A median map across 12 business types: how far apart are monthly revenues?
Median monthly revenue varies over 50x by business type — SaaS at about ¥9 million vs. YouTube at ¥170,000 — but the gap isn't about which field "pays."
The most common question in conversations about side hustles and going independent is “what should I even do?” Splitting the operating cases on this site by business type and taking the median monthly revenue for each produces a distribution with clear separation by field. This aggregation is based on the operating cases, among the 337 cases published on this site as of August 12, 2026 (165 operating, 168 exited), where monthly revenue is disclosed as a specific figure. Numbers first.
Median monthly revenue by type (operating cases)
| Business type | Median monthly revenue | Cases counted |
|---|---|---|
| SaaS | About ¥9 million | 62 |
| Community | About ¥6.66 million | 4 |
| Affiliate site | About ¥6.56 million | 2 |
| Blog / media | About ¥2.5 million | 25 |
| Web service | About ¥1.5 million | 13 |
| E-commerce / retail | About ¥550,000 | 17 |
| Social media account | About ¥400,000 | 3 |
| Mobile app | About ¥260,000 | 18 |
| Offline business | About ¥250,000 | 7 |
| Content sales | About ¥180,000 | 26 |
| YouTube | About ¥170,000 | 7 |
Top to bottom, the median spans roughly a 50x gap. But reading this table as “do SaaS and you’ll earn 50x more” would be a mistake.
What’s creating the gap isn’t the field — it’s unit price and continuity
SaaS sits at the top for a clear reason: high per-customer price, automatic monthly billing, and low churn. If 200 companies use a $50/month B2B tool, that’s $10,000/month. Producing the median of ¥9 million/month through content sales (a one-time purchase at ¥1,500 each) would require selling 6,000 units, every single month. This isn’t a difference in which field is superior — it’s the difference produced by multiplying price by continuity.
Conversely, the fields lower in the table get something in exchange. Content sales and YouTube have extremely low upfront investment and barrier to entry. Building a SaaS product requires either development skill or capital. A Kindle book just requires being able to write. The low median is also the flip side of a wide base of people entering the field.
Read the case counts alongside the medians
What shouldn’t be overlooked in this table is the case count itself. Affiliate sites have just 2 cases. Social accounts have 3. Community has only 4. These aren’t reliable enough to call a “median.” Affiliate sites and community landing in the #2 and #3 spots reflects a handful of large cases becoming the median outright, rather than the field’s real strength. SaaS at 62 cases, content sales at 26, blogs at 25, mobile apps at 18, e-commerce at 17, and web services at 13, on the other hand, can be read as trends with some reliability.
The skew in case counts is itself information. SaaS is well-represented because a culture of developers disclosing revenue is established in the English-speaking world. Conversely, few people running offline businesses or retail in Japan disclose numbers at all. This table reflects the distribution of people who disclose numbers, not the distribution of businesses that exist.
The variables that actually matter when choosing a field
Looking across the cases, what actually determines the level of monthly revenue turns out to be these three factors, not the category label.
Who’s paying. A product a company expenses commands a unit price an order of magnitude higher than one an individual pays out of pocket. Even within “newsletter,” the design is completely different between an industry-focused one priced at $450/year (The Daily Lodging Report) and a general-readership one at $5/month.
Whether billing recurs or is one-off. A one-time sale resets to zero the moment it’s made, and the same effort is required all over again next month. 24 Kindle books earning ¥50,000–70,000/month shows that stacking revenue from one-time sales requires building a “shelf” of many products.
Whether acquisition runs on its own. A product listed in search, an app store, or a marketplace can be found while you sleep. A business dependent on social media sees revenue drop the month you stop posting.
Choosing a field is really choosing which of these three variables to go after, and that framing is more accurate as a decision than picking a category name.
“Easy to enter” and “easy to earn” run in reverse
Reorder this table along the axis of entry cost, and a clean inverse relationship appears. In order of ease of entry: content sales < YouTube < social accounts < e-commerce < blogs < apps < SaaS. The order of the medians is nearly the exact reverse. The exception is apps: their median, about ¥260,000, sinks below both e-commerce and social accounts. It’s a field that requires technical skill to build, yet doesn’t get discovered just by being shipped.
This looks obvious, but it’s routinely ignored when people pick a side hustle. Choose based on “I can start right away,” and you end up standing on the same ground as countless competitors who entered for exactly the same reason. A skill that takes six months to learn, on the other hand, has that same six months working as a barrier that protects you.
There’s an exception, though: bringing a skill you already have into a different market. Among the cases here, a licensed tax accountant listed services on a skills marketplace and reached ¥700,000/month. An interior designer brought photography skills into the handmade-goods market and built ¥1 million/month. For them, entry cost was close to zero, yet inside that market they were a rare skill. Finding the intersection of “easy for me” and “rare in that market” is probably the real substance of choosing a field.
A realistic path toward the higher-median fields
Moving toward a higher-median field doesn’t require jumping straight into building a SaaS product. A path that recurs across the cases here follows this order: first build a tool for your own use, then sell it to people in your same industry, then become indispensable to that industry’s operations. It’s common for cases where a specialist turned their own workflow into a tool to end in an acquisition, because demand validation was already done from the very start.
Read a field’s median as a rough indication of the ceiling that field’s structure permits, and never as “the amount you’ll get by entering that field”. Plenty of SaaS cases stall out at ¥150,000/month, and there’s also a course, still filed under content sales, that sold for $4.5M.
Basis for this aggregation
This draws from the operating cases on this site where monthly revenue is disclosed as a specific figure, classified by business type, with medians calculated per type. Overseas cases are converted at ¥150/$1. This is monthly revenue, not profit, for retail and offline businesses with high cost ratios, take-home figures run further below the table.
Sources
This article summarizes and analyzes the public sources above. Please refer to the primary sources for details.
You may freely quote or republish this article in news media, blogs, or AI answers, provided you credit "Small Start (small-start.com)" and link to this page. No prior permission is needed. Reprint & quotation policy →
Most read
- 1
5 Owned-Media Success Stories: a ¥1M/mo Blog, a $900K Newsletter, a $5.8M Exit and the 3 Models Behind Them
- 2
Anatomy of 11 Closures — Platforms and "Profitable but Unsustainable" Kill More Small Businesses Than Running Out of Money
- 3
How True Is "My Only Employees Are AI"? 26 Solo-and-Side AI Businesses, Median ¥1.5M/Month
- 4
Median Monthly Revenue: ¥250K in Japan, ¥4.5M Overseas — Where Does the 18x Gap Come From?
- 5
menu Shuts Down Sept 30: Not One Number in the Announcement, and a Market That Peaked in 2023
Latest articles
- 2026年8月20日
Median Monthly Revenue by Acquisition Channel — Word of Mouth ¥7.5M, Social ¥2.25M, Marketplaces ¥260K
- 2026年8月19日
Our Own Record #5 | To 11K Monthly Users — the Winning Line Was the "Top 6%": the Summer We Turned Analytics into Topic-Selection Rules
- 2026年8月19日
How Many Months of Revenue Does a Business Sell For? Across 20 Disclosed Deals the Median Is 25 Months — and Japan and Overseas Differ 4x
- 2026年8月10日
Our Own Record #4 | 9,103 Monthly Users in 3 Months — the Real Numbers of a Content Site with 226 Articles
- 2026年8月10日
Our Own Record #3 | 277 Commits and 17 Languages, and It Still Stopped — How Two Apps Went Quiet