first thing is we

00:03:16.620 have to develop the muscle. We got to see all these frustrations that are happening around us

00:03:20.740 that sometimes we just go through life and ignore. When people start businesses, they usually struggle

00:03:24.980 with finding good ideas. I have the problem of saying no to a lot of great ideas. Why? I've

00:03:30.540 developed this muscle. These are the steps to develop this skill. So what I want you to do

00:03:35.360 is tomorrow, as you go through the world, I want you to write down anything that frustrates you.

00:03:41.180 I'm talking about simple things like, why do I have to plug my phone in? Why do I always have

00:03:45.640 to open and close my garage door. Pretty much anything that makes you go like, oh, every day.

00:03:50.540 As you build this list, what you'll develop is the skill to see opportunities. Because people

00:03:56.120 only pay for three things. They pay for money, things that'll make them or save them money.

00:04:00.360 They pay for time, productivity. How do I get my time back? It doesn't matter if I'm a CEO or I'm

00:04:05.580 a manager or I'm an entry level employee. And then they pay for status. How does this thing make me

00:04:11.100 look. If you miss all three of those, then you haven't built a painkiller. You've just built a

00:04:15.940 vitamin. The point is to develop volume and write down as many things as you can to develop the

00:04:22.260 skill of seeing potential ideas that become painkillers. And the way I see it is million

00:04:27.540 dollar businesses aren't built on $10 problems. You have to have serious problems that solve

00:04:32.800 serious pain for somebody out there that's hurting. If you got a migraine and I got a pill

00:04:37.020 and it's a hundred dollars to get rid of that migraine you're gonna buy it from me and here's

00:04:40.560 a crazy hack for this call the consulting companies already working with those companies

00:04:44.400 that are getting proof through what they're paying for to solve those problems because those are the

00:04:49.000 services that they sell they wouldn't sell a service if there wasn't demand for it so that's

00:04:53.840 how we do it manually but here's how we use ai to accelerate things i use clod but any ai can work

00:04:58.920 we got to find the pains so write this into your clod i want to start an ai first businesses in

00:05:04.640 and insert your industry.

00:05:06.260 That's where I already spend my time

00:05:07.960 and it's where I want to build.

00:05:09.240 That means it's something you know.

00:05:10.480 So give me 10, this is key,

00:05:12.660 observable and nuanced pain points

00:05:14.580 that people in this industry deal with every single day.

00:05:17.920 Not the obvious stuff everybody already knows,

00:05:20.280 the specific ones you'd only catch

00:05:22.040 if you were actually in the room.

00:05:23.440 Only give me pains that are getting worse, not better,

00:05:26.020 and rank them by how much people are willing

00:05:27.760 to pay for them to be solved.

00:05:29.220 And ask me any question you need to get clarity.

00:05:31.740 So this is the exact step that I use

00:05:33.740 to make sure that it's bulletproof. By definition, if you ask the AI a generic question, you will get

00:05:39.360 a general response, which will get you a very basic business. So we want to be very specific

00:05:44.440 around the pain points. With this, you'll get real specific problems people will pay for.

00:05:50.000 So now we figured out what we should sell that's a painkiller, not a vitamin. Here's what you're

00:05:54.840 going to ask AI. Question number two, find your edge. Finding a painkiller is just the start.

00:06:00.660 Question number two will help you figure out how to solve it in a way that only you can.

00:06:05.060 For example, when my brother started building homes, he had an idea.

00:06:08.940 How about I build the cheapest home?

00:06:11.200 Because if it's the cheapest home, then everybody's going to buy it.

00:06:13.520 Obviously, people had a pain and they wanted to buy a home,

00:06:16.120 but he built the wrong thing because he didn't figure out what his edge was.

00:06:19.340 And it's why it didn't sell for almost nine months.

00:06:22.260 He realized his edge was working backwards from the customer and building something they wanted.

00:06:27.220 He was willing to do something nobody else in the home industry at the time was doing,

00:06:30.980 which is market research.

00:06:33.080 He decided to grab all the potential buyers,

00:06:36.040 go look at the dozens of different open houses on a Sunday

00:06:39.620 and ask them what aspect of those homes do they like

00:06:43.660 and took the best ideas and put it into a new design.

00:06:46.920 And he didn't then just go out and build the third and fourth home.

00:06:50.040 Instead, he designed it, he got renderings for it,

00:06:53.000 and then he pre-sold 16 homes with deposits

00:06:56.420 before he ever put a shovel in the ground.

00:06:58.680 That's what allowed him to scale

00:07:00.280 to be one of the top home builders

00:07:02.040 and now real estate developers in Eastern Canada.

00:07:04.680 The other thing he realized is that it takes them,

00:07:07.200 on average, less than 100 days to build those kinds of homes.

00:07:09.720 So what he decided to do is make a commitment to his customer,

00:07:13.320 a promise of building the home in 99 days or less.

00:07:16.680 And what's funny is I remember the first time

00:07:18.800 he did a home show and other builders

00:07:20.560 would walk by his booth with his 99 day promise and they would say like that's not that impressive

00:07:25.120 because we do the same thing but they weren't willing to claim it he figured out what his edge

00:07:29.780 was early on for me my obsession and passion for just writing code and building software and

00:07:35.540 building any tool that anybody found useful that's what i love now i want to teach you how to do this

00:07:40.640 for yourself for you it could be your skills it might be time it might be relationships that other

00:07:46.700 people don't have you gotta mine what you have to figure out what gives you an edge in the market

00:07:51.820 the reason why i was able to help my brother go through this process is because it's something i

00:07:55.660 had to figure out the hard way it's why i teach other people the same way that i did it and

00:07:59.420 oftentimes it's teaching business owners how to integrate ai into their business so the question

00:08:04.220 you got to ask yourself is how do you figure out what your unfair advantage is essentially what do

00:08:09.020 you know how to do better than almost anybody one of the things that my brother had that most

00:08:13.900 home builders didn't have back then is a dedicated architect and because most people build in homes

00:08:19.260 just buy plans off the internet especially if it's a spec home and then just build it there was no

00:08:23.900 advantage how do you compete in a red ocean like home building to create your own blue ocean which

00:08:30.380 is one that doesn't have competition where you can differentiate and you're better this one's about

00:08:35.420 you it's what's unique to you the thing that nobody else can do and the hard part oftentimes

00:08:41.180 is you can't see your own advantage

00:08:43.060 because you're too close to it.

00:08:44.300 It's really easy for you.

00:08:45.620 So you just assume everybody knows how to do this.

00:08:47.700 You don't realize how special it is.

00:08:49.180 So it's easier to get an external source to tell you,

00:08:52.160 which is where we bring an AI.

00:08:53.500 So here's what we need to do.

00:08:54.660 The first step is you gotta connect everything you have

00:08:57.640 that it can use for research.

00:08:58.980 I'm talking your Gmail, your Slack, your calendar,

00:09:01.500 your social medias.

00:09:02.400 Make sure that it knows everything about you

00:09:04.340 so that it knows you better than you know you.

00:09:06.360 For example, my AI has access to everything.

00:09:09.100 I'm talking meetings, transcripts, trainings,

00:09:12.500 all my financials, my whole life,

00:09:14.520 all the decisions I've made for the last 15 years.

00:09:17.240 I've slurped it all in so that it knows everything.

00:09:19.800 So in that same chat,

00:09:21.160 where we already have our painkiller identified,

00:09:23.420 here's what we write.

00:09:24.360 Based on everything you know about me,

00:09:26.140 where do I have an unfair advantage?

00:09:28.000 What can I do better than 97% of people?

00:09:30.680 And the second step is if you need to give it more context,

00:09:33.560 make it interview you.

00:09:34.960 This is where you write.

00:09:35.920 Now interview me about the last 10 years of my work.

00:09:38.420 10 questions one at a time then tell me what my unique edge is and like here's the pro tip

00:09:43.720 use your voice it's way faster and it provides more context so i use a tool called whisper flow

00:09:49.020 where it asks the question it prompts me and then i can reply talking to it it transcribes it enter

00:09:54.340 way quicker talking now once you get an answer the third step is to make sure it fights you on

00:09:59.240 the answer essentially you don't want it to just be a yes man so this is where we write be brutally

00:10:03.900 honest. Is that actually rare or am I overrating myself? Where would someone who's done this their

00:10:09.260 whole career actually crush me? You want to make sure that it's not just making you feel good.

00:10:14.460 Now you cross-reference those answers against the first question. You've got the painkiller we

00:10:19.900 picked and the edge we just found. Where do they overlap? Give me the one place where my edge is

00:10:25.460 the reason this problem gets solved better than anybody else could solve it. If they don't overlap,

00:10:30.520 say so. So pain alone is a market. Edge alone is a resume. Where they cross is actually where

00:10:37.660 the business is. And that's the reason somebody will pick you over everyone else selling the

00:10:42.100 same thing. For me, my edge is helping other business owners use AI to scale so they don't

00:10:47.620 build a business they grow to hate. Why? Because I went through it personally. And I know the pain

00:10:51.480 that I felt and how hard it was to figure out the path forward. It's why I teach so many people to

00:10:55.140 do it, which is why I built my AI company OS. Full playbook on how to integrate AI into every

00:11:00.420 department of a business. I'm talking marketing, finance, delivery, all of it. And if you want it,

00:11:05.180 just find me on Instagram and DM me AI business and I'll send it over. When I worked backwards

00:11:10.060 from a real pain a customer had to figure out what I need to sell. And then I added my unfair

00:11:14.980 advantage, which is the fact that I would do this for free. Then I had an edge that made me

00:11:19.240 successful, which brings us to our third question. We have to find real buyers and test the market.

00:11:25.360 question one told us what to solve question two told us how we stand out but unfortunately neither

00:11:31.180 one of those get us paid the only thing that actually makes it worth it is getting a stranger

00:11:36.240 to part with their money you need to get a wallet which ai can help us get as you build a business

00:11:41.100 people will tell you that's such a great idea you should build it come back when you're done i'll

00:11:45.060 use it i'll pay you for it i'll be your first customer and i will tell you they're lying to

00:11:49.220 you they don't mean to but they're not being honest because if you actually ask them for money

00:11:53.640 in that moment, they wouldn't give it to you.

00:11:55.800 That's why we have to do this test.

00:11:57.560 When I started my fourth company, Flowtown,

00:11:59.320 I wanted to know if they would actually pay for it.

00:12:01.540 So what I did is I created a landing page.

00:12:03.900 I had them go through the whole signup,

00:12:05.900 go through all the different screens,

00:12:07.280 and even put in their real credit card

00:12:09.180 to validate and confirm they're willing to pay.

00:12:11.520 And guess what?

00:12:12.120 20% did.

00:12:13.300 That number told us everything.

00:12:15.600 And then I would just say,

00:12:16.480 unfortunately, our servers are over capacity

00:12:18.460 and we're not able to take new customers on.

00:12:20.420 But here's the deal.

00:12:21.100 There were no servers that were over capacity.

00:12:23.020 there was no product that was even built that simple idea that's why i call it the wizard of

00:12:27.180 oz it's the man behind the curtain i simulated the real transaction to validate to ensure a customer

00:12:34.300 was willing to pay with a credit card before i built anything there's nothing more expensive

00:12:38.780 than building something that nobody will pay for and that's just something i've done in every

00:12:43.180 company i've built since then we've launched 12 companies in the last 18 months and we've done

00:12:47.820 this process with each one of them we wanted to test before we built did they have a painkiller

00:12:53.740 problem or a vitamin problem question three for me is both validating that there's demand for what

00:12:58.620 i want to solve and for the way i've solved it it's not would somebody pay it's would they pay

00:13:04.140 for it the way i'm solving it and an even better question is how do i get a real payment over the

00:13:09.100 next seven days here's what you need to understand you need to prove it before you build it this is

00:13:14.860 why we need to figure out people fall down the path all the time i gotta register my instagram

00:13:19.260 account i gotta build the perfect design i have to have a logo a business card all before they

00:13:23.740 ever even find out if anybody wants this so the right question is does anybody wants this enough

00:13:29.580 to pay me right now because here's the way i look at it certainty brings currency you sell before

00:13:35.660 you build every single time this is how we use ai to test even faster for a third question and

00:13:42.300 And there's four parts.

00:13:43.440 First, we have to ask the cheapest possible test.

00:13:46.260 We don't wanna spend a bunch of money to validate anything,

00:13:48.360 so we need to figure this out.

00:13:49.540 So the prompt is this.

00:13:50.960 Based on the painkiller and the edge we landed on,

00:13:53.740 design the cheapest market test

00:13:55.320 that gets me one real payment within the next seven days.

00:13:58.400 No preexisting website, no brand,

00:14:00.400 no logo or business registration

00:14:02.240 before I spend hours building anything.

00:14:04.620 And then number two is have it write the offer.

00:14:07.120 So ask it, write the one page offer.

00:14:09.220 A clear promise, timeline and price.

00:14:11.500 nothing else third make it build the actual path for a payment so write this prompt now give me

00:14:17.820 the 10 people i should contact first the exact message to send and the follow-up to ask for the

00:14:22.760 money not a call not a demo cash number four and the final step then make it find the hole before

00:14:28.900 the market does so we want to write why would some people say they want this and never pay ask it what

00:14:34.500 is my riskiest assumption and how do i test for that one first now it's giving you a cheap test

00:14:39.660 you've got the offer and you bulletproof it so you know they want it. Send that message that it

00:14:44.820 gave you the offer out today and take deposits from whoever's willing to lean in. Now, here's

00:14:50.240 an important thing. We want to go manual before we start automating anything, before we even know

00:14:56.180 what people are going to say. So we do the first 10 by hand before automating a thing. And I know

00:15:01.400 with AI, you might want to start automating or even scaling this up. They're like 10, I can do

00:15:06.240 100, stop yourself. If you scale things up too quick, it's the wrong move. So for test, we have

00:15:12.160 to pre-sell. So now we know what we're selling that's a painkiller. We have an unfair advantage

00:15:18.260 on an edge and we're pre-selling to get people to pay us with money to prove they actually want

00:15:23.280 what we have the way we're going to solve it. Now, here's the thing. You could do all those

00:15:27.160 three things and do them perfectly and still walk away with nothing because there's two mistakes

00:15:32.000 that kill all of it the first one i was guilty of for the first seven years number one is you fell

00:15:38.380 in love with your solution i wanted to build stuff i was interested in the technology what i should

00:15:45.720 have did is fell in love with my customers problems because that's what i needed to learn

00:15:50.760 i needed to understand exactly what they felt was a painkiller and how they wanted it solved not what

00:15:56.380 I wanted to do. Theory and ideas will tell you what could work. A real person opening their

00:16:02.000 wallet tells you what actually does. That's why we go through this whole process so we're not

00:16:06.180 delusional. So the first time you go through it, make sure that you're not lying to yourself and

00:16:10.080 you really test your assumptions. Because the second thing is we got to make sure that AI

00:16:14.380 doesn't lie to us. Because as great as AI is, it is world class of being a people pleaser. It will

00:16:20.880 sit there and tell you you're smart and you're awesome and it'll be a cheerleader for you and 0.98

00:16:25.420 you could do no wrong it doesn't want to tell you that your idea is dumb it would never say that 0.81

00:16:29.960 instead it's going to tell you to go do it it's going to tell you exactly how to do it even if

00:16:33.720 it's not the best one that is a trap and you don't even notice that it's happening because you never

00:16:38.660 question it and everybody wants ai to agree with them everybody needs especially in the early days

00:16:43.120 when we're starting business we're just like so fragile that any negative feedback could derail

00:16:47.580 us and we just decide not to do it but please don't allow yourself to fall prey to this you 0.84

00:16:51.940 have to push back on the ai and ask it to be a ruthless mentor be brutally honest that's the 0.68

00:16:57.420 language i use in my prompts and that's why for every single answer make sure you give this to

00:17:02.360 it you got to flip it back now push back on me tell me why this fails where am i kidding myself

00:17:07.360 in the military they call it red teaming you essentially attack your idea or your plan based

00:17:13.220 on what you think the enemy will do and make that a habit and do that as often as possible you have

00:17:19.240 to put yourself on the other side of the coin you have to say like why won't this work how do i think

00:17:24.480 about it and then from that you can actually do the things that are going to protect you so doing

00:17:28.320 those things will help you not only find the right thing to buy and give you your edge but it's going

00:17:32.080 to allow you to validate so you don't waste so much time energy building something nobody wants

00:17:35.940 now here's what i've learned building companies for the last 30 years is essentially realizing

00:17:41.140 that my whole life is the byproduct of questions questions i ask myself the quality of my questions

00:17:47.320 dictate the quality of my life.

00:17:49.480 And I wanna encourage you to be more curious,

00:17:52.180 to seek continued growth,

00:17:53.920 to decide that a properly timed question,

00:17:56.780 even a question you repeatedly ask yourself,

00:17:58.500 on the backside of that is potential

00:18:00.240 for you to become even more.

00:18:02.020 So out of those three strategies,

00:18:04.200 figure out what you wanna sell, your edge,

00:18:05.800 or how to test it, which one did you like best?

00:18:07.980 I need you to leave me a comment below

00:18:09.360 to let me know that you picked that up

00:18:11.180 because the next time you start your business,

00:18:12.820 because it may not be your first time,

00:18:14.120 you're gonna use that strategy to avoid making that mistake.

00:18:17.000 And remember, if you want my full playbook on how I integrate AI into every department of my

00:18:21.240 business, just DM me AI business on Instagram and I'll send you the company OS. And if you

00:18:26.160 want to learn how to make your first 100K, click this video next.

Dan Martell - August 26, 2026

How to start a 1-person business with AI (ask these 3 questions)

Episode Stats

Transcript