The Stakeholder Cheat Sheet I Built From Meeting Transcripts
- Amy Westlake
- May 5
- 3 min read
Situation
I was several meetings into a working relationship with a program owner I didn't know well. Every time I needed to send him something or prep for our next conversation, I found myself doing the same thing: staring at a blank draft, trying to reverse-engineer what tone would land, what to lead with, what to avoid.
I was wordsmithing toward an unknown target.
It wasn't that I had bad information. I'd been in meetings with this person. I'd watched him react to things. I just hadn't done anything with what I'd observed — because in the moment, you're managing the meeting, not taking notes on the human.
The AI Move
I had the meeting transcripts in NotebookLM already. On a whim, I asked it to build me a communication profile.
I didn't give it a template or a list of categories. I just asked: based on these transcripts, what can you tell me about how this person communicates and what he responds to?
What came back was more structured than I expected. Communication style. Priorities. Sensitivities. What motivates him. What creates resistance. And the part I use most: a dos and don'ts list.
I now pull that list up before every meeting with him. Not because I've forgotten who he is — but because having it written down means I'm not running that analysis in the back of my head while I'm trying to be present in the room.
I saved it as a Google Doc in my program folder, added it as a source in NotebookLM, and started pulling it into Gemini prompts whenever I need to draft something to him. The message that comes out isn't generic. It's calibrated — to his pace, his priorities, the things that make him dig in.
The Shift
The obvious shift was in my communication — more confident, less second-guessing.
The less obvious one was what happened to the profile itself. After the steering committee meeting last week, I updated it. I try to update it after every significant encounter. And recently I started tracking something new: email behavior versus meeting behavior. How he communicates in writing versus how he shows up live. That's still early — I don't have a strong read yet — but I suspect those are two different profiles, and I want to know where they diverge.
I also started building profiles for groups. Same framework: tone, sensitivities, dos and don'ts. When I'm prepping comms to a full steering committee, I'm not writing to one person's preferences — I'm writing to a room. Having a group profile changes what I lead with and how I frame tradeoffs.
The Pattern
Every meeting is generating data about the people in it. Most of that data lives nowhere — in your gut, in half-formed impressions, in the vague sense that something works with this person and something else doesn't.
The profile makes the implicit explicit. It takes what you already noticed and turns it into something you can actually use — and build on.
You're not learning something new about the person. You're finally writing down what you already knew.
The Implication
Think of one stakeholder you communicate with regularly but feel like you're still guessing at.
Find two or three transcripts from meetings with them. Ask your AI tool of choice: based on these conversations, build me a communication profile. Include their style, what they care about, what creates friction, and a practical dos and don'ts list.
Save what comes back somewhere you'll actually find it. Use it the next time you draft something to them.
Then notice whether the message feels different to write.
What I'm Testing Next
Group profiles are working for tone and communication style. What I haven't built yet is a profile that captures group decision-making patterns — how a particular leadership team actually makes decisions, where resistance tends to come from, what kind of framing gets things across the line. That's the next version.
Comments