Why Your Japanese Documentation Is Quietly Hurting Your Product
There is an art form in Japan called dorodango.
This compound word breaks down simply: doro means dirt, and dango means dumpling. Dorodango are made by shaping ordinary dirt into a sphere, then polishing it, layer by careful layer, until the surface becomes smooth, dense, and mirror-bright. I have held dorodango that were decades old and passed down through generations. These objects began as common earth and, through nothing but patience and attention, became something genuinely beautiful. Something worthy of display.
I think about dorodango often when I think about Japanese documentation.
Japan has shaped my understanding of craft more than any course or credential. I first moved there in February 1989, and I have been returning ever since. I’m always drawn back by the people, the language, and a cultural orientation toward quality that is difficult to articulate but impossible to miss once you’ve experienced it. In Japan, the gap between doing something and doing something well is not considered optional. It is the point.
When a US company enters the Japanese market with documentation produced by Google Translate or a bulk localization vendor, Japanese users notice. They may not complain. They may not even consciously articulate what’s wrong. But the impression is made, and it is not the one you worked so hard to create with your product.
If you’re looking to enter the Japanese market or improve your performance there, you want to avoid mistakes that may halt your progress almost before you start. Here’s what some of those pitfalls look like in practice, and what to do instead.
Red Flags
Machine translation that propagates errors at scale
The most insidious problem with machine translation isn’t any single mistranslation. It’s those errors that propagate. Run your documentation through something like Google Translate, and every mistake it makes gets replicated across every page, every UI string, every customer-facing surface simultaneously. A systemic misunderstanding of your product’s terminology doesn’t produce one bad sentence; it produces a bad document, a bad interface, and a bad impression, consistently and at scale.
I have seen this cause genuine product failures in the Japanese market. Not because the product was bad, but because the documentation and UX writing surrounding it communicated, at every touchpoint, that the company hadn’t taken Japan seriously enough to get the words right.
Technical terminology being treated as vocabulary rather than convention
Japanese has a sophisticated system for handling foreign technical terms, and machine translation routinely gets it wrong in ways that are immediately obvious to native readers.
Consider the word “Windows.” Any Japanese technical reader knows this refers to Microsoft’s operating system, rendered in katakana as ウインドーズ. Machine translation, applying vocabulary logic rather than technical convention, sometimes renders it as 窓, the Japanese word for an ordinary glass window. The result isn’t just incorrect. It’s jarring, as category errors always are, like someone referring to an Apple computer as a piece of fruit.
I encountered an equally striking example while working on a point-of-sale system. The documentation referred to the “right light,” meaning the indicator light on the right side of the device. A translator rendered this as ライトライト, which is a meaningless, direct phonetic transliteration of the English words “right” and “light” placed side by side. It communicates nothing to a reader, and worse, it signals that nobody who understood Japanese reviewed the document before it shipped.
These are not edge cases. They are the predictable output of treating Japanese as a translation problem rather than a writing problem.
Cultural tone that misreads the audience
Language is only part of localization. The images, design choices, and implied cultural assumptions that surround your documentation matter as much as the words themselves; sometimes more.
I worked with a company that had produced a shopping app for the Japanese apparel market. Their marketing materials featured businessmen in black suits and ties, smiling. The intent was professional and aspirational. The effect on Japanese readers was closer to unsettling. In Japan, black suits and ties are strongly associated with funerals. The image didn’t communicate professionalism. It communicated that nobody involved in producing the materials understood Japan well enough to catch a significant cultural misstep.
Documentation carries the same risk. A tone that reads as refreshingly direct in the US can come across as blunt or disrespectful in Japan. Humor that works in English often doesn’t translate, not because Japanese people lack a sense of humor (they certainly don’t), but because the cultural context that makes something funny often doesn’t cross the language boundary. Getting the tone right requires not just fluency in the language but genuine familiarity with the culture.
Typography that makes complex kanji illegible
Japanese documentation presents a technical challenge that Latin-script documentation does not; complex kanji rendered at fixed small sizes become illegible. Simple kana and common kanji survive the compression reasonably well. But the more complex characters, like those that often carry precise technical meaning, collapse into indistinct pixel blots at small font sizes, forcing readers to infer meaning from context rather than read it directly. For example, the kanji for person (人) is easily read, but the kanji for machine (機) or depression (鬱) becomes incomprehensible at small font sizes.
This happens most often when companies use interface elements or document templates designed primarily for Latin characters, with Japanese support added as an afterthought. The font sizes that work perfectly for English text are simply too small for complex kanji. The result is documentation that is technically present but functionally unreadable, exactly where precision matters most.
Green Flags
Native Japanese technical writing, not translation
The difference between translation and writing is the difference between a dorodango and a handful of dirt. Both start from the same material. Only one has been shaped, refined, and polished into something worthy of the audience it serves.
Native Japanese technical writing begins in Japanese by a writer who understands both the technology and the language, rather than beginning in English and then being translated by linguists who often have no technical skills to speak of. The result reads naturally because it is natural, not because a translator found adequate equivalents for English phrases, but because a writer who thinks in Japanese chose the right words from the beginning.
This is what we offer at Inkwright, Inc. Our Japanese documentation isn’t translated. It’s written.
Technical terminology is handled by convention, not guesswork
A native Japanese technical writer knows that ウインドーズ is Windows, that マウス is a computer mouse, and that 右の電気 is the light on the right side of a device. These are conventions, not vocabulary questions, and handling them correctly requires familiarity with Japanese technical writing as a discipline, not just fluency in the language.
Ask any localization vendor you evaluate how they handle technical terminology. The answer will tell you immediately whether you’re talking to someone who understands the problem.
Cultural fluency that goes beyond language
The best Japanese localization partners don’t just translate your words; they translate your intent. They know which images will resonate and which will misfire. They understand the appropriate tonal register for your audience and product category. They can tell you when your documentation’s implicit assumptions about how users think and behave don’t map cleanly onto Japanese professional culture.
This kind of fluency isn’t available from a translation mill. It comes from people who have spent significant time in Japan, who understand the culture from the inside, and who care enough about the outcome to say something when something isn’t right.
Typography and layout designed for Japanese readers
Quality Japanese documentation is designed for Japanese readers from the beginning, not adapted from a Latin-script template. That means font sizes large enough for complex kanji to remain legible, layout choices that account for the different visual weight of Japanese text, and a general attentiveness to the reading experience that treats Japanese users as the primary audience rather than an afterthought.
The cost of getting it wrong
Japan is one of the world’s most demanding technology markets. Japanese users have high expectations for product quality, and documentation is part of the product. Poor documentation doesn’t just frustrate users; it also communicates how much you value their business.
The dorodango passed down through generations in Japanese families weren’t preserved because dirt is precious. They were preserved because the care taken with them is precious; because taking something ordinary and making it exceptional, through patience and attention and genuine craft, is itself a value worth honoring.
That’s the standard Japanese users bring to your documentation. It’s the standard we bring to producing it.
If your company is entering the Japanese market, or is already there and wondering why adoption isn’t where it should be, the documentation is worth a careful look. We’d be glad to help.
Inkwright, Inc. provides technical writing and ghostwriting services for technology companies, in English and Japanese. Get in touch.