A few weeks ago, I was walking home from the office one evening, and a sentence landed in my head:
"This campaign should start with the pain of forgetting, not the speed of typing."
I fumbled for my phone. It was at the bottom of my bag, tangled in earbuds. By the time I unlocked it, opened Notes, tapped "New," and started typing, the sentence was already gone. What I wrote down that night wasn't the same thought.
This has happened to me more times than I can count.
It's not just me
A teammate of mine runs every morning. Halfway through her route she calls her own voicemail to leave herself reminders. An engineer sends himself Slack messages from the passenger seat, one-handed, full of typos.
We don't have a shortage of ideas. We have a capture problem.
We don't think in paragraphs. We think in fragments, in half-sentences, in sudden five-second insights that don't feel important until later. And by "later," I mean after they're already gone.
Why the best thoughts don't wait for us to be ready?
I read later that a Stanford study by Marily Oppezzo found walking can significantly improve creative thinking, with creative output increasing by an average of about 60% compared with sitting. I didn't need a study to believe it. Every time I leave my desk, ideas start lining up.
The best thinking happens in the in-between. Walking to lunch. Waiting for coffee. Driving home. Lying in bed with the lights off, right before sleep. These look like empty moments. They are not. They are where a lot of real thinking actually lives.
The problem is that "in-between" is exactly where our tools stop working.
I've tried every tool. They all failed.
I've spent the last two years trying to solve this for myself with what already exists.
Voice Memos. Fast to start, impossible to revisit. Two weeks later I find a 47-second recording called "New Recording 23" and I have no idea what I was thinking.
Apple Notes. Unlock, open, new note, type. Four steps minimum. By the time I start typing, the sentence in my head has already changed shape.
AI Notetakers. Built for meetings. Not built for a person walking alone at night.
A physical notebook. I tried carrying one for a month. I was never holding it when I needed it.
For a while I thought this was my fault. I should write faster. I should remember better.
Then I came across a concept from cognitive science called cognitive offloading: using external tools or actions to reduce the mental burden of remembering or processing information. Relying on tools to hold thoughts isn't a weakness. It's how the human mind extends itself.
So the issue wasn't me. The issue was with the tools themselves.
Every one of them makes the same assumption: that the thinking is already finished by the time you pick them up. Keyboards are built for paragraphs. Notes apps are built for organization. Voice recorders are built for long audio. And you can feel that assumption every time you reach for one. Capturing a single sentence on your phone right now takes eight separate steps: take it out of your pocket, unlock it, find the app, open a note, decide where to put it, start typing, fix the typo, maybe add a title. That flow is fine for writing a document. It's terrible for catching a thought. A thought does not wait eight steps.
Because most thoughts don't arrive finished. They arrive half-formed. In fragments. In a single sentence that hasn't found its own shape yet. A tool built for paragraphs can't catch something that isn't a paragraph yet.
Why we're building SpeakON
Here is a number that stuck with me. In human conversation, the average response time between one person finishing and another replying is around 200 milliseconds. That is how fast our brains are built to move.
Our tools work on a different clock. From "I have an idea" to "I've written it down" is closer to 20 seconds. A 100x gap between how fast we think and how fast we can capture what we think.
SpeakON is our attempt to close that gap.
It's a MagSafe AI voice button. It snaps onto the back of your phone. You press it, you talk, you move on. No unlocking. No opening an app. No choosing a folder. Two seconds later, what you said is in your inbox — transcribed, searchable, cleaned up if you want it to be.
That's it. That's the whole product. We didn't want to build another voice input tool. We wanted to build a button for thought capture. Something that moves at the speed of the idea.
What I'm hoping for
I still think about that sentence I lost walking home. It came back to me, eventually, in a slightly worse form. But I also think about the other ones — the ones that never came back.
Some of those I only remember because I stumbled across them in an old draft months later. Most of them, I have to assume, I simply lost.
None of these were finished ideas. They were seeds. And seeds are easy to lose.
But if capture becomes frictionless, the seeds don't have to die. A salesperson's elevator thought becomes the opening line of next quarter's pitch. A PM's half-sentence on a run becomes the first ticket of the next sprint. A founder's three-in-the-morning realization becomes the first slide of Monday's all-hands.
That's what we're trying to build. Not a better recorder. Not a smarter note app. A button.
So that the next time a sentence lands in your head while you're walking home from the office, you don't lose it before you get to the door.
