Lovable reaches $400 million ARR — with 146 employees

Lovable broke through the $400 million mark in annual recurring revenue (ARR) in February. In January that figure still stood at $300 million. A hundred million more in a single month, with — according to the company itself — 146 employees. CEO Anton Osika told Bloomberg TV the company is running "five months ahead of forecast".

A reminder of just how fast all of this is moving: the first $100 million ARR was reached in July 2025, $200 million in November, $300 million in January, and $400 million in February.

The numbers side by side

Moment	ARR
July 2025	$100 million
November 2025	$200 million
January 2026	$300 million
February 2026	$400 million

In December, the company raised $330 million at a $6.6 billion valuation, with CapitalG and Menlo Ventures leading and NVIDIA, Salesforce, Databricks and HubSpot joining in through their investment arms. Lovable itself is keeping the door open to $1 billion ARR later this year, but isn't publicly committing to it.

$2.7 million in revenue per employee

The figure that stands out most to us isn't the $400 million, but the division: roughly $2.7 million in revenue per employee. At a software company known for its efficiency, such as Salesforce, that figure sits around $500,000.

That ratio is exactly where the broader trend becomes visible. A product team that uses AI to build needs fewer people to deliver the same output. That applies not only to the company making the tool, but also to the companies using it. We see it in our own projects: a customer portal or internal dashboard that used to be a months-long project is now a matter of weeks.

What this means for SMEs and mid-sized businesses

Three practical consequences we're already seeing come up in conversations.

1. Internal tools are no longer a luxury

Many businesses work with spreadsheets and separate folders because custom software was too expensive. That calculation has changed. A lead overview, a planning tool, a quote generator: it now costs a fraction of what it used to. We build them the same way ourselves.

2. Demand is shifting towards direction, not hands on deck

When building goes faster, the mistakes you make become visible faster too — and get repeated faster. The value lies in the choices made beforehand: who is this for, what should someone be able to do, how do you measure whether it works. Building is the easy part after that.

3. Quality gaps are getting bigger, not smaller

We look at websites every week that were thrown together quickly and where the basics are wrong: no thought-out page structure, duplicate or missing titles, no schema, copy that doesn't do anything. The tool doesn't determine whether something is good — the brief and the attention paid to it do.

Our approach at this pace

We use AI tooling as standard in our projects, but with a fixed checklist that doesn't change:

• Structure first. One H1, logical headings, clear URLs and internal links.
• Measure speed, don't assume it. We check core speed and stability metrics on every page.
• Accessibility. Contrast, focus states, alt text and keyboard navigation are part of delivery, not a later sprint.
• Findability and content. Search intent determines what goes on a page, even if the page goes live in a day.
• Maintainable. Can someone else work on it in a year? If not, it's not a finished delivery.

The bigger picture

If a company with 146 people is running $400 million in recurring revenue, that's above all a signal about the direction software development is heading: less time spent building, more time spent on what you're building. For business owners, that's good news — as long as you don't skip the basics.

Want to know which piece of software has suddenly become achievable for your business? Book a strategy session — we're happy to think it through with you.