I've discovered a new wave of Luddites. This time it's not weavers with sledgehammers, nor programmers whining about neural networks taking their jobs. It's product managers.
Their argument is singular, repeated like a mantra: "If vibecoding actually worked, we'd see a massive surge of new, successful commercial IT projects across every sector. But we don't. Instead, every founder is just tinkering with something in their own corner." The conclusion they draw: vibecoding is a dud, a marketing bubble, a toy.
And here is where you want to grab them by the collar and shake them gently.
You're weighing the new harvest on an old scale
The absence of a flood of "vibecoded" commercial products is not proof that the tool doesn't work. It's proof that you haven't understood the world you're standing in.
The software market hasn't changed cosmetically. It has changed fundamentally — at the level of economics. That old model, where someone takes a technology, slaps a new product on it, pushes it "to market," and makes millions — that model simply doesn't exist anymore. It didn't get "harder." It's gone.
Yet the product manager keeps applying the old ruler to the new reality. He waits for a "rise in the number of successful small businesses," fails to find it in the old coordinates, and gleefully declares: it didn't take off. He's measuring the yield of a new field with the weighing scale that hung in the shop for twenty years. And pretends the field is to blame.
It's one thing when a burnt-out programmer talks like this — at least his skin is in the game, he's genuinely scared for his salary. But a product manager, whose job is to be the first to read where value has shifted and where the money is now? A product manager who still hasn't grasped that everything has shifted? That's not nearsightedness. That's professional disqualification.
Why no sane person builds a product "for the market"
Let's break down the economics, since everything rests on them.
There are no vibecoded commercial products not because they can't be built. There are none because no one in their right mind would build them "for the market." Releasing a new commercial product is now pointless — because any high schooler can revibecode it for free tomorrow evening.
And here is the main thing that doesn't fit in the Luddite-product-manager's head: it doesn't matter WHAT or HOW you assembled this product.
Whether it was hand-crafted by Himalayan virgin programmers or Claude for twenty bucks an evening — no difference. Whether you brought in gray-bearded architects and team leads with thirty years of experience to design it, or threw it together with a single prompt on your knee — no difference. Your expertise, your heroic backlog, your six months of development no longer form a protective moat around the product.
Absolutely no difference.
The barrier to entry — time, money, rare engineers — used to be the product itself. You were selling not so much the feature as the fact that replicating it was expensive. Now replicating it is cheap. Effectively, free. Which means what you were going to charge money for "on the market" slips through your fingers the moment you ship it. Product-for-market as a business model didn't break partially — it zeroed out.
That's why founders are "tinkering with something in their own corner." It's not a sign that the technology doesn't work. It's exactly the behavior the new economy encourages. They aren't idiots. The idiot is the one who expects the old ritual of "ship to market → collect millions" from them and, failing to see it, declares the technology dead.
Where vibecoding actually makes money
If the value is no longer in selling software outward — where did it go? Inside.
Vibecoding is a tool for application INSIDE an organization. And it doesn't bring market revenue; it removes cost from yourself.
The first is payroll. A development department that used to require ten people is now covered by fewer. This isn't abstract "efficiency" from a slide deck — these are specific lines in the payroll that are no longer mandatory.
The second is licensing fees, and here the Russian eye is unaccustomed. We're used to pirating everything or making do, but in the EU and the Emirates, an average company legally pays for software like an adult: dozens of products, a license for EVERY workstation, per-seat, every month. Multiply by headcount. An internal tool, vibecoded for your specific task and replacing the function of a paid external product, cuts that subscription out of the budget entirely.
This is where the technology already works and brings money. Not as a storefront on a marketplace, but as a deduction from your own costs. It's just that this money doesn't show up in the "successful commercial launches" statistics the product manager is stuck on. He's looking in the wrong place, finds nothing, and concludes there's nothing to see.
The competence test
So I'll answer the original thesis directly. "We'd see a rise in the number of successful small businesses"?
You won't. And you never will — not because vibecoding doesn't work, but because you're looking for its effect where, by the new economics, it cannot be. Value moved from the external market to inside the company perimeter, and public launch statistics simply don't catch it.
The ability to see this shift — where value now arises and why the old product-for-market model is dead — is, frankly, a baseline competence for a product manager. It's literally what they get paid for.
So you can use this as a fitness test for your product manager. Give them the original argument about "we'd see a rise in small businesses" and watch what they answer. If they nod and agree that the absence of a launch flood means the technology is a gimmick, you know exactly who in your team needs replacing.