All systems
System 01AI Meeting Systems

Prepare for meetings faster using AI-assisted workflows.

Reduce manual research, improve preparation quality, and generate better meeting context using practical AI systems.

~3h saved / week
Beginner → Intermediate
ChatGPT · Claude · Perplexity · Notion
What this system solves

The real reason you walk into meetings under-prepared.

It's rarely about effort. It's about not having a repeatable system for turning scattered information into clear context — fast.

  • Repetitive preparation
    Re-doing the same kind of research before every call instead of running a system.
  • Fragmented research
    Context lives in five tabs — LinkedIn, the website, the CRM, the last email, your notes app.
  • Poor meeting context
    You walk in knowing the agenda but not what each person in the room actually cares about.
  • Generic questions
    Default discovery questions that don't move the conversation past surface-level answers.
  • Manual follow-up
    Writing the same recap and next-steps email from scratch after every call.
Who it's for

Built for people who run on meetings.

Sales professionalsConsultantsRecruitersFoundersAccount managersGTM teams

If your week is shaped by 8–15 conversations and the quality of each one compounds — this system is for you.

Workflow overview

Six steps. Roughly ten minutes.

Run them in order the first few times. Once the rhythm clicks, you'll cherry-pick the steps you need.

Step 01
Research the company

Understand the business before the conversation, not during it.

How to run it: Use Perplexity or ChatGPT with browsing to summarise the company: what they sell, who they sell to, recent news, funding, and strategic direction.

Step 02
Research stakeholders

Know who's in the room and what they're likely to care about.

How to run it: For each attendee, generate a short profile: role, tenure, recent posts, and the priorities most relevant to your meeting.

Step 03
Generate a meeting context brief

Consolidate everything into a one-page brief you can scan in 60 seconds.

How to run it: Combine the company and stakeholder research into a single structured brief — context, attendees, likely outcomes, and risks.

Step 04
Generate strategic questions

Ask questions that surface real constraints, not surface-level answers.

How to run it: Prompt the model to draft 5 sharp, open-ended questions tailored to the company, the role of each attendee, and the meeting goal.

Step 05
Build a meeting notes structure

Capture the right things during the call — without having to think about format.

How to run it: Generate a lightweight notes template (context, key points, decisions, owners, next steps) and keep it open in Notion or Apple Notes.

Step 06
Generate a follow-up draft

Send a clear, specific recap within an hour of the call ending.

How to run it: Paste your notes back into the model and ask for a short follow-up email: what was discussed, what was agreed, and the single next step.

Example prompts

Four prompts. Run the whole system.

Tested across GPT-5, Claude Sonnet, and Gemini 2.5. Replace anything in {curly braces} before sending.

Company brief
Act as my research assistant. Produce a one-page brief on {COMPANY}.

Sections:
1. What they do, in plain English (3 lines).
2. Who they sell to and how they make money.
3. Recent signals from the last 6 months: funding, hires, launches, public posts.
4. Strategic direction — what they appear to be betting on.
5. Three things that would matter to them in a meeting about {TOPIC}.

Tone: calm, direct, factual. No hype. Cite sources where possible.
Stakeholder profile
For each of these attendees, produce a short profile:
{ATTENDEES}

For each person include:
- Role and likely scope of responsibility
- One recent public signal (post, talk, hire, launch) if known
- The one thing they probably care most about in a meeting on {TOPIC}
- A talking point that would land with them specifically

Keep each profile to 4 lines maximum. No filler.
Strategic questions
I have a meeting with {ATTENDEES} from {COMPANY} about {TOPIC}.

Draft 5 strategic, open-ended questions I can ask. They should:
- Surface real constraints, not surface-level answers
- Be specific to this company and these people, not generic
- Mix at least one question about current process, one about success criteria, and one about what's already been tried

Avoid leading questions and anything that sounds like a sales script.
Follow-up draft
Below are my notes from the meeting with {ATTENDEES} at {COMPANY}.

Notes:
"""
{NOTES}
"""

Write a follow-up email that:
1. Opens with one specific thing they said (not "great chatting today")
2. Recaps the 2 most important points in their language
3. Names the single next step we agreed on, with a date
4. Ends with one clear ask

Constraints: under 140 words. Match my tone — direct, warm, senior. No "circle back" or "synergy".
Example outputs

What the system actually produces.

Two artefacts from a real-world prep cycle. Pricing review, three attendees, thirty-minute call.

Acme × You · Thu 14:00 · 30 min · Pricing reviewMeeting brief
Context
  • Acme is reviewing their current vendor stack ahead of a 2026 budget cycle.
  • Pricing review was triggered by a 22% YoY increase on their last renewal.
  • Likely outcome they want: a clear path to flat or reduced spend without losing functionality.
Stakeholder summary
  • Sarah Chen · VP Operations
    Owns vendor budget. Cares about predictability and reporting.
  • Marcus Lee · Head of RevOps
    Day-to-day user. Cares about workflow continuity.
  • Priya Singh · Finance Partner
    Joined last quarter. Cares about unit economics, not features.
Strategic questions
  1. 01What would a successful renewal look like to each of you, in one sentence?
  2. 02Where is the current setup costing you more time than money?
  3. 03What did you try last year that didn't work?
Follow-up draftSent within 1h
Subject: Quick recap + next step

Sarah — picking up on your point about predictability heading into the 2026 cycle, here's what I took away:

1. The 22% jump on the last renewal is the real trigger, more than functionality.
2. Marcus is comfortable with the current workflow — the question is commercial, not product.

Proposed next step: I'll send a flat-rate option scoped to your current usage by Tuesday, with two scenarios for Priya's unit-economics view. Workable?

— You
Common mistakes

Where this system tends to break.

Trusting AI without verification
Models hallucinate names, titles, and recent events. Always sanity-check anything you'll quote in the room.
Generic questioning
Asking the model for 'discovery questions' produces a script. Ask for questions tailored to this company, this person, this topic.
AI-generated corporate language
If the brief reads like a press release, your meeting will too. Strip filler and rewrite in your own voice.
Overcomplicated prompting
Long, layered prompts produce vague output. Shorter, structured prompts with clear sections work better.
Replacing judgment
AI can structure the prep. Deciding what matters in the conversation is still your job.
The human layer

AI should support your thinking — not replace it.

Use this system for the parts of preparation that are repetitive: research, structure, drafting. Keep the parts that matter — judgment, communication, strategy, the relationship in the room — yours.

AI is for
  • Operational leverage
  • Workflow acceleration
  • Preparation quality
You are for
  • Judgment
  • Communication
  • Strategy & relationships
Issue 001Shipping weekly

Weekly practical AI systems for modern work.

Useful workflows, operational systems, and practical AI implementation strategies for modern professionals. No hype, no trend-chasing — just systems that work.

Next dispatch · Monday 06:00 CET

No spam. Unsubscribe anytime.