I priced the 443-person company I would have had to hire

In 2016 I built an illustrated story library the conventional way. I hired a writer,
illustrators, translators and narrators, and I managed it book by book. It found readers on its
own — the app is still live on Google Play,
100,000+ downloads and 975 reviews, with no marketing budget. Then it stopped. Not because nobody wanted it. Every page was a commissioned painting, and I could
not afford page 19. I shipped 18 stories out of 50 and that was the end of it. I came back to it alone. It is now twelve libraries and 2,341 illustrated books, each in
thirteen languages and narrated, shipped as 24 mobile apps from one React Native codebase plus a
React/TypeScript web reader. ## The number people actually ask for "How many people is that really?" came up often enough that I stopped guessing and priced it: role
by role, at mid-market salaries, sized against real throughput — an illustrator finishing ten
paintings a month, a translator covering 50,000 words a month, against the real catalogue. 443 people. $12.87M in year one. Illustration alone is 255 of them. Translation is 120. ## Two things I think generalise 1. The judgment is the job. The tooling executed against twenty years of knowing what good looks like. It did not supply that, and the times I let it supply that, the output was obviously worse.

1. The uncomfortable part, said first: those 443 roles were never jobs anyone lost. They were jobs nobody could afford to create. What did not exist before is not the staff — it is the 2,000+ books that were never going to be written at any price I could pay. I published the whole ledger, including where the method is weak, so you can check the arithmetic. Read the full breakdown → Happy to answer anything here — the pipeline, keeping a character's face consistent across a book, judging a translation in a language you don't read, or what the rejection rate actually is.