Open source and venture capital : how free software builds billion-dollar ventures

· 4 min read · by Adrien Lopez

Can you build a profitable company on software you give away for free? From Red Hat to Mistral AI, here is how open source ventures create value, which business models work, how venture capital invests in them, and what licence changes teach founders.

Open source ventures

"Why would anyone invest in a company that gives its product away?" For a long time, this was the main objection to open source startups. Today, open source runs the internet, the cloud and AI, and commercial open source software (COSS) companies attract billions in venture capital. Here is how it works.

Free software and open source: a quick reminder

  • Free software, as defined by the Free Software Foundation, guarantees four freedoms: to run the program, to study and modify it, to redistribute copies, and to distribute modified versions.
  • Open source, as defined by the Open Source Initiative, focuses on the same licences from a more pragmatic, development-oriented angle.

Licences fall into two families: permissive (MIT, Apache 2.0, BSD), which allow almost any reuse, and copyleft (GPL, AGPL), which require derived works to stay free. The choice of licence is a strategic business decision.

Why open source wins

  • Distribution: developers can try the product in minutes, without a sales call. Adoption is bottom-up.
  • Trust: the code can be audited, which matters for security and digital sovereignty.
  • Community: contributors improve the product, report bugs and create integrations.
  • Talent: engineers want to work on public code that builds their reputation.
  • Standards: open projects tend to become the default (Linux, Kubernetes, PostgreSQL, React).

Business models that work

  1. Support and subscriptions. The historic model of Red Hat, acquired by IBM in 2019 for about 34 billion dollars: the software is free, the enterprise subscription (support, certifications, long-term maintenance) is paid.
  2. Open core. A free community edition and paid enterprise features (SSO, audit, advanced security). Used by GitLab, Mattermost and many others.
  3. Managed cloud / SaaS. The company hosts its own software as a service: MongoDB Atlas, Elastic Cloud, Supabase, Vercel for Next.js. Customers pay for convenience and reliability.
  4. Dual licensing. Free under a copyleft licence, paid licence for companies that want to embed it in proprietary products.
  5. Services and integration. Consulting, training and custom development around an ecosystem: the model of many European integrators, and of Nextcloud, which sells enterprise subscriptions to governments and companies.
  6. Open models, paid platform. In AI, Mistral AI publishes open-weight models while selling API access, enterprise deployments and its assistant, Le Chat.

How venture capital looks at open source

VCs invest in COSS companies because open source adoption is a powerful leading indicator: GitHub stars, downloads, contributors and community activity show product-market fit before revenue. Specialised funds and indexes track the fastest-growing open source startups.

What investors look for:

  • a project with real adoption, ideally led by its original creators;
  • a clear monetisation path that does not conflict with the community;
  • a cloud or enterprise offer with strong gross margins;
  • a defensible position: brand, expertise, ecosystem, not just the code.

The licence change lesson

Some companies, facing cloud providers reselling their software, moved from open source to "source-available" licences: Elastic (2021), HashiCorp with the Business Source License (2023), Redis (2024). The community answered with forks: OpenSearch, OpenTofu, Valkey, the latter two hosted by the Linux Foundation. Elastic later re-added an open source licence (AGPL) in 2024, and Redis did the same in 2025.

Lessons for founders:

  • choose your licence from day one with your business model in mind (AGPL protects better against cloud resale than MIT);
  • your community is your moat: breaking its trust can create a competitor overnight;
  • governance matters: a foundation can reassure large contributors and customers.

Open source, AI and Europe

AI has brought the open versus closed debate back to the centre. Open-weight models allow companies to run AI on sovereign infrastructure, fine-tune it on their data and avoid dependency on a single provider. For Europe, which has few hyperscalers, open source is a lever for technological sovereignty and for building global ventures, from Mistral AI to Nextcloud, and from Hugging Face (founded by French entrepreneurs) to countless startups.

Read more in our guides on sovereign cloud and AI assistants compared.

Advice for founders

  • Start with a real problem developers have, and solve it in the open.
  • Measure community metrics but focus on active users in companies, the future customers.
  • Make the paid offer obvious: hosting, security, compliance, support.
  • Raise when adoption is proven: open source traction is a strong signal for seed and Series A investors.
  • Stay humble with the community: transparency about the roadmap and the business model builds lasting trust.

Resources

AD Ventures supports founders and companies building on open source and web technologies. Let's talk.

Tags

← Back to the blog