Maria joined the sales call with her US team at 10 am Swedish time. For forty-five minutes she was sharp, present, articulate. She anticipated objections, reframed a competitor comparison, closed the prospect on a pilot. Off call, she felt good about it.
Then she opened Slack to send a follow-up summary to her US counterpart.
She drafted it. Read it. It sounded smaller than how she'd actually spoken. She rewrote it. Better, but still cautious. Third rewrite: tighter, friendlier, closer to how she'd sounded on the actual call. She sent it and moved on. Twenty-eight minutes had passed since the call ended.
This isn't a vocabulary problem. Maria speaks English with native-level fluency in real-time conversation. The friction is specific to written async communication: when she shifts from speaking to typing in a language that isn't her first, her own voice gets quieter.
Why Speaking and Writing Create Different Friction
This pattern shows up consistently among senior PMs, engineers, and salespeople who are genuinely fluent in English when speaking but feel exposed in writing. The cause is structural, not personal.
Speed isn't evenly distributed across communication modes. The average person speaks at 150 words per minute but types at 40 words per minute. That's a three-fold difference that's invisible to native English speakers. For non-native speakers, it becomes a real tax on cognition.
When you speak, your brain handles language generation and monitoring simultaneously. You anticipate what the other person needs. You adjust vocabulary mid-sentence. You read facial expressions and shift tone in response. You don't need to front-load everything. You iterate as you go.
Writing, especially in a second language, demands the inverse. You must commit to every sentence before you send it. There's no midstream adjustment. No tone-of-voice negotiation. No reading the room. Everything you want to convey has to be crystallized on the first try.
For someone who learned English after childhood, this cognitive difference is pronounced. Your brain is doing simultaneous translation (from how you think, in your native language, into English) while also managing the motor task of typing (which is 3.75 times slower than speaking). Meanwhile, the stakes of written communication in an async workplace feel higher, because every message becomes a permanent record.
So you rewrite. Three times. Four times. Not because the grammar is wrong, it isn't. You're searching for the version of yourself that sounds like the person who just closed that deal on the call. You're trying to recover the presence that evaporates when you move from speaking to typing.
The Compounding Effect on Async Teams
In an async-first culture, this time tax becomes significant. Your US teammate doesn't hear your voice during a Slack exchange. They form their impression of you through text. That matters. It shapes how they think about your ideas, your confidence, your seniority.
So the pressure to get the written version right is real.
Slack message that took 4 rewrites: 22 minutes. Email to a US partner: 35 minutes. Comment in a project doc: 18 minutes. Proposal language: 45 minutes.
Across a week, non-native English speakers in async teams spend an additional 6 to 10 hours on writing that could've taken half the time if the medium didn't create this friction.
That time gets hidden. It doesn't show up in a Jira ticket. It doesn't appear as "waiting for approval." It's just time spent searching for the careful version of yourself, knowing that version isn't actually you.
The Existing Tools Miss the Point
Most tools built for non-native English writers are focused on correctness. Grammarly catches grammar. Duolingo teaches you rules. Both assume the problem is English skill.
The actual problem isn't skill. It's medium.
Your English is fine. Probably better than fine. But writing in a language that isn't your native tongue, under the pressure of async workplace communication, creates a natural caution that speaking doesn't. This shows up as fewer contractions, more hedging, softer language, longer sentences. You end up sounding more formal, more measured, smaller than you are.
Grammarly doesn't help with this because there's nothing grammatically wrong. The writing is correct. It's just not you.
Your most recent performance review probably confirmed this dynamic. Many senior non-native English speakers describe feedback like "excellent communication on calls" paired with "could be more direct in writing" or "needs to be more concise in Slack." It feels like a capability gap. It isn't. It's a medium effect. The gap between how you sound when you speak and how you sound when you write isn't a reflection of your English proficiency. It's a reflection of the cognitive load of translating and typing simultaneously in a language you learned as an adult.
What Changes When Voice Stays Voice
There's an alternative to the rewrite cycle: speak the way you speak on calls and let the system convert that directly into message format.
This isn't dictation software. Dictation gives you a phonetic mess that you then have to clean up and rewrite anyway. This is direct voice-to-message conversion. You say the thing the way you would say it in a meeting. The system listens, understands the intent, and delivers it to Slack as a finished sentence. No intermediate transcription. No manual cleanup. No rewrite loop.
Recitey integrates voice capture directly into your Windows environment, so your spoken message appears as formatted text in Slack without the intermediate step of dictation software and manual rewriting. It's built on Whisper, the speech recognition system trained on diverse global English accents, which achieves 96.3% accuracy on standard English speech tests.
For Maria, this changed the calculus. After that sales call, she spoke her follow-up message the way she would've said it to a colleague sitting next to her. It appeared in Slack in under ninety seconds. No three rewrites. No searching for a more careful version of herself. Still her.
The message had presence. It sounded like the person who was on the call.
The Trade-Off You Make
You give up the sense of control that comes from typing every word and editing iteratively. Your first draft might be slightly rougher than your fifth rewrite would've been. You're optimizing for "sounds like me" instead of "sounds most polished."
For someone who's fluent in speech but hesitant in writing, this trade reverses the problem. You stop spending time searching for a more careful version of yourself. You get the version of yourself that speaks, and it turns out to be the one that lands better.
You also lose the buffer that writing gives you. You can't sit with a message for an hour and polish it incrementally. But research on async communication suggests that incremental polish often erodes voice and personality, especially for non-native English speakers. The version you speak is usually better.
Who This Is Built For
Senior knowledge workers in Sweden, Germany, the Nordics, and other non-English regions who are fluent on calls and spend 20+ minutes per Slack message rewriting. PMs who run crisp standups and then spend 45 minutes on a Slack update. Engineers who explain systems clearly on calls and then second-guess their written decision docs. Account executives who close calls with confidence and then rewrite their follow-up three times.
Founders building distributed teams across time zones. Consultants writing async proposals to US clients. Anyone who's felt the specific friction of being brilliant in one format and careful in another, knowing the difference isn't ability but medium.
The better frame isn't "improve your written English." It's "stop translating yourself for the medium." Your English is already good. The medium is the only thing slowing you down.