Quick Answers
What does it mean for a platform to be "localized" but not usable?
It means the pages that explain and sell a product are translated, while the product screens further downstream — signup, checkout, account setup — are never checked at all. A visitor can read fluent Japanese the entire way and still hit an English-only screen at the exact moment they're supposed to convert.
How do you actually catch a gap like this?
Not by reviewing the marketing pages again. The only reliable way is to walk your own signup, checkout, or application flow yourself, screen by screen, as a user, in the language you claim to support — instead of assuming the flow is covered because the pages around it are.

TL;DR

We localized the entire marketing surface of a platform we run into Japanese — the site, the blog, the landing pages — and considered the job done. It wasn't. The signup screen, the one page where a visitor actually becomes a customer, was still in English, on a platform that advertises support for nine languages. We found this by accident, looking at a partner's dashboard: steady clicks, zero signups. Our own localization checklist had never covered that screen, because it only covered the pages meant to explain and persuade, not the page where persuasion turns into action. The fix wasn't better translation. It was walking our own funnel ourselves, as a user, in the language we claimed to support.

Key Takeaways

The number that shouldn't have been zero

I was looking at a partner's dashboard, mostly to admire the numbers. One partner had a healthy click count next to their name — steady referral traffic, the kind you're glad to see. I scrolled over to conversions.

Zero. Not low. Zero.

I sat with that for a second, because zero is a strange number to trust on the first look. I checked it again. Still zero, for a market we'd already told ourselves was localized.

We had already called this one done

Earlier in the year, we'd put real effort into localizing the platform's marketing surface into Japanese: the site, the blog, the landing pages. It felt like a finished project — a box we could stop checking. That confidence is exactly what made the gap invisible for as long as it was. Once a project feels done, you stop looking at it the way you looked at it while it was still in progress.

Nine languages. One screen was missing.

The dashboard number sent me back to the funnel itself, and I noticed something I hadn't actually done before: I had never clicked all the way through our own signup flow, myself, as a user, in Japanese. I'd checked the landing page. I'd checked the article that sent people there. I had never checked the last screen.

It was English. The one screen where someone actually becomes a customer was in English, on a platform that advertises support for nine languages. Japanese wasn't a low-priority gap on that screen. It was simply missing — never added, because that particular screen had never been part of anyone's localization checklist.

The checklist was the problem, not the translation

Our own checklist covered the pages that explain the product and the pages meant to persuade someone. It never covered the page where persuasion turns into action. "Localization," in practice, had quietly come to mean articles and landing pages, and the screens further downstream were assumed to be someone else's job, or already fine, or not worth a separate line item. None of that was true.

There's research that lines up with what that zero was telling us: CSA Research's 2020 survey across 29 countries found that 76% of consumers prefer product information in their own language, and 40% won't buy from a site that's only in another language (CSA Research, n=8,709, source). We didn't need the number to know it was true. We needed it to admit we'd built a process that didn't act like we believed it.

Looking Japanese and being usable in Japanese are not the same thing

A visitor can read your article, trust your landing page, click through with real intent — and still hit a wall right before the outcome the whole funnel was built for. That gap doesn't show up in a content review. It shows up in a partner's dashboard, as a number that's exactly zero.

So the question worth asking isn't really "have you localized." It's narrower than that: have you personally walked your own signup, checkout, or application flow — not the marketing pages around it, the actual flow — as a user, in the language you claim to support? Not skimmed. Walked, screen by screen, to the end.

The shift in one line: keep checking whether the content is translated, and add a step that checks whether the flow itself is walkable — done by someone who goes through it as a real user would.

What we fixed, and the question we're keeping

We fixed ours. The signup page went live in Japanese on July 24, 2026. What stayed with us wasn't the fix itself — it was how long that one screen had sat there, one click past everything we'd already checked, while we felt confident the whole thing was done.

A quieter question for your own funnel

You probably know whether your marketing pages are localized. Most teams can answer that one.

Here's the one most teams haven't asked. When did you last walk your own signup, checkout, or application flow yourself, screen by screen, as a user, in the language you claim to support? If the answer is "we're not sure," that's not a failure — it's just the check you haven't run yet. When you want a starting point, a Japan readiness check is built to look at exactly that.

Frequently Asked Questions

We already translated our whole site — could this still be a problem for us?

Yes, if the translation effort stopped at the pages built to explain and persuade. Screens further downstream, such as signup, checkout, account setup, or in-app onboarding, are often outside the scope of a content-focused localization pass, and they're exactly the screens a visitor has to get through to become a customer.

How do we check for this without a big audit?

Walk the flow yourself. Switch the interface to the language you advertise, open the signup, checkout, or application flow as a new user would, and go through it screen by screen to the end. It takes minutes and needs no tooling.

Isn't a platform that supports nine languages already covering this?

Not necessarily. Language support is often implemented page by page rather than as a single global switch. A platform can genuinely support nine languages across most of its surface and still miss one screen, especially a screen that sits outside the marketing site's translation scope.

Is this the same thing as a bad translation?

No. A bad translation is a quality problem on a page that exists in the target language. This is a coverage problem — the screen was never translated at all, because it was never on the list of pages anyone checked.

What's the cheapest first step?

One test: complete your own signup or checkout flow in the language you're advertising, from the first click to the confirmation screen. If you can't get through it cleanly yourself, a visitor in that market won't either.