22 Udemy Courses, 8 Bestsellers, $2,124 a Month on Average: Building a "Shelf of Courses" on a Global Platform

This is an article in which a practitioner (22 courses, five years as an instructor) answers the question “Is being a Udemy instructor profitable?” with a revenue report attached. The author is Kazuya Isobe (a life coach certified by the US-based Transformation Academy), who runs business-oriented courses on coaching, video production, blog building, and more. The value of this case lies not only in the headline total of $2,124 a month on average (about ¥319,000, converted at ¥150 per dollar, the 2025 average), but in disclosing the “low batting average” too: courses that don’t sell make about ¥1,000 a month.

The Skeleton of the Revenue Report

8 Pillars out of 22 — a Game of Batting Average

The core of this case is that it does not hide the per-course disparity. Of the 22 courses, the revenue pillars are the 8 bestsellers. The rest make around ¥1,000 a month. The hit rate is just under 40%, and you cannot know which course will land until you publish it. Isobe himself writes that when he started, his mindset was roughly “make 100 courses and it’s fine if one hits” — he openly acknowledges the design is premised on volume. In Takizawa’s nine courses too, the first course ended at $165. Being a Udemy instructor is not a “pour everything into one course” craft but a statistical game of publishing multiple courses and reinforcing the hits.

He names three guiding principles: prioritize solving students’ problems over money (value > money), take many shots, and hold an audience framing, an angle, that other courses lack. The third is about differentiation that avoids head-on collision with competitors, suggesting that the variable that moves the hit rate is concentrated in the angle of the topic even before the content of the lectures.

Converted to a pace, 22 courses in five years is 4–5 per year, steady production of one course every two to three months. And the gap between a bestseller (¥50,000–150,000 a month in the chart) and an unsold course (¥1,000 a month) is 50–150x. It is a power-law distribution in which a few hits, not the average, determine the whole. Given that distribution, “make one course and see how it goes” is statistically disadvantaged, and his “100 courses” intuition reads not as exaggeration but as adaptation to the distribution.

Working Backward from ¥481–666

The disclosure of ¥481–666 per student is practically useful. Courses sold through Udemy’s own traffic essentially sell during sales, so the average revenue per new enrollment is ¥1,300–1,800, of which 37% goes to the instructor. In Udemy’s sale-driven culture, you must run your business math on this effective unit price, not the list price: making ¥150,000 a month requires 250–300 new enrollments every month, a number that anchors any back-of-envelope plan. Notably, the 97%/37% revenue split and the sale-dependent structure match the disclosure by Takizawa (2 years in, 9 courses) exactly. With two independent instructors reporting the same numbers, these can be treated as platform mechanics, not personal anecdotes, when planning.

Meanwhile, via coupons the instructor issues himself, prices can be set freely between ¥1,300 and ¥27,800, and 97% stays with the instructor. This is where the breakdown of Isobe’s roughly ¥319,000 income matters: about ¥150,000 came from Udemy’s own traffic, and about ¥160,000 from coupons via his own blog, roughly half and half. For the same course, routing sales through your own audience multiplies the effective unit price several times over. Ride the platform’s distribution while growing reader touchpoints (a blog) outside the platform to increase sales on the 97% side. This dual sales channel is the blueprint behind ¥319,000 a month.

Minimal Equipment — the Screencast Choice

The production-side disclosure is also instructive. His courses are primarily screen recordings (screencasts) requiring no on-camera presence. Recording and editing use WonderShare DemoCreator, and the microphone is a HyperX QuadCast, a minimal setup. While the passive-income quality of “earning even while you sleep” is cited as a benefit, the entry-level capital outlay compresses down to one microphone and one piece of software. Video courses conjure images of camera gear and on-camera hurdles, but business and tool-tutorial courses work with just a screen and a voice.

Beyond money, he cites two benefits: acquired authority and content-production skills. Simply having courses is proof that you are “someone who can teach,” and that credibility flows back into your main line of work. The dual channel is a circuit for revenue and, at the same time, a circuit for trust.

The Downsides, in His Own Words

Isobe does not dress this up as a get-rich story. He spells out the downsides. Exposure of your real name and voice, with the identification risk that carries. No fast money. It is an accumulation model with a slow ramp. And recording-environment constraints: Udemy’s audio-quality standards are strict, and while staying in the Philippines he had to borrow a friend’s place to record because of stray dogs and motorbike noise. Add the structural fact that 14 of the 22 courses hover around ¥1,000 a month. That hit rate itself. The 8 hits are, above all, a result stacked on a five-year time horizon.

Conditions for Reproducing This

What generalizes: working backward from the ¥481–666 effective unit price, a “volume x angle” design for finding hits, and the dual-channel structure of building your own routes (a blog or similar) to make sales on the 97% side. That it can be started with no on-camera presence and minimal gear also applies directly to side-hustlers in Japan.

What does not generalize easily is the time horizon. This income sits on five years and 22 courses of accumulation. It is not a model that produces ¥300,000 a month in year one. Also, the choice of coaching and video production as genres is continuous with his own profession, teachable expertise has to exist first. Learning the platform (what gets searched, which angles land) requires tuition paid in unsold months, just as in Takizawa’s case. Note too that students on a course platform are also “prospects who have listened to you lecture for hours”. The two-stage structure of using Udemy as base income plus touchpoint and connecting it to a higher-priced main business (coaching, consulting, production) is the same structure as tax accountant JUN, who connected Coconala to his practice’s client acquisition.

Related Reading

• A field report of 9 Udemy courses — details on revenue splits and the learning curve
• Coconala x tax accountant — a parallel case of connecting a platform to main-business client acquisition