Building was the easy part
You spent a month building the SaaS you had in mind. Hours, days, weeks, nights. You finally have something that works. You tell yourself the hard part is done, and selling it is now just a formality.
The hard part starts now.
Building a product and selling it are two different jobs. You just did the first one with AI, and you did it fast. The second one cannot be vibe coded. Nobody is going to generate your first customers while you sleep.
The good news is that selling your first 10 users has nothing to do with selling your next 10,000. You do not need growth hacking, funnels, or an ad budget. You need to talk to humans. And that can be learned.
The "I launch and wait" trap
Here is the classic scenario. You finish your app. You add a nice "Sign up" button. You post on LinkedIn or X. You wait.
Nothing happens.
Three likes, one comment from a friend, zero signups. You start thinking nobody cares about your product. That is usually wrong. The problem is not necessarily the product. The problem is that you confused "publishing" with "selling".
Publishing is passive. You put something into the void and hope. Selling is active. You find one specific person, with one specific problem, and you show them that your app solves it. In the beginning, that happens one person at a time. Manually. It is slow, uncomfortable, and exactly how everyone starts.
Find your first 10, not your first 1,000
The beginner mistake is to go broad. "My app can help everyone." No. An app for everyone is an app for nobody.
Think about the most specific person possible. Not "freelancers". More like "the freelance graphic designer who struggles to follow up on unpaid invoices." The more specific you are, the easier it is to find them and talk to them.
Then go where they already are. Not into the void of your feed. Into real places.
The communities they already hang out in. Discord servers, Slack groups, subreddits, niche Facebook groups. You do not show up to spam your link. You show up to help. You answer questions, give advice, and when someone describes the exact problem your app solves, you mention it.
Your direct network. You probably know three people who have the problem you solve. Write to them. Not a generic message. A real message, for them, that says "I built something for this, do you want to try it?"
Cold outreach, done properly. Ten personalized messages are worth more than a thousand copy-pasted ones. You take the time to understand the person before writing. It takes five minutes per message. That is normal.
Your goal at the beginning is not ten sales. It is ten conversations. The sales come after.
For the record, I found my very first customer by doing physical outreach. His name was Greg and he ran a Mexican food truck. I talked with him for half an hour, and he eventually signed up for my app to "help me out".
That gave me the leverage to go to the next people and say: look, this food truck signed up, so why not you?
Talk before coding one more feature
Your vibe coder reflex is to answer every objection with code. "Oh, that is missing? I will add it tonight." Stop right there.
Before adding anything, talk. Speak with ten people in your target audience on the phone, on a video call, or even just in DMs. Check whether the need is real.
You are not selling during these calls. You are listening. You are trying to understand:
- How they solve the problem today, without you.
- What really bothers them about their current solution.
- Whether they would pay to make that problem disappear.
Those ten conversations are worth more than ten new features. They tell you what to sell, which words to use, and what price to start with. Often, you discover that the feature you were most proud of is used by nobody. And the small thing you rushed in two minutes is what people actually care about.
It is the same logic as the product itself: building fast does not mean building the right thing. The market decides, not your intuition.
On the website for my restaurant SaaS, I had built a very complete MCP server to manage a restaurant with AI. I spent a lot of time on it. It brought me zero customers.
On the other hand, by listening to customers, reacting quickly, and shipping two or three small improvements made for them, they recommended me and brought in others.
Charge from day one
The biggest mistake I see among vibe coders is: "I will make it free to get users, and monetize later."
Later never comes. And worse, free lies to you.
I know because I made that mistake. 60 customers in two months, and not one of them truly committed. Some even refused to work with me: they preferred to pay for something that felt serious. Free had not only deprived me of revenue, it had destroyed my perceived value.
A hundred people signing up for a free app proves nothing. People say yes to anything that costs nothing. The only proof that your product has value is someone taking out a card. One customer at EUR 20 teaches you more than a hundred free signups.
You do not need perfect pricing. You need a price. Set one, even if it is approximate. You will adjust it. A good starting point:
- A simple price, one or two tiers maximum. No five-column pricing table.
- Better slightly too expensive than too cheap. It is easier to lower a price than raise one.
- A free trial if you want one, but one that eventually asks for a card.
And when you add a payment button, real money and user trust enter the picture. That is precisely where tech problems can hurt you. I wrote about how an AI-generated Stripe checkout can leak data or fail silently. Before charging people, make sure you really charge them, and that you do not charge them twice.
The thing nobody tells you: selling breaks your app
As long as your app only runs with you alone, everything is fine. You click where you are supposed to click, in the order you are supposed to click. You are the only user, and a very polite one.
Your first real customers are not polite. They click everywhere. They put emojis in fields that expected numbers. They sign up twice, close the tab in the middle of payment, and use a password your app never saw coming.
Selling means inviting chaos into a house you built for one calm person.
That is when the cracks appear. Auth that lets an account through without a verified email. The app that slows down as soon as ten people use it at the same time. The form that loses data if someone clicks "Submit" twice. None of this showed up when you were alone. All of it shows up the day you sell.
That is not a reason not to sell. It is a reason to know where you stand before selling. The nuance matters. You can launch with cracks, as long as you know which ones exist, and which ones may cost you a customer.
The checklist before adding the payment button
You do not need everything to be perfect. You need the essentials to hold. Before sending your link to a first paying customer, check these points.
Signup and login really work. Not just for you. Test with an address you have never used. Test "forgot password". Test what happens when someone signs up twice.
Payment charges once, and only once. Run a real payment from end to end. Check that the money arrives, that the customer gets a receipt, and that a double click does not charge them twice.
Users cannot see each other's data. This is the most common and most serious bug in vibe-coded apps. Create two accounts, log in with one, and make sure it sees nothing from the other.
You know when it breaks. If your app goes down on a Sunday at 11 p.m., do you find out, or does your customer tell you in an angry email on Monday? Put at least one alert in place.
You can answer a customer in under an hour. An email address, a chat, a phone number. At the beginning, support is you. And a customer who struggles but feels that you are responsive is a customer you can keep.
If you can check these five boxes, you can sell with a clear head. If you are not sure you can check even one, you will find out with your first real customer, and that is the worst possible moment to discover it. I made a more complete checklist for going from prototype to product if you want to go deeper.
The rule to remember
You do not sell by publishing. You sell by talking to one person at a time, and making sure the product holds when they arrive.
Vibe coding gave you something rare: a finished product, quickly, without a team. That is a real head start. But you lose that head start the day a customer pays, clicks, and lands on a crack you did not know was there.
Sell. Really sell, now. But look under the hood before inviting people in.
Want to know whether your app can hold before selling it?
If you are about to put your SaaS in the hands of real customers, you should know first what is solid, what is risky, and what needs fixing first. Book an audit. In 3 days, you know exactly where you stand and what to do next, before a customer tells you the hard way.
And if you want concrete advice for building and selling with AI without getting burned, sign up for the newsletter. No spam, just what you need to know.
