Speaker 1663 people signed up for my software in the first eight days and about 414 of them handed me their credit card but so far it has made me zero dollars zero and i'm thrilled about that let me explain so about 15 months ago i paid 800 bucks for a domain name on a hunch and then i hit record while i figured out if the idea behind it was even any good and about 95 000 of you watched or listened to that episode last week that idea finally went live and it's called lazybooks.com and it's me taking a swing at quickbooks a company that i hate with all my heart i don't hate anyone come on but i don't like quickbooks as a software so it's david versus goliath and of course i'm the david in this equation and in this episode i'm going to give you the entire autopsy why i started it what it cost every number from launch week good and bad we'll talk about the stuff that broke on day one the money that i wasted and then the part that i care about most which is how you can do this exact thing test an idea and launch it with no audience and about 100 bucks stick around for that because the cheapest test in this whole video cost nothing and it's sitting in your phone right now and yes of course this is my company so i'm biased but i'm going to try really hard to make this the least ad-like thing i've ever made it's kind of got me in a tough position because you know a lot of business guys on the internet talk talk talk and talk and talk and they're not actually building you know a lot of business guys on the internet talk talk talk and talk and talk and they're not real businesses or running real businesses and i am so i want to talk about it and i think you guys want to hear it but i don't want you to think this is just an ad so i'll walk that line very carefully so this won't just be roses but you know good bad and ugly and all that okay so first of all why would anyone want to start an accounting software company on purpose and especially someone like me who got c's in accounting 101 201 all those my my mba program i did poorly in accounting i am terrible at accounting and my business partner slash the guy who actually developed this thing not vibe coded but real coded this thing justin he doesn't know anything about accounting software either and also why would you start a software as a service business a sas business in the age of ai that's supposed to be killing sas businesses right well we'll talk about that and when it comes to why i would start a software business you know i've started like 80 businesses or whatever and at any given time i've got a whole pile of them and every single one of them needs its own set of books and with quickbooks every company is its own freaking subscriber subscription its own login its own bank connection its own monthly bill so i'd sit there logging in and out of the same software over and over paying for it over and over sorting through dozens of reports when i only really need a profit and loss statement and i don't know why they don't just list it at the top and then of course the bank connection would always break and they kept raising their prices and if you've used it you know you reconnect the bank it works for two weeks it breaks so i don't speak accountant i don't know what a chart of accounts is i literally don't t-table debits credits i'm so terrible at this stuff i just don't know what a chart of accounts is i literally don't want to know did we make money this month or did we lose money this month so then on august 1st of this year quickbooks raised their prices yet again the middle plan went from 115 a month to 140 the expensive plan went from 275 to 340 and the starter plan is now 38 bucks per company so let's say you've got three little businesses side hustle and llc for whatever three times 38 114 a month 1400 a year on the cheapest plan to do your own bookkeeping in the age of ai and they're totally garbage so that fact is the whole business that's my thesis for starting this every entity you have on quickbooks is a separate bill and so remember that because later i'm going to show you how to find that one fact for your idea okay now me being annoyed at something is not necessarily a good reason to start a company i've started plenty of companies on a hunch and you can go watch my six biggest failure episode from i don't know six months ago or something to see how that went but i want to tell you how i tested it with real people before i let myself fall in love with it to be clear i've never written a line of code in my life i'm not technical yada yada yada remember that too because it'll matter here in a minute but step one was i wanted to ask ai to tell me that i was wrong so back in august of 2025 i opened up four different ai tools and gave them all the same job go read hundreds of real reviews of quickbooks in zero on all the review sites like g2 or whatever and on reddit and tell me what people hate the most and what they can't live without and they all came back with the same two things number one the bank connection breaks number two this is the one that got me the price and the complaint was very specific people can handle expensive but what makes them go crazy is a bill that keeps going up every single year and they do it because they can they have pricing power they know that their customers are tied in they're integrated their banks are connected they know the software they use the software it's connected to shopify and plaid and all these other tools and so they exploit that one of the ai tools i used even scored how people feel about these companies on reddit quickbooks got a 38 out of 100 zero got a 48 and so basically everyone's a little bit miserable which if you're a guy looking for a business idea this is great news and people tell you chris counting software is a commodity what's wrong with commodities commodities equal massive market corn is a commodity gold is a commodity we're not talking about solutions right we're not talking about there's no perfect type of business there's just trade-offs cotton is a commodity because all we wear are t-shirts and if million small businesses in this country and 28 million of them have no employees. It's just one person and a bank account. That's who I'm going to build this for. So all these angry people on Reddit wrote my marketing for me basically a year before I had anything to sell. We wanted to be cheap with a price that doesn't just go up. Step two, I built the landing page before building the product. I did it on camera. It took eight minutes on a site called Card that cost me like two bucks a year per website. And the headline said, I'm quoting myself here, QuickBooks sucks, LazyBooks doesn't. Be notified below when it goes live. Not my best copywriting, but it has been collecting emails ever since. Step three, and this is the one I want you to steal. This one came later this past March while the thing was still being built and before anybody outside could even touch it. I asked real business owners two questions. Number one, which of these features can you not live without and which ones could you happily skip? That's the whole survey. And then I listed all the features. I put it in type form and I sent it to a group of business owners from my audience. 433 of them answered in the first 12 hours and by the end of it, 589. And here's what they said they can't live without. Reports, 62%. Just show me the profit and loss, baby. And I agreed with that. Connecting my bank. Can't live without it, 46%. Sending invoices. Didn't expect that. We weren't going to build this until later on. 41%. Reconciling, okay? Which is accountant speak for making sure the books match the bank. 37% want that. And then way down at the bottom, which is the feature I was kind of banking on, was ask the AI questions about my books. 11%. That's it. In 2026, when every software company on earth is slapping AI on the front of the box. And I'm glad I asked that question because it saved us a ton of time and money because we were about to build the flashy thing, chatbot, and we ended up building it before launch, but we didn't build it first. But almost 600 people looked me in the eye and was like, dude, we want reports and we want them to be cheap. So that's what we built. There is AI in it and it does the boring job. It looks at every transaction, guesses the category, and you can tap approve. But you're the boss. The AI is just kind of doing the grunt work. And here were the features that people said they could do without. Mileage tracking, Shopify integrations, inventory tracking. So we skipped all of that. Payroll was kind of weird because it showed up on both lists. Some people can't live without it. Some people don't care, but I'll get to that in a minute. So my brother-in-law, Justin, built this. He's a real actual developer. He's got a great job. He makes great money and he wrote the code. I brought the problem that I was dealing with and the audience and the research from my four different AI tools and a bunch of opinions. And then he did the hard part. So if you're not technical, you've got two choices. Number one, find a Justin. We all know a developer or number two, vibe code it. And I was a little nervous about vibe coding this because there's sensitive data involved. And so I'm glad I worked with Justin. But we all know Justin's or you can keep the first version of your product small enough or less feature rich that, you know, something like Replit could do a pretty good job building it securely. But this took a long time, mostly because Justin's a busy guy and he's working on other stuff. I bought the domain in August of last year. It was a premium domain for sale for 800 bucks, but it was worth it. We had a beta product up this past April and we launched November 9th. And the vast, vast, vast majority of those 15 months were just, not working on it at all. It's not like it's something that was worked on every day or every week. And I say that because, you know, a lot of people on the internet, myself included, are talking about things getting vibe coded in a weekend. Now, a lot of things can get vibe coded in as little as an hour, but this one, I just didn't want to risk it. The bank login goes through Plaid, payments go through Stripe, and we don't store your bank or card info at all. So we decided to charge 19 bucks a month because I do think that software businesses are going to and the prices are going to go down. And so we just thought, let's just start low. That covers up to three businesses. Every business after three is five bucks a month. First 60 days are free. And the price you start at is your price for as long as you stay. Why 19 bucks? Well, I just kind of said that on camera 15 months ago and I didn't want to change it. So we stuck with it. And our costs on this are very low. We pay Plaid 30 cents per month per bank account that our customers connect. And our, our customers connecting on average, like two accounts a month. So we're credit card, debit card, bank account, whatever. Some fives, a lot of them won. So call it a dollar per month to Plaid. And then, you know, Stripe takes 3% so that's another 60 cents and then our ai credits are really cheap so even at 19 bucks we can still make good money of course it doesn't account for our time but we don't have any employees yet and we really want ai agents to do as much as possible and they're already doing a ton of work which i'm about to get into i know there's a couple other guys out there for 19 bucks quickbooks has a free plan now that does basically nothing but we don't do payroll no inventory no mileage no actually job costing i think is almost ready we are going to come out with an iphone and android app here soon we have a an mcp like a clod chad gpt grok gemini connection which is super cool and an api but we will be on the clod and chad gpt app store and the iphone app store here soon so that's pretty cool okay launch day monday november 9th here's exactly what i did and exactly what happened i sent it to my newsletter and i posted a tweet that was the launch uh the tweet did about 200 000 views and 250 replies and if you're watching this i'll show you what signups look like so these were page views pages on launch day thousands i think mid-thousands these were organizations that added their credit card per hour right on launch day till today number of people that have canceled quickbooks 176 as of right now this is cool number of new customers we have and where they're coming from so facebook brave browser twitter abc newswire.com i'll talk to you soon about that bing google youtube duck duck go so we are almost exactly 50 50 split between mobile and desktop after the u.s canada and great britain and australia are three biggest markets here's our error rate almost nothing 86 of organizations add their credit card and 63 of users add their credit card day one our signups for 342 day two cut in half almost exactly 170 day 358 then 20 then 10 that's basically what every launch on earth i've ever experienced sounds like it's a firework big boom and then it's dark again and by day six we're down to like 10 signups a day but i'd stop promoting it and we're going to do a product hunt launch and more newsletter mentions and facebook ads and google ads we got a bunch of stuff planned but over this last 10 days we've just been focused on handling feature requests and tech support stuff so if i only showed you day one you'd think i was a genius but uh after eight days we had six hundred and six hundred three signups 414 of them put in a credit card it's about 63 or whatever 181 for my newsletter 75 from twitter 63 i have no idea 15 came from google so the newsletter beat everything else combined and i wanted to be straight with you about that because most people don't have a big newsletter those came from years and years of building a list of 270 something thousand people who trust me enough to try something so that is an unfair advantage and i'm going to spend the one i mean everyone has one but if you don't have that one there's another thing hiding in that data and that's google signups they doubled on day two from 15 to 31 and i think that's because people heard the name from somewhere word of mouth they saw the tweet they they wanted to make sure that google like blessed it so oftentimes people will see a link or see a tweet and then they just go to google and type it in so it's not like a pure organic customer but it's close enough but it's just a lesson that you better own the first result for your own name on day one sounds obvious but check that google search console is your friend and one of my favorite numbers of the whole first week was out of the first 357 accounts 30 of them set up two or more businesses on the first day and that's about eight percent of everyone so people really are out here running more than one thing and not one person has complained about the price people keep asking about payroll 14 different people to be exact number two is job costing which is tracking money by project which i think is ready now but payroll is a big thing and it's a big thing and it's a big lift there's a lot going on there so we're not really in a rush to do that but on day one things broke in public so the tool that imports your old quickbooks data threw out errors for some people our support email bounced the contact button was broken the chat inside the app showed an error for a couple people at least one person told me that they had already canceled quickbooks before they found out that our import was broken so so that's great but it all got fixed really quickly thanks to justin so anytime you're launching something just let me know in the comments and i'll try to break it stress test it ask your ai agent to do the same thing sign up as a stranger with a brand new email email your own support address before you tell anyone about it and there's one that i caught just in time the week before launch my announcement was written and ready it said 90 days free and the website still said it was in beta but we weren't in beta anymore and it was supposed to be 60 days free and not 90 days but you know we fixed all that okay but what do we spend on the launch press releases so this is the first time i've ever done this you know those dumb hokey press releases that nobody reads i think they actually matter and i'm seeing evidence in only within the first week or so that they do so i spent 660 bucks across two wire services one was called ab newswire and the other was einpresswire.com i paid an extra 90 bucks to get on usa today's website now why well for seo or ai seo aeo geo whatever it is seo and ai seo there's two types of backlinks okay there's do follow links and no follow links do follow links basically pass through seo juice right to google and bing that says hey this guy's legit because this other website that's legit talks about him okay and a no follow link is a backlink a link on another website back to you that basically doesn't give you anything it doesn't do anything for you so when you pay for these press releases some of them will be do follow links and some of them will be do follow links and it gives you some seo juice and it tells chat gpt and claude hey you know this guy is something you know a couple people are talking about him you do get some do follow links but then you get a whole bunch of like syndicated no follow links as well which is mostly just noise doesn't really help you but for a product like this where the lifetime value of a customer is probably around a thousand bucks paying 660 bucks for a couple good backlinks seems to be worth it to me because if i can just get one customer maybe two then it more than pays for itself but it doesn't make sense for every business anytime you start any business that has a website one of the first things you do should be to add it to google search console okay search.google.com that's gonna basically allow you to see how your website is ranking for certain keywords and terms and whatnot and so here is my if you're watching google search console dashboard for lazy books you can see it's down because a week ago it was really high from the top so a lot of people were looking for it and now it's normalized but you can see these are the terms i'm ranking for ai accounting como exportar quickbooks excel right a spanish version of you know exporting quickbooks to excel from excel alternatives to quickbooks i'm ranking 29th on google for that so at the very bottom of the third page but hey it's something and we are ranking second for the phrase ai accounting so that's cool and how did i do that i'll get to that in a minute and then here you can see what pages are appearing what countries are seeing so it's kind of like google analytics but more focused on seo and showing you what you're ranking for pretty cool but after i paid for these press releases i can tell you that the number of websites linking to me went from about 80 to over 200 in a week which sounds awesome right but most of those again are just syndicated copies of the original press releases that i paid for and nobody is reading them and about 80 of those links are the no follow links that i paid for and i can tell you that it's not going to be easy for me to get to the top of this list because i'm not sure if i'm going to be able to get to the top of this list i'm going to be able to get to the top of this list i'm going to be able to get to the top of this list i'm going to be able to get to the top of this list i'm going to be able to get to the top of this list i'm still mostly invisible we did get listed on 15 different software directories in one day for free i used muse my ai agent to do that for me just a bunch of random directories that just list off random software tools like this it doesn't really hurt it's an asymmetric bet a lot of them wanted like 100 to 200 bucks but that backlink didn't give me any more seo juice than the free one so we didn't pay for any of those and i would advise you to do the same so take the free listing and pay only if the site sends real people so what now all right signups have crawled to about 10 a day we want some seo juice going we want people searching for us and we've got time you know seo won't pay off for months it's all right let's get it going right now i use grokbot with my data for seo plugin and i've got a link for that in the description grokbot went and looked for almost 2 000 different things and keywords that people are typing into google about bookkeeping software and it all of them to combine add up to about 640 000 searches a month and advertisers are paying 32 bucks a click for those 32 bucks for one click so they're very valuable searches so instead of paying for them on google ads we're using ai to write articles and we already have 126 different articles planned 82 are already written and we published three of them a week on like an agentic schedule automation and here's how i keep it from making stuff up because it's going to want to do that every article has to check every competitor's price against that competitor's own website the same day a lot of these are listicle articles about you know comparing prices and features and stuff and if one single fact can't be checked then the article just stays a draft and then i'll go in there manually and verify it myself on the 38 were right two are wrong so on one of them the ai wrote about a quickbooks plan that does even exist anymore. If I'd published that, that one accounting nerd who noticed would never trust me again. Every article also says in plain words, hey, I built this away what I'm saying with bias in mind. And I'm hoping that makes people trust it more. Two, the robots, right? The AIs. More and more people are asking Chad GPT and Claude what software to use where they used to ask Google. So the week I launched, we asked Chad GPT and perplexity 10 questions that a buyer would ask. Lazy books got named zero times out of 10. So what do you do? You just take the same sentence about yourself and repeat it on every page in every directory you can find. Lazy books cost $19 a month for up to three businesses. And the AI tools will repeat what they see written the same way in a lot of different places. And I won't know if it works for, you know, 90 days. Now getting people to actually use the thing. This is the one that matters most. Out of everyone who signed up, the number I can see who connected to the bank account is a little over 100,000. And I think that's a lot of people. I think that's a lot of people. I think that's a lot 150. So for the next six weeks, my job is to work on the people who have not connected to bank account card on file with no bank connected. Okay. And the fourth thing that comes next is for us to keep building what people are asking for payroll and job costing. They also asked for accrual books, which just went live today. We only did cash basis accounting, but now we offer accrual as well. And one lesson from my own tools this week was I had an automation that was supposed to copy every new customer over to my email list. It was turned on. It said enabled, but it had never run, not even once. So for about four hours, new customers were just falling on the floor until I asked my AI tool why the list wasn't growing. So just because you have something scheduled doesn't mean that it's actually running or working. You got to check on that. So on January 8th, we'll know how many of these hundreds are hopefully maybe by then over a thousand people with a card on file will actually stay. We could be at 8,000 a month. We could be 4,000. I don't know. We're going to test all kinds of different marketing concepts. Like we haven't even scratched the surface, but I'm just looking at it right now. It's a pretty slow way to make money. It'll grow. It'll compound over time because people don't switch bookkeeping software very often. But if you came to this episode for get rich quick, I don't know if software is going to do it. Okay. But the part I promised, so you probably don't have a newsletter or a big Twitter following. So how do you copy any of this? Well, start with five tests. We'll start with the cheapest and I'm going to talk about the price, the tool, and the first step for each one. So the test number one, the 10% rule, okay? Cost is zero. The tool you use is your phone. So there's 35 million small businesses, 33 million, whatever, and about 260 million adults. So about one in every 10 adults, every nine adults, whatever, you know, owns a business of some kind. Your uncle that has the lawn business, the girl from high school who does lashes, your dentist, barber, whatever. Open your phone contacts and start scrolling and you're going to find 20 in just a few minutes. Text them two questions. Okay. These are the same two questions that I use. Number one, what is the one thing about blank you couldn't live without, right? Insert software tool. And then what is the feature that you would happily skip? That's it. Okay. No pitch. Just ask. And people like being asked their opinion. These 20 answers will teach you more than any prompting to chat GPT will do. And this works brilliantly with software because if your idea is a cleaning company for offices or a bookkeeping service or, you know, washing trucks, your customer is a business owner and you already know a bunch of them. Test two, you're going to let AI read the complaints and it will cost zero to 20 bucks. You can use chat GPT or Claude, whatever. And here's the prompt, something like go read 200 recent reviews of X. Then you name the big company that you'd be up against. Give me the top 10 complaints and the top 10 things people can't live without. Then go type that same company name into Reddit and into a couple of Facebook groups and read what real people are mad about. And you want to see those one or two complaints that show up over and over. For me, it was the price that keeps going up and the bank connection keeps breaking. Test number three is something called PickFu and it'll cost you about a hundred bucks. This is a site with millions of people who get paid to answer polls. You pick who you want, like small business owners in the U S or female, small business owners between 35 and 45 that are moms, whatever. Ask your question. And it's about a buck an answer. So a hundred owners of your exact target market is a hundred bucks. You get that back in about an hour. Ooh. And actually that reminds me, my friend Brandon had the idea to use PickFu to find customers, right? So you're asking them a bunch of questions about your software, but you're also pitching your software in the questions and for a buck. And if your lifetime value is a thousand bucks, it's not bad. In fact, I'm going to open a tab right now. So I don't forget to do that tomorrow. Cause that is a for like an hour. And this is not an ad for PickFu. I've used it for years. All right. Test number four, a poll ad. This will cost 50 to a hundred bucks. You're going to need Facebook ads, but you don't even need to log into Facebook ads. Use Muse, Instinct, or Grokpot to log into Facebook ads for you. You just run an ad to business owners in your town. And all it does is ask a question. Quick poll for local business owners. What do you hate most about your bookkeeping? They'll tap it and answer two questions. You're paying for answers. You're paying to play. And every one of those you can go back to when the thing is ready to sell. And you can do polls in any number of different ways. You could just do it as like an engagement post where you pay per comment, bunch of different ways you can do it. Or you can just AB test, right? Like five infographics. Like I hate QuickBooks because the bank account keeps disconnecting or because the price keeps going up and whichever one like gets the most reaction, you could consider that to be the winner. So test number five, the card test. This is the big one. So this will cost you about 11 bucks and for a domain name and a couple of bucks for a website. But you can do this in chat, GPT, cloud, whatever. You don't need card for this anymore. Put up one page, say what it is, you say the price, just a simple landing page. And then you ask for something that costs them a little bit. An email is fine, but a credit card is strong. And the reason why, so I mean, this is not anything super groundbreaking. It's a landing page where you're selling your thing before you have the thing. If people actually pay you money before you're ready, you just refund it. Fine. It's a completely different thing to vote with your wallet than it is to vote with a couple clicks, right? They're putting their money where their mouth is and it means so much more. In a survey, people tell you what they want, right? But on launch week, people really tell you what they want because now comes time to pay for it. Polls tell you what to build, but cards, credit cards tell you if anybody even cares. And here's the thing with these tests. Let's say you run all five of them. You spend two or 300 bucks and you just get crickets, nothing. Congratulations. You either win or you learn. You win or you win. You're going to win no matter what, right? And a lot of times, if not most times, it's much better to learn that an idea would not work than to get a false positive that an idea would work. And then you go miles down the wrong road, only they have to turn around. Let's talk about the directories. There are dozens, if not hundreds in every category. Google Maps, your Chamber of Commerce, Yelp. Those are all directories. AI agents can clean up listing you on all these directories. You do not have to do it by yourself. You can do it by yourself anymore. Reddit is a directory of sorts. You've got to be human about it. I have an old Reddit account that has a lot of karma, but it didn't have any karma or reputation in the groups that mattered for this launch. So in certain subreddits, they will completely ban you if you even mention your product. In other subreddits, you can link to your product and they won't care. It just depends. But write a comparison page, right? Your product against the big product, right? QuickBooks in this case. Over time, that will rank. You got to be patient It's slower, but it works. So eight days in, 10 days in, whatever, almost 700 signups, 400 and something cards, $0 collected, 10 people a day, support email that balance. So we won't know until January if this thing is going to make any money. But I'm thrilled because 15 months ago, this was just an $800 domain name and a guy on camera, me saying QuickBooks sucks. But I'm going to come back in January and show you the real number, how many of those cards turned into 19 bucks a month. Even if it's ugly, we'll show it. So. Go ahead and subscribe if you want to hear about that. If you want to check out the product, it's at lazybooks.com. I think you'll love it. We're getting amazing feedback. Really, really good feedback. Good user interface, good features. You're going to love it. And it risks you nothing. So business is hard. Yes, but it's fun and it's totally doable. So we'll see you next time on The Kerner Office.