The archive · Developer & Business Tools · Product decision · 2020–2026
Obsidian: a 7-person, zero-VC note app that hit ~$25M ARR by selling local Markdown files
Two Waterloo founders built a notes app that stores everything in plain Markdown on users' drives; without investors it reached ~$25M ARR with seven people.
Obsidian
What the business is
Obsidian is a free, local-first Markdown note-taking app with paid add-ons (Sync, Publish, Catalyst) and a plugin ecosystem; revenue comes from optional services and licenses.
Starting capital：No external funding: self-funded by the founders and profitable from day one (company statements cited 2026).
How it started
In March 2020, University of Waterloo alumni Erica Xu and Shida Li — who had already built the outliner Dynalist — were stuck in quarantine and unhappy with every note app they tried. They wanted speed, offline access, and data that would outlive any company, so they wrote the first beta of Obsidian (first code: January 31, 2020).
What happened
The app grew through word of mouth and a plugin ecosystem; super-user Steph Ango joined as CEO. Paid Sync and Publish services funded everything. In August 2023 the team publicly committed to three rules: never grow past 10–12 people, never take VC, and never collect user data.
How it ended up
Still running and scaling as of September 2026 with a team of about seven to nine full-time people (plus a cat), millions of users, and an estimated ~$25M ARR.
Background
Obsidian began in March 2020 when Erica Xu and Shida Li, Waterloo alumni who had already built the outliner Dynalist, were stuck in quarantine and disappointed by every notes app on the market. They wanted speed, offline use, and files that would outlive any company, so they built it themselves: a local-first Markdown app where all notes live on the user's own drive as plain text.
The app spread through word of mouth and a community plugin system modeled on VS Code. Super-user Steph Ango was invited to join as CEO, and paid add-ons — Sync, Publish, Catalyst — funded everything from the start. In August 2023 the company publicly committed to three rules: never exceed 10–12 people, never accept venture capital, and never collect user data.
The bet on refusing lock-in held: by 2026 Obsidian had over 4 million users including 10,000+ organizations, an estimated ~$25M ARR (third-party estimate, not officially disclosed), a reported ~$350M valuation, and still only about seven to nine full-time employees. It deliberately stayed out of the AI feature race, leaving AI to community plugins, and kept selling control of one's own notes instead of a captive cloud.
What has to be true
- Local plain-text files meant users could leave at any moment, so the product had to earn retention every day — and the company's principles built unusual trust.
- A plugin ecosystem turned thousands of users into an unpaid engineering corps, letting a team of three engineers serve millions.
- No investors meant no growth-for-growth's-sake: features stayed aligned with paying users, not an exit cycle.
- Charging for convenience (Sync, Publish) rather than the core app made the free tier a genuine funnel instead of a hostage situation.
What can be applied
Data ownership can be the product: when users can leave anytime, trust becomes the moat, and a tiny team out-earns venture-backed rivals by charging for convenience, not captivity.
Aftermath
As of September 2026 Obsidian remains independent, profitable, and free of VC: roughly seven to nine full-time people including three engineers, millions of users and 10,000+ organizations, an estimated ~$25M ARR, and a third-party-reported ~$350M valuation. It has deliberately avoided an AI pivot, leaving AI integrations to community plugins, and is hiring a fourth engineer while holding to its 10–12 person cap.
Sources
- Obsidian: $25M ARR, 8 People 1 Cat No Funding, Not Chasing AI
- Three Engineers, Zero Financing, No Meetings: Obsidian, a $350M-Valued "Small but Beautiful" Company
- Small teams got dangerous
- Obsidian Blog
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