It started with a single message from a friend in Brazil. “Hey, can you make Wordit work in Portuguese?”
I said yes. A few days later, I pushed the update. But I didn’t do what you’d expect. I didn’t add all the proper accented characters — é, ã, ô, ç. I stripped them out.
You’ve probably been told that good localization means honoring every orthographic nuance. That if you’re building for a native audience, you must respect the sacred rules of their language. That’s what the textbooks say. But my friend, who actually uses the tool, didn’t want that. He wanted simplicity. He wanted to type without fighting a keyboard layout that doesn’t match his phone.
So I made a choice that would make linguists cringe: I removed accentuated words to keep it simple.
Here’s the thing about product design: Sometimes the best way to respect a language is to let go of its sacred rules.
This isn’t about laziness. It’s about friction. Every time a user has to hold down a key to find the right accent, every time their autocorrect guesses wrong, you’ve introduced a tiny gate. For a utility like Wordit — a tool for quick thinking and wordplay — that gate is a wall. My friend didn’t want to play “correct Portuguese.” He wanted to play.
There’s a dangerous idea floating around that accessibility means adding more options. More fonts, more characters, more settings. But real accessibility often means subtraction. Removing the things that aren’t essential. Letting the user go fast.
I saw this firsthand. I built the feature in an afternoon, pushed it live, and my friend started using it immediately. Not a single complaint about missing accents. Because he already knew what he meant. The tool filled in the gaps.
This is the twist nobody talks about: True localization isn’t about being correct — it’s about being usable. A perfectly accented interface that makes you hunt for characters is a failure. A stripped-down version that lets you flow is a win.
Of course, someone will say this is cultural erasure. That I’m dumbing down a rich language. But I’d argue the opposite. By removing the barrier, I made the language more accessible to the people who love it. Not every tool needs to be a dictionary. Some just need to work.
Take a side: If you’re building for a specific audience, ask them what they actually hate. Then cut that. Neutrality is death in product design. Commit to the user’s real experience, not the ideal.
So here’s my golden quote, the one you can screenshot and send to your team: “The best localization removes the things that slow people down — even if those things are letters.”
Wordit now speaks Portuguese. Just not the Portuguese you learned in school. And that’s exactly how my friend wanted it.
FAQ
Q: Why remove accents? Aren't they essential for correct Portuguese?
A: They are essential for formal writing, but not for a fast-paced word game. The user's priority was typing speed, not grammatical precision. The tool still works — players naturally understand the intended words.
Q: What's the practical takeaway for product builders?
A: Ask your actual users what they hate, not what they theoretically need. Then remove that friction. Often the 'correct' version is the slowest version. Ship the stripped-down one first, add complexity only if demanded.
Q: Isn't this just lazy engineering?
A: No. It's intentional design. Lazy engineering would be ignoring the language entirely. Thoughtful design means choosing which features serve the user and which serve the ego. Removing accents for a specific audience is a deliberate trade-off, not an oversight.