You were sharp on the call. You explained the approach clearly, asked good questions, landed the key point. The client was engaged. You felt present.
Then you drafted a follow-up message in Slack to your team.
You read it. It sounded careful. Smaller than how you actually talk. You rewrote it. Better, but still not quite right. Third pass. Fourth. By the time you sent it, the message took 40 minutes and still sounded like the cautious version of yourself.
This is not a vocabulary problem. This is not a grammar problem. This is a medium problem.
The Call Is Where You Sound Like Yourself
Maria is a senior account executive at a B2B SaaS company in Stockholm. On sales calls, she's sharp. She handles objections with precision, builds rapport quickly, closes deals. Her last two performance reviews flagged "exec presence on calls" as a strength. She knows she's good on the phone.
But everything she writes in Slack, email, Salesforce, and proposals goes through three, four, sometimes five rewrites. Not because the grammar is broken. Not because the vocabulary is wrong. But because the written version sounds smaller and more careful than how she actually communicates.
She knows the performance review feedback about "more concise in writing" is not really about writing. It's about the medium. When she writes, something shifts. The complexity that flows naturally in speech becomes hesitation on the page.
She's not alone in this.
Async Writing Isn't a Slower Version of Speaking. It's a Different You.
Non-native English speakers often describe this gap more honestly than native speakers do. You know the difference between how you sound on a call and how you read on Slack because you feel it every single time.
It's not that you lack words. It's that writing forces you to pre-package your thoughts. You have to commit to a phrasing before you send it. On a call, you can correct yourself mid-sentence, add nuance, feel the other person's reaction and adjust. In Slack, you're alone with the draft.
So you rewrite. You're checking not for grammar but for voice. Does this sound like me? Does this sound competent? Does this sound like I belong in this conversation?
The math is brutal. If you spend 5 minutes per Slack message on rewrites, and you send 15 substantive messages a week, that's 75 minutes of rewrite tax every week. Over a year, that's 65 hours. That's two weeks of work, spent on edit cycles that didn't add new information, just anxiety reduction.
Grammarly Doesn't Solve This
You've probably tried Grammarly. It catches missing articles, suggests tense changes, cleans up comma placement. These are real problems if you're still learning English.
But Grammarly misdiagnoses the real issue. It assumes you need correction. What you actually need is speed. You don't need someone to tell you that "it's" not "its." You need to move from thought to sent message without the identity-protective rewriting cycle in between.
Grammarly also erases what survives in your writing despite being non-native. Your directness. Your precision. The way you phrase complex ideas. These are not grammar errors. They're your voice. Grammarly treats voice as noise to be smoothed out.
The Real Tax Is Translation Delay
Here's what happens in your head: you think a thought in your native language, translate it to English, read it, feel it's not quite right, retranslate, try again. This is different from a native speaker's block, which is usually perfectionism. Your block is identity. Does this English version still sound like me?
Native English speakers have blank page anxiety. You have translation fidelity anxiety. Different problem, different solution.
Worse, when you finally send that message, it has lost something in the rewrite cycle. The conversational edge is gone. The clarity remains, but the warmth is muted. You sound competent but careful. That's close to who you are, but not quite.
What Happens When You Dictate Instead
The alternative is to speak it first and preserve the voice intact.
When you dictate, you bypass the translation-and-rewrite cycle. You skip directly from thought to draft. Whisper-v3 reaches 96.3% accuracy on LibriSpeech, which means the dictation captures what you actually say. Yes, there are rough edges. Every first draft has them. But the voice that comes through is closer to your actual voice. The Slack message sounds more like the person who crushed the call 15 minutes earlier.
Then a quick polish for typos and colloquialisms that don't read well in text. Two seconds. Not two minutes. Not 40 minutes.
The Slack message to your team captures the presence and clarity you had on the call. The follow-up email to the client carries the competence and directness that landed the deal. The deal note in Salesforce reflects the nuance you were actually thinking, not the watered-down version you typed out after 4 rewrites.
Where This Actually Works in Practice
A developer at a fintech company was getting feedback that his technical decisions were sound, but his documentation was hard to follow. He started dictating the technical explanation, the same way he'd explain it to a colleague at the desk. Rough draft in 90 seconds. Clean version in under 2 minutes. The documentation clarity didn't change, but it sounded less defensive. It sounded like someone who knew what he was doing, not someone proving he knew what he was doing.
A consultant drafting a proposal in Google Docs used to take 2 hours on 800 words. She'd write, read, rewrite for tone, rewrite for logic, rewrite because the sentences didn't flow. She started speaking the proposal structure first. Thirty minutes to draft. Fifteen to polish. She still rewrote, but the rewrite was now about precision, not identity.
Maria sent a Slack message to her team without the rewrite cycle. It took 4 minutes instead of 40. The message sounded sharper. More direct. More like the person who just closed the deal, not the person pretending to have confidence on async.
It integrates into Slack, Google Docs, Salesforce, browser, and terminal, plus every other Windows app via the system clipboard. So you're not context-switching to a separate writing tool. You're dictating where you already work.
The Point
You sound like yourself on calls because you're not alone. You're in response mode. You're meeting someone in real time. Async writing forces you into a different stance, solitary, pre-packaged, identity-protective.
Dictation doesn't fix writing. It fixes the medium. It gives you back the presence you have when you're speaking, even when you're writing alone.