Revenue Records from 9 Udemy Courses — The First Course Took 100 Hours and Made $165, but What It Taught Produced a ¥1.5M Course
Udemy bestselling instructor Naoto Takizawa publishes real data across 9 courses. His first course took roughly 100 hours to make and earned $165 (about ¥20,000) in a year — yet a later, iteratively improved course cleared $10,000 (¥1.5 million) in profit. Monthly students grew from 34 in the first month to 563 within a year. The breakdown of revenue-share rates (97% for self-driven sales vs. 37% for Udemy-driven sales) makes this an unusually valuable look at how the system works.
This is a translation of the Japanese original. The Japanese version and cited primary sources are authoritative for all figures.
The published numbers
| Item | Figure |
|---|---|
| Courses published | 9 courses |
| First course | ~100 hours of production → $165 (about ¥20,000) in one year |
| A later course that took off | Over $10,000 (¥1.5 million) in profit |
| Monthly students | 34 in month one → 563 about a year later |
| Revenue share | 97% via the instructor’s referral links / 37% via Udemy search and recommendations |
| Impact of sales | 98–99% of purchases happen during sales. New daily enrollments: ~10 normally → 50+ during the year-end sale |
How to read “$165 for the first course” correctly
One hundred hours for ¥20,000 — at ¥200 an hour, this number alone says “quit.” But what Takizawa’s record shows is that the real deliverable of the first course was not revenue but learning: what gets searched, which titles and thumbnails get clicked, where students drop off. The follow-up courses that incorporated this learning generated over $10,000, and monthly students grew 16x in a year, from 34 to 563.
Multiple courses also create a structure of mutual cross-referral. A student who takes one course browses the same instructor’s other courses. The more courses you have, the lower the customer-acquisition cost per course — exactly the same catalog compounding as the “build a shelf” strategy in Kindle publishing.
What the 97% vs. 37% split reveals: the price of customer acquisition
Udemy’s cut is 3% on sales the instructor drives themselves, and 63% on sales Udemy drives. That gap is a price list proving that customer acquisition is the most expensive step in the chain. Since the variable behind a 3x revenue difference (by his own account) is not lecture quality but traffic source, an instructor’s real skill set is the ability to teach multiplied by the ability to own acquisition channels (social media, an email list).
The concentration of purchases during sales (98–99%) also means list prices are merely anchors. Unit prices drop, but riding Udemy’s sales and recommendation engine is effectively the distribution strategy — designing around this “platform gravity” is the realistic approach.
Related reading
Sources
This article summarizes and analyzes the public sources above. Please refer to the primary sources for details.