From 30 yen in revenue to 8 years later: how running 3 apps in parallel got an indie developer to 200,000 yen a month

Most indie developer revenue posts cut out just the numbers from a year that went well. This article covers the opposite kind of record. Posted to Qiita in September 2025 under the title “How my indie development revenue surpassed my starting salary,” the author (@nakapon9517) lays out revenue by year, going back to 2017 — five of those years read “¥0.” After five zeros in a row, the numbers climb: 30 yen, 1,000 yen, 40,000 yen, 200,000 yen. What ended the run of zeros is the center of this case.

Revenue laid out year by year

The article opens with just this table.

The table itself doesn’t specify units, but the body text says: “monthly revenue has now exceeded 200,000 yen, and I’ve finally managed to surpass what I earned as a new graduate.” So the 200,000 on the right end is a monthly figure, and “surpassed my starting salary,” the article’s title, refers to that monthly revenue. Growth is roughly an order of magnitude each year, and across the three years from 2022 to 2025, the multiple works out to roughly 6,600x. But since the starting point was 30 yen, the multiple itself barely matters. What’s worth reading is the fact that 30 yen appeared after five straight years of zero.

What he was doing across those eight years

He currently runs three apps: “Bosuku” (iOS/Android), a bowling score-tracking app; “ShareTore” (iOS/Android), where users training and dieting connect with each other. And “Gazzy” (iOS), which shows monthly walking/running distance on a leaderboard. He also relocated to Okinawa in 2024.

The tide turned right after he’d fully lived through one failure

The turning point sits at the transition from 2022 into 2023. In 2022, the author started building “a service with every feature bundled in,” meant to unify Twitter DMs, LINE open chats, Slack, and other messaging channels into one. Development took about a year. He got it released and started running it himself, and that’s when he first realized something. “Building and maintaining a service just like LINE myself turns out to be insanely hard,” with the constant weight of chasing competitors’ features, storing enormous amounts of data, and managing the risk of leaks. On top of that, “I hadn’t even thought about how hard monetizing a social service would be, so revenue was completely flat.” His motivation ran out, and the service was shut down.

The following year, 2023, he set five rules before touching code again:

• Build a service you’ll use yourself
• Develop only the bare minimum feature set at first
• Release the first version within one month, even unfinished
• Don’t depend on ad revenue — build around subscriptions as the main model
• Research the market and competitors beforehand

Bosuku came out of this rule set. He writes that within roughly a month of release, he noticed store impressions and installs climbing. In revenue terms, this is where the run starts: 1,000 yen in 2023, then 40,000 yen in 2024, then 200,000 yen in 2025.

Why these five lines worked

The five rules are essentially the mirror image of the 2022 failure. The opposite of “took a year” is “release within a month.” The opposite of “everything bundled in” is “bare minimum.” The opposite of “hadn’t thought about social monetization” is “subscription as the main model.” What worked was neither grit nor new technology but flipping each cause of the 2022 failure, one by one. What ended five years of zero was five lines flipped out of a one-year failure.

Structurally, three things stand out. Cutting the development cycle to a month simply increased how many times he could test market response. On a one-year cycle, you only get one shot at validation. The shift from ads to subscriptions lowered the required user count by an order of magnitude. Ad revenue scales roughly with user count, so making a social-style service pencil out requires a large inventory of users. A subscription can work with a small number of continuing users. Built-in market and competitor research put an upfront filter on the 2020–2021 symptom of “something I want but stop using myself.”

Subsequent effort has concentrated on relentlessly polishing the one thing that hit. On ASO: refining the app name, description, and keywords, plus A/B testing screenshots. On monetization: developing paywalled features and A/B testing the purchase screen. On the experience side: speeding up launch time and page transitions, offline support, early error detection via Sentry and Crashlytics, social login to survive device switches, and support for 9 languages including Japanese and English. He designed from the start around reaching a US audience, and even placed his database region in the US. Reaching 200,000 yen a month on a niche theme with a limited competitive population like bowling rests on this “count overseas users from day one” design.

The tactics that didn’t work are listed too

The same list also includes tactics whose cost-effectiveness was clearly poor. Reaching out to celebrities and YouTubers for collaboration got “1 reply out of 20 to 30 people contacted.” DMs to potential users totaled roughly “200 to 300 people, cumulatively”, with no mention of response rate. Paid ads on Apple Search Ads, Google Ads, Yahoo, and X Ads. Press releases to Applishow, Approom, and SeekUps. Promotion on TikTok, Instagram, Facebook, and Reddit, entering a Gemini hackathon and a RevenueCat hackathon. And launching on Product Hunt are all listed alongside each other, with no breakdown of which contributed to revenue.

The author’s own summary is unsentimental: “Unfortunately, in my indie development, not a single big lucky break ever happened.” Our editorial reading agrees with his verdict. No viral moment or major media coverage flipped the numbers, and the year-over-year jumps are the cumulative result of dozens of tactics run in parallel. In our judgment this is a record of design rather than of luck: he kept the number of at-bats up until one connected.

What’s reproducible here, and what isn’t

What’s easy to reproduce is the methodology: release within a month, cut features down, design around paid subscriptions rather than ads, and collect and analyze competitors’ store reviews. None of these require significant capital. Even the release process and ASO knowledge accumulated over five years of zero revenue can now be learned in far less time than it took him.

But some subtraction is needed on the condition side. The author has worked as a contractor doing React Native as his main job since 2019, acquiring practical cross-platform development skills at work and redeploying them into his indie projects. That’s precisely why his development costs were effectively zero. And the eight years of time itself is the largest resource invested, sustained during that period by freelance and contract work. The 9-language support and US-region infrastructure also assume he could handle English-language store review and support largely on his own.

There’s a caveat on the absolute numbers too. 200,000 yen a month is roughly a “new graduate’s starting salary” level, not a large sum as the culmination of eight years running three apps. The value of this article isn’t in the amount, but in the record of what happens over that timeline if you don’t stop even after five years of zero. Also easy to overlook is that the author frames this as “a means to fulfill the parts of life outside indie development”, a framing consistent with the modest absolute number.

Related reading

• Photo AI — the revenue trajectory of an app a solo developer built with public build-in-public design
• Kindle Publishing — a personal record built not on one hit but on volume