Lovable is no longer just a fast-growing AI coding startup. At a reported $500 million annualized revenue run rate, it is becoming a serious test of whether software buyers will build instead of buy.
Lovable has put a hard number behind the vibe-coding boom: more than $500 million in annualized recurring revenue, one million new projects a week, and more than 50 million projects created on its platform. For a Swedish company that only opened Lovable to the public in late 2024, that is not normal SaaS growth. It is a signal that the market may be changing faster than the incumbents expected.
According to Business Insider, Lovable said on June 9 that it had passed $500 million in ARR, up from $400 million earlier this year. That matters because the story is no longer only about hobbyists asking AI to make landing pages. Lovable says most of its users are non-technical, but many are building software they want to monetize or use inside real businesses.
That is where the bigger argument begins. If non-engineers can build websites, e-commerce storefronts, CRMs, inventory systems and HR tools without waiting for a traditional product or engineering cycle, the pressure does not only fall on developers. It falls on Figma, Webflow, no-code platforms and every SaaS vendor selling workflow software that a team might now try to replace.
Software companies have always sold one powerful promise: do not build this yourself. Buy the product, get the updates, avoid the maintenance, and let someone else carry the operational headache. That bargain has worked for decades because custom software was expensive, slow and usually required a technical team.
Lovable is attacking that bargain from the other side. Its first user report draws on anonymized activity from January 2025 to May 2026 and a survey of more than 14,300 users. More than half of respondents said they were building a business, while nearly a quarter said they were working on side projects they hoped to monetize. That is not the same as saying they will succeed. But it does show that users are not treating these projects as throwaway experiments. They are trying to turn prompts into products.
This is why the $500 million figure is more important than the one million weekly projects. Usage can be noisy. Revenue is harder to ignore. People may experiment for free, but businesses pay when a tool is saving time, opening a new revenue line, or replacing something they already spend money on.
There are already examples pointing in that direction. By late 2025, Lovable had highlighted enterprise use cases from companies including Deutsche Telekom, Uber and Zendesk, where teams were using the platform to speed up prototypes and internal workflows. Those examples matter because enterprise adoption is where vibe coding moves from interesting to dangerous for the old software stack.
Still, building is the easy part compared with keeping software alive. Dependencies change. Security issues appear. Integrations break. Users ask for more features. Many companies buy SaaS because they want someone else to be accountable when things fail. Lovable’s long-term challenge is not whether people can create apps quickly. It is whether those apps can be maintained safely enough for real business use.
Fast Revenue Also Raises Hard Questions
Lovable’s funding history shows why investors are paying attention. In December 2025, the company raised a $330 million Series B led by CapitalG and Menlo Ventures at a $6.6 billion valuation. That followed a $200 million Series A in July 2025 at a $1.8 billion valuation. The company had previously reported $100 million in ARR in July, doubled that by late 2025, and then climbed to $300 million and $400 million before its latest disclosure.
That kind of revenue velocity changes how the market looks at defensibility. If a company can grow this quickly, the opportunity is obviously large. But the same speed attracts rivals. Cursor has become one of the best-known AI coding companies for developers, while Bolt and Replit are chasing builders who want fast app creation without traditional setup. OpenAI and Anthropic also sit uncomfortably close to the category because many of these tools depend on frontier models that those companies control.
Lovable’s answer appears to be breadth. It is not only selling code generation. It is building hosting, databases, authentication, payments, integrations, collaboration and governance around the act of creating software. That matters because prompts alone are not a moat. A workflow that helps teams go from idea to deployed product, with security and maintenance built in, has a better chance of becoming infrastructure.
The company also has to prove trust can scale with revenue. In April, Lovable acknowledged that a backend permissions change had accidentally re-enabled access to chats on public projects. The company said it reverted the change and made public project chats private again. That response was necessary, but the episode also showed the risk of putting business software creation into the hands of millions of users who may not understand where visibility, data access and security boundaries begin.
This is the balance investors now have to judge. Lovable may be one of the clearest signs yet that AI app builders can create a new software market, not just decorate the existing one. But the next phase will be less forgiving. Revenue growth will still matter, but retention, project survival, security, enterprise governance and real production usage will matter more.
If Lovable can prove that millions of projects do not simply become millions of abandoned experiments, the company will have done more than build a fast-growing AI tool. It will have shown that the software market has a new buyer, the non-technical builder with a budget, a problem and no patience for the old queue.
Also read: Wayve could give London a private-market test that actually matters • OpenAI moves closer to an IPO as Wall Street waits for AI numbers • NinjaOne reaches $12.3 billion as AI pushes IT automation higher
