You can't ask a note a follow-up question

Notes only go one direction

Think about the last podcast you listened to that was actually good.

You probably took notes. A few lines in your phone at a stoplight, maybe a page of them when you got home. And they're still sitting there.

Here's the problem with notes. They only capture what you understood in the moment. When the guest mentions a study you've never read, or throws out a term you sort of know, or makes a claim that doesn't quite sit right with you, all you can do is write it down and keep listening. The question you had never gets answered. It just becomes a line in your phone.

That's why the notes die. Not because you didn't care. Because there was nothing else to do with them.

Make it a conversation, not a capture

The shift is treating AI as a notebook you can talk back to.

Same episode, three different jobs:

Before you listen, get the map. Drop the transcript in and ask for a summary. Now you know what's coming, and you can listen for the parts that matter to you instead of waiting to find out if the episode is worth your time.

While you listen, chase what catches you. Guest mentions a concept you want to understand better, go deeper on it right then. Something sounds relevant to a problem you're having, ask what it looks like in your setting. You're not saving questions for later anymore, later is where questions go to die.

When something sounds off, check it. Speakers make assumptions. They generalize from their own athletes, their own resources, their own level. When a claim doesn't sit right, ask what the research actually says, and whether the evidence supports it or the speaker is filling in a gap with opinion.

That last one is the biggest change. You've always had that instinct when something sounded shaky. Now you can do something with it while you still care about the answer.

Inside the Lab: run it in the Project you already built

This all happens in the Continuing Education Project from a couple issues back. No new setup, and it already knows your situation from your Program Card.

Drop the transcript in, ask for the summary, then listen.

The questions you ask after that are just the ones you already have. What does he mean by that term? Is there research behind this or is that experience talking? What would this look like with 60 athletes instead of 20? She said this works better than the alternative, what's the actual evidence there?

Ask them the way you'd ask another coach, and keep going until you've got what you need.

What it looks like in the real world

Say a guest spends ten minutes on in-season monitoring and lands hard on daily wellness surveys. Sounds good, but something bugs you about it.

So you ask AI. What's the evidence behind daily wellness data and using that to make changes to training plans? It comes back with what the research supports and where it thins out, and you find out the strongest evidence is in populations that don't look much like yours.

Then you keep going. What would you actually do with a flag from one of those surveys? Now you're working through your own follow-up, not the speaker's. By the time you're done you know what you'd keep, what you'd drop, and why.

That's an hour of podcast turning into something you can act on, instead of four lines in your phone you'll never open again.

Where the coach stays in charge

Verification runs both directions. AI can tell you when a speaker is stretching, but it can also be confidently wrong. If something matters enough to change what you're doing, ask for the actual studies and go look at them yourself. Use it to find the source, not to replace reading it.

A good conversation isn't a decision. You can spend twenty minutes going back and forth on monitoring and still be exactly where you started, which is fine. Understanding something better is the point. Whether it belongs in your program is a separate call, and that one's yours.

The Coach's Scouting Report

• Signal. A note only holds what you already understood. A conversation gets you what you didn't.
• Noise. Writing down questions you never answer. Taking a speaker's claim at face value because they sounded sure.
• Next Rep. Next podcast, run the transcript through first and get a summary before you listen. Then ask three questions you'd normally have just written down.

See you next Sunday.