The archive · Developer & Business Tools · Financial decision · 2012
Meteor's 2012 bet: $11.2M to make a JavaScript framework its only product
Meteor raised $11.2M from Andreessen Horowitz (July 2012) to build its open-source JavaScript framework as the only product — no consulting, no anchor app.
Meteor
What the business is
Meteor was building an open-source, full-stack JavaScript web framework for dynamic client-server apps — one language across browser and server, with reactive templates and optional packages — and planned to earn later from services that deploy and manage apps built on Meteor.
How it started
Meteor's founding team, which included the creator of EtherPad, set out to build a full-stack JavaScript web framework as a product in its own right, not as the by-product of an app. The first public Meteor release shipped in April 2012; the team ran meteor.com on Meteor and built personal projects with it, and commenters later noted the company had been a stealth participant in Y Combinator's S11 cycle.
What happened
On 2012-07-25 Meteor announced an $11.2 million round from Andreessen Horowitz, framing the money as a development budget: the core team would focus on the framework for years, with no consulting work and no separate commercial app built on top of it. The Hacker News thread turned that promise into the central debate — whether a framework built for its own sake can succeed, since Rails and Django grew out of real products. Founders answered that every feature had to be implemented by hand several times in apps the team was building before being packaged as an optional Smart Package, and that meteor.com and their side projects ran on Meteor. Marc Andreessen joined the thread to explain a16z's reasoning: Meteor needs to exist as a great way to build modern client-server web apps, the market is large, and the bet echoed his own JavaScript history at Netscape.
No ending yet — it is still running.
Background
Meteor was a company whose product was a developer tool: an open-source, full-stack JavaScript web framework meant to let one team build modern, dynamic client-server web apps in a single language. The first public release shipped in April 2012, meteor.com itself ran on Meteor, and features were added only after the team had built them by hand several times in real applications, then packaged as optional Smart Packages.
On 2012-07-25 the company announced an $11.2 million round from Andreessen Horowitz and called the money a development budget: the core team would focus entirely on Meteor for years, without consulting or building another product to sell on top of the framework. Skeptics on Hacker News argued that successful frameworks such as Rails and Django were born as the side products of real applications, and that a framework built for its own sake tends to over-engineer.
Founders pushed back in the thread: the team ran its own site and tools on Meteor, every feature had to be implemented by hand in real apps before it was packaged, and they counted on the community to call them out if the money made them lazy. Marc Andreessen explained that a16z wanted a great way to build modern client-server web apps to exist, saw a large developer-tools market, and felt a personal echo of his JavaScript days at Netscape. The announcement thread drew 249 points and 107 comments on 2012-07-25.
What has to be true
- The raise was itself the bet: $11.2M of runway to build an open-source framework as the company's only product, with no consulting and no flagship app to fall back on.
- The founders' packaging rule answered the classic objection: add a framework feature only after implementing it by hand in several real apps.
- Andreessen Horowitz wanted the platform to exist: Marc Andreessen argued Meteor was a real breakthrough, the market was large, and the investment echoed JavaScript's origins at Netscape.
- The unproven question was adoption: whether developers would rally around a framework nobody was forced to use, and whether that community could later become revenue.
What can be applied
Raise to make the framework the product, but keep the dogfood rule: Meteor's defense against 'built for its own sake' was that nothing shipped until it had been built by hand in real apps.
Aftermath
As of 2012-07-25 Meteor had just announced the round and was live: an open-source framework, its own site running on it, a 107-comment Hacker News thread, and deployment services for Meteor apps planned. What came after the announcement — adoption, later rounds and the company's eventual direction — is beyond this batch's sources.
Sources
spotted an error? The archive wants to know.
Your turn
You just read one. Describe what you are building, and see who is betting on the same thing.
Free account · 3 free questions · no card