Entrepreneurship KB

Powered by Agentic Workers

Case Study: Notion — From Kyoto Noodles to $600M ARR

11 min read

Case Study: Notion — From Kyoto Noodles to $600M ARR

Notion is the case study for patience, craft, and platform thinking. Founded in 2013 by an immigrant founder from China, the company nearly died in 2015 — Ivan Zhao laid off every employee, borrowed $150K from his mother, and moved to Kyoto, Japan to rewrite the entire product from scratch while living on instant noodles. The product was rebuilt four times before finding product-market fit. Today: $600M ARR, 100M+ users, $11B valuation. The journey from $3M ARR to $600M took 6 years — but the journey from founding to $3M took 6 years first. Notion is proof that some of the biggest outcomes require the longest setup times.

Timeline

YearEventScale
2013Notion Labs founded in SF by Ivan Zhao + Simon Last$2M seed, 4-6 people
2013-2015First product (no-code app builder) failsBurned through seed
2015Near-death: lays off everyone, borrows $150K from mother2 people
2015-2016Kyoto rebuild: complete rewrite on React2 people, instant noodles
Mar 2016Notion 1.0 launches — #1 Product HuntDocs + wikis + templates
Mar 2018Notion 2.0: databases — the breakthrough~8 employees
Aug 2018Akshay Kothari joins as COO (10th employee)Profitable
Sep 20191M users~27 employees, $3M ARR
Apr 2019$10M Series A at $800M (267x revenue)27 people
Jan 2020$50M Series B at $2B (Index Ventures)~50 employees, $13M ARR
2020COVID quadruples user baseInfrastructure nearly collapses
Oct 2021$275M Series C at $10B (Coatue + Sequoia)20M users, $31M ARR
Feb 2023Notion AI launches ($10/member/month)AI add-on
Jun 2024Notion Sites, Charts, LayoutsPlatform expansion
Sep 2025Notion 3.0: AI Agents + MCP$500M ARR, ~1,000 employees
Dec 2025$600M ARR, $11B valuation100M+ users, 4M+ paying

Total raised: ~$343M. Revenue grew 19x in 4 years (2021-2025) while valuation moved only ~10%.

The Founders

  • Ivan Zhao (CEO) — Born in Urumqi, China (Xinjiang). Learned to code via Math/CS Olympiad. Studied traditional Chinese watercolor painting. Mother moved them to Vancouver for high school — he learned English from SpongeBob SquarePants. UBC for cognitive science and art. Product designer at Inkling in SF. Read Douglas Engelbart’s 1962 “Augmenting Human Intellect” and decided to devote his career to tools for thought. “If you study physics, you know who Newton is. But in tech? Most people don’t know who Engelbart is.”
  • Simon Last (co-founder) — CS student at University of Maryland. Zhao found his portfolio on Hacker News, offered an internship, and convinced him to drop out. “His dad’s a school principal. We shook hands, [I had to show] I am not a crazy person.”
  • Akshay Kothari (COO, joined 2018) — Co-founder of Pulse (news app sold to LinkedIn for ~$90M in 2013). Early angel investor in Notion. Joined as 10th employee to professionalize operations.

Mapping to Frameworks

product-market-fit: Four Rebuilds to Get There

Notion’s PMF journey is unusually long — 6 years from founding to $3M ARR. The product was rebuilt four times:

  1. V1 prototype (2013): Built while Zhao was at Inkling. Concept demo.
  2. First real product (2013-2015): No-code app builder on Google Web Components. Unstable, crashed constantly, caused data loss. Users were confused. “People don’t want to build apps.”
  3. Kyoto rewrite (2015-2016): Ground-up rebuild on React. Became Notion 1.0 — docs, wikis, to-do lists, 30+ templates. Simplified the vision to familiar metaphors.
  4. Notion 2.0 (March 2018): Added databases, Kanban boards, tables, calendar views. This was the real breakthrough.

The critical pivot: instead of making people build apps (which nobody wanted), Zhao wrapped the same LEGO-like power in familiar productivity metaphors — notes, docs, wikis — and let users discover the depth gradually. He called this “sugar-coating the broccoli.”

David Pierce’s WSJ review of 2.0 — “The Only App You Need” — was the inflection point.

do-things-that-dont-scale: The Kyoto Survival Story

This is the most extreme “do things that don’t scale” story in the wiki. When the first product failed:

  • Laid off every employee (4 people beyond cofounders)
  • Borrowed $150K from his mother
  • Moved to Kyoto, Japan — cheaper than SF, chosen partly on a whim
  • Rented a two-story house with only a shoji screen between bedrooms
  • Coded ~18 hours/day for roughly a year
  • Lived on instant noodles, rarely went outside, rarely dressed

“Neither of us spoke Japanese and nobody there spoke English, so all we did was code in our underwear all day.”

“A brand-new code base is always happy. There’s no legacy code.”

This is the right kind of stubborn — Zhao was attached to the goal (tools for thought that empower everyone) but willing to completely abandon the approach (no-code app builder) and start over.

product-led-growth: The All-in-One Workspace Playbook

Notion’s PLG engine has several distinctive mechanics:

The blocks architecture: Everything is a block — text, images, database rows, even pages. 500+ block types across 6 categories. Converting between types preserves data. This makes Notion infinitely composable: “When you add a new LEGO brick, it might pollute the whole thing. So you have to think system-first.” The architecture is the moat — competitors can copy features but not the underlying system.

Generous free tier: Individual users get essentially full functionality for free, creating a massive top-of-funnel. Notion grew to 1M users before raising a Series A.

Bottom-up enterprise adoption: The individual-to-company customer ratio shifted from 90:10 to 50:50 as individuals brought Notion into their workplaces. No enterprise sales team was needed for the first wave — people just started using it.

Templates as distribution: Public templates are both content marketing and product demonstration. Every shared Notion page is an advertisement. The template marketplace creates an ecosystem of creators who market Notion for free.

Public pages as viral loop: Every page published publicly includes Notion branding. Notion Sites (June 2024) formalized this — publish any page as a website with custom domain.

focus: Design as Philosophy, Not Decoration

Ivan Zhao’s design philosophy is the most articulated of any founder in the wiki:

  • “Beauty isn’t decoration. It’s fundamental to how the tool works.”
  • “The goal isn’t to add things. It’s to remove everything that shouldn’t be there until what’s left is inevitable.” — inspired by the Japanese concept of “ma” (negative space)
  • “If someone’s drunk, can they understand it? You care about your design. Your pixels. Your product. But nobody else does.”
  • “Designers spend too much time on edge cases. Most of the time, what matters is the dumbest path. Make that great.”
  • On constraints: “We don’t allow green furniture [in the office]. You can’t control where people move things around. But if you cut green from the palette entirely, everything still looks okay.”

This echoes Linear’s craft-as-moat approach but adds a deeper cultural dimension — Zhao draws explicitly from Japanese minimalism, Bauhaus design, and architecture.

competitive-strategy: “Our Competitor Is the Entire Industry”

Notion’s competitive positioning is unique: they don’t compete against any single tool. “If you’re building Legos, are you competing with a toy airplane or a toy car company? It’s both.”

On copycats: “Imitation validates that we found something real. But copying features is easy. Copying the underlying philosophy — the commitment to craft, the willingness to rebuild from scratch when something isn’t right, the patience to wait for organic growth — that’s hard.”

The all-in-one strategy creates a different competitive dynamic:

  • vs. Confluence: Notion wins on UX/design; Confluence wins on enterprise compliance (HIPAA, FedRAMP) and Jira integration. 75% of Fortune 500 use Atlassian vs 35% for Notion.
  • vs. Coda/Airtable: Coda has stronger automation; Airtable stronger databases. Notion wins on the full bundle.
  • vs. Google Docs: Notion offers structure (databases, wikis) that Docs can’t match. Google has broader enterprise penetration.

ai-era-entrepreneurship: AI as Enterprise Accelerator

Notion AI (launched February 2023) followed a disciplined rollout:

  1. $10/member/month add-on — testing willingness to pay
  2. Initially 10-20% adoption → crossed 50% by mid-2025
  3. May 2025: Bundled into Business/Enterprise tiers — forcing enterprise upsell
  4. September 2025: Notion 3.0 with autonomous AI agents and MCP integrations

AI’s strategic impact: it pulled Notion upmarket. Enterprise buyers who would never have paid for “another docs tool” paid for AI-powered knowledge management. Over 50% of ARR now comes from AI-enabled customers.

leverage: 27 People, $800M Valuation

The team efficiency numbers are extraordinary:

PeriodEmployeesUsersARRARR/Employee
2018~8GrowingProfitable—
2019~271M$3M$111K
2020~504M+$13M$260K
2025~1,000100M+$600M$600K

27 people serving 1 million users at an $800M valuation. This is code leverage — the blocks architecture meant a tiny team could serve massive scale because the product was essentially a platform where users built their own solutions.

distribution: 95% Organic, Zero Paid Ads

Notion achieves 95% organic traffic and 82% organic download penetration — meaning most installs happen without paid advertising. The distribution engine has three layers:

1. Programmatic SEO via templates. The Notion Marketplace has 30,000+ templates, 99% user-generated. Each template has its own SEO-optimized page targeting niche keywords (“product requirements document template,” “habit tracker template”). Templates generate ~70% of Notion’s ~7M monthly organic visits. This is Zapier-style programmatic SEO powered entirely by community content.

2. Community-led growth. 300+ ambassadors across 50+ languages. Reddit: 280,000+ members (10x Figma’s). TikTok: 1 billion Notion views. Community events happen “almost every single day” globally. Head of Community Ben Lang: “Many brands pay influencers to read scripts… when we tried that, ROI wasn’t there. Creators must use Notion first.”

3. Creator economy as distribution. 2,000+ template creators. Thomas Frank’s “Ultimate Brain” template: 15,093 sales at $129 = ~$1.9M revenue from a single template. Creator “Easlo” crossed $500K+. Templates sold on Gumroad, Etsy, YouTube channels — all driving new signups back to Notion. Every creator is an unpaid salesperson.

The three viral loops: collaboration invites (each shared page creates a potential user), template duplication (one-click copy creates accounts), and public pages (every published page is a Notion ad).

international-expansion: 80% Non-US

Notion’s user base is 80% international — one of the highest ratios for a US-headquartered SaaS company. Particularly strong in:

  • Japan: DAU increased 500% in the year before Oct 2021. Localized Oct 2021. ~11% of user base. Data residency added May 2026.
  • South Korea: Mobile MAUs increased 8x (Aug 2020 → Aug 2024). ~12% of user base. Fans created unofficial Notion “textbooks.” Data residency added May 2026.

Strategy: localization + regional pricing + community-first (find passionate local users, give them ambassador status, let them build organic communities). No paid marketing in these markets.

Enterprise: Bottom-Up to Top-Down

50% of Fortune 500 now have teams using Notion. The adoption path: individual contributors discover it personally → bring it to their team → spreads to department → enterprise deal. Zhao: “We focused too much on what we wanted to bring to the world. We needed to pay attention to what the world wanted from us.”

Enterprise features added over time: SSO/SAML, SOC2, advanced permissions, admin controls, workspace consolidation. But the initial wedge was always bottom-up.

Engagement Caveats (Honest Assessment)

Notion’s breadth of adoption is stronger than depth: weekly app engagement averages under 15 minutes (vs. Slack’s 25 minutes), and user churn rate is 31% (vs. Slack’s 14%, though better than Monday.com’s 47%). Mobile MAU growth of 681% (2021-2025) far outpaces the category average of 66%, but per-user engagement remains a challenge.

The Counterintuitive Lessons

  1. Four rebuilds is fine. Most founders give up after one failed version. Zhao rebuilt the product four times over 6 years before hitting real traction. The “fail fast” advice has limits — sometimes you need to fail slowly and thoroughly.

  2. “Growing into your valuation” is real. Notion was valued at $10B on $31M ARR (322x multiple, peak ZIRP 2021). By 2025, $600M ARR on an $11B valuation (18x). The multiple compressed 95% while the business grew 19x. Patience works — but only if the underlying growth is real.

  3. Sugar-coat the broccoli. Notion’s actual product is a no-code app builder — the original failed vision. But wrapped in notes/docs/wiki metaphors, people adopted it willingly. The packaging matters as much as the product.

  4. The setup time can be most of the journey. 6 years to $3M ARR, then 4 more years to $600M. Notion’s growth was non-linear because the platform architecture needed to mature before network effects could compound.

  5. Borrow from your mother if you have to. The $150K family loan during the Kyoto period kept Notion alive. Sometimes the most important fundraising isn’t from VCs.

  6. AI as enterprise trojan horse. Notion’s AI add-on didn’t just generate revenue — it fundamentally changed who buys the product, pulling enterprise customers who would have dismissed “a docs tool.”

  7. Let your users build your distribution. 30,000 templates, 300+ ambassadors, 2,000+ creators, 280K Reddit members — Notion’s community does most of the marketing. The key: don’t pay influencers to read scripts; find people who already love the product.

See Also

Sources

Graph View

Backlinks