You took the call and it went well. You meant to send something the next morning. The recap, the thing you said you would look into, the date you promised to confirm. Then the week happened. It is Thursday, the thread has gone quiet, and sending it now feels worse than not sending it.
Or the other version. A call starts in a few minutes with someone you last spoke to months ago, and you are scrolling your own inbox trying to remember what you agreed to.
Those are two halves of one job, and it is the first job I would hand over in a small company. Not because meetings are special. Because the prep is work you already do badly under time pressure, and the follow-up is a loop you already drop.
What prep actually costs when you do it by hand
Prep is not one task. It is a string of separate lookups in separate places, and every one of them starts with you deciding to go get it.
Who am I meeting. What does their company actually do now, not what it did when you met them. What did we last say to each other, and who owed who something when we hung up. What has moved since then, on their side or on mine, that changes this conversation. And the one that gets skipped most often: what do I want out of this call.
Each of those is a tab, a search, a scroll back through a thread for the line where you promised something.
None of it is difficult. That is exactly why it slides. It is work you have to start yourself, at the moment you have the least appetite for starting anything, because the last call ran long and this one is already starting.
So the cost is not the twenty minutes. The cost is that the gathering sits entirely on you, every time, which means in a busy week it does not happen. You walk in cold and spend the opening of somebody else’s meeting rebuilding context you already had.
What it becomes when something else does the gathering
It becomes one page you read before you walk in.
Who they are and what their company is doing now. What you last talked about and who owed what. What has changed since. What you want out of this one.
Assembled before you asked for it, waiting when you sit down.
That page is not better than what you would write with an hour and a clear head. Often it is thinner. The difference is that it exists on the days you would never have written it, which are the days you needed it most.
The half that matters more
Prep is the part everyone thinks of first. The follow-up after the meeting decides whether the meeting was worth having.
It goes like this. The meeting ends. Something asks you what happened.
That question is the load-bearing part, and the failure it fixes tells you why. Nobody skips a follow-up because the email was too hard to write. You skip it because the meeting ended, the next thing started, and by the time you had a quiet moment the detail had gone soft and the moment had passed.
Being asked at the right time is a different thing from having it on a list. A list waits for you to look at it. The question does not wait.
The capture takes 30 to 60 seconds
Then you talk.
Thirty to sixty seconds out loud, while the call is still in your head. What they said. What surprised you. What you promised and by when. You are not writing anything, you are answering a question, which costs a fraction of what a blank draft costs.
From that, the follow-up drafts itself. You read it, fix the line that is not quite you, and send it.
It is a simple thing. It gets drafted quickly and saves the important notes, so you do not forget and can follow up quickly later.
The payoff is mental load, not minutes
You can count the minutes saved. That is the boring version and it undersells this.
Now you do not have to remember everything. You can offload that part and review it later, when you need to send an email or do any other follow-up.
That is mental load, and it is the expensive part. Not the prep time. Every task left hanging that only you can close, in a founder already carrying too many.
The machine keeps track, not you.
One job, one week
You do not need a strategy for this. You need one job, running for one week, against the meetings already on your calendar. Take the front half on its own if that is all you want. Be prepped for every call next week and pay attention to how those calls open.
Then look at the rest of your week for work with the same shape. The jobs you already do badly under pressure, and the loops you already drop. That is your list, in priority order, written by your own worst weeks.
If you want help turning that list into a build order, that is what the AI Org Roadmap is for. Start at davidchung.ai.