The 90-second client call brief

The 30-minute pre-call ritual nobody admits to

Most people walk into client calls the same way. Twenty minutes of email archaeology. A Drive folder hunt. Scattered Notion notes. A quiet hope that nothing important slipped through.

There's a better way and it takes 90 seconds.

The old way vs the connected way

The pattern is familiar. Open Gmail, search the client name, scroll through months of threads trying to remember what was promised. Open Drive, hunt for proposals and shared docs. Open Notion, dig up the last call notes. Stitch it all together in your head before joining. That's 30 minutes if you're disciplined. Often longer.

Sometimes you just wing it and pay for it mid-call when a client mentions something you completely forgot you agreed to handle three weeks ago.

The shift is small. Connect Claude to Gmail, Drive, and Notion. Run one prompt 90 seconds before the call. Walk in fully prepared.

What makes this work isn't the AI. It's the context retrieval. Most of what makes knowledge work exhausting isn't thinking. It's finding. The right email thread, the right document, the right note from three months ago. Claude connected to your apps collapses that fragmentation into one coherent brief in seconds. You stop spending mental energy on memory management and start showing up with actual presence on the call.

In partnership with

Turn AI Into Extra Income

You don’t need to be a coder to make AI work for you. Subscribe to Mindstream and get 200+ proven ideas showing how real people are using ChatGPT, Midjourney, and other tools to earn on the side.

From small wins to full-on ventures, this guide helps you turn AI skills into real results, without the overwhelm.

Get Your Free Guide

The exact prompt that does it

Here's what one consultant now runs before every client call:

Point 5 is the one that earns it. Claude reads tone across multiple email threads and surfaces things you'd miss skimming. Frustration that's been building quietly. An expectation that was never quite said out loud. A question they've now raised twice. That kind of synthesis used to require careful re-reading. Now it happens automatically.

Points 2 and 3 do the heavy lifting on trust. Walking into a call already knowing what you owe someone, and having a specific response ready for what they last raised, changes how the call opens. You're not catching up. You're continuing a conversation you already fully understand.

Your competitors already read this every morning.

The AI Report keeps 400,000+ executives ahead of every major AI move — in 5 minutes a day. Trusted by leaders at the world's top companies. The question isn't whether AI is changing your industry. It's whether you'll see it coming.

Get the edge free

*Ad

How to set this up

Setup takes about 2 minutes per connector. No code. Works with Gmail, Drive, Calendar, Notion, Slack, HubSpot, and 200 plus others through Claude's built-in integrations. Most people connect Gmail first, then add Drive once they see how the brief improves with document context layered in.

Permissions stay scoped to what your account can already see. Connecting Drive doesn't open anything your login couldn't access before. Claude reads within your permissions, not around them.

Narrow the window for long client relationships. Six plus months of history? Limit the prompt to 90 days. The brief gets sharper. Too much history creates noise. The recent window keeps everything actionable.

Customize per client by adding specifics. "Flag anything they've asked twice that I haven't answered." "Include any pricing discussions verbatim." Claude folds those in naturally. For high-stakes calls, add "note the overall tone trajectory across the last five emails" and it will tell you whether the relationship is warming up or cooling down.

One thing worth knowing. Claude won't send anything or make changes without showing you first. The brief reads and synthesizes. Nothing goes anywhere.

Start with Gmail

If you only connect one thing this week, make it Gmail. The client call prep prompt above is the one that pays for itself fastest. You'll feel the difference on the very first call you run it before.

The mental model shift happens the first time you walk into a call knowing exactly where things stand, what the client expects, and what you owe them. Not from 30 minutes of hunting. From 90 seconds of reading. Once that becomes your normal, the old approach feels like showing up to a meeting you forgot to prepare for.

Stop using Claude as a smarter search box. Connect it to your actual apps and it becomes something that reads across your real data and synthesizes what matters before you even join the call.

The questions everyone asks

Can I lock this to read-only so it doesn't accidentally send anything?
Yes. Claude's Gmail access can be set to read-only. No drafting, no sending, nothing leaves your system without your hands on it. The tools are read-only by default, so you'd have to explicitly add send functionality to the prompt for it to even try. Start read-only. Add send later if you want.

What if Claude misses something important in the brief?
Honest answer: Claude nails the broad strokes (usually 80 to 90 percent) but can miss tone shifts buried in a long email chain. Most people do a quick 2 to 3 minute skim of the original threads before the call to catch what the AI missed. Bonus move: ask Claude to flag any promises you made but haven't delivered. That accountability layer catches gaps the summary alone wouldn't surface.

Can you use this to draft follow-ups after the call?
Yes. Run the same pattern in reverse. After the call, give Claude a recap of what happened and let it draft a follow-up email grounded in the original context. Closes the loop and saves you from writing emails later.

What breaks when you reorganize folders or rename Notion pages?
The trap: if you hardcode folder paths in the prompt and your structure changes, the brief comes back thin. Fix is simple. Keep discovery hints (folder paths, Notion page IDs) in a separate config file instead of burying them in the prompt. One update to the config and you're good even after a major restructure.

My honest take

The expensive part of knowledge work was never the thinking. It was the finding. Every time you walk into a call cold, you pay a tax in attention, energy, and trust. Connectors don't make you smarter. They just stop charging you that tax.

Try it once before your next call. Five minutes of setup. Ninety seconds of reading. Then notice what changes about how the call opens.

Check out the original post here.

How 2M+ Professionals Stay Ahead on AI

What’s the secret to staying ahead of the curve in the world of AI? Information. 

Luckily, you can join 2,000,000+ early adopters reading The Rundown AI — the free newsletter that makes you smarter on AI with just a 5-minute read per day.

Sign up to start learning.

*Ad