Modules/Module 6/Lesson 4
Lesson 4 of 6 ~10 min read

AI for Meetings

6.4 — AI Before, During, and After Meetings

A team in a meeting room with a laptop open on the table

What You'll Learn

Meetings are expensive. A one-hour meeting with six people doesn't cost one hour — it costs six. And yet most meetings are poorly prepared, poorly documented, and poorly followed up. AI can improve all three phases: before the meeting, during it, and after it.

In this lesson, you'll learn which tools to use, how to use them, and how to build a meeting workflow that actually saves time rather than adding to it.

Before the Meeting: Preparation

Good meeting preparation is rare. Most people show up having skimmed the invite and improvised from there. AI can help you prepare quickly and well.

Agenda Creation

If you're running the meeting, use AI to turn a rough idea of what you want to cover into a structured agenda.

Prompt:

"I'm running a 45-minute team meeting to review Q2 results, align on Q3 priorities, and assign owners to three key projects. Write a tight agenda with time allocations and a brief description of what each section should achieve."

Output: a professional agenda you can send with the invite, that signals to attendees you've thought this through.

Pre-Read Briefing Documents

For meetings involving external guests, board members, or senior stakeholders, AI can help you write briefing notes quickly.

Prompt:

"I'm meeting with [client/stakeholder] on Friday to discuss renewing their contract. Here's the background: [paste context]. Write a 1-page briefing document summarising the relationship history, current issues, and our objectives for the meeting."

Preparing Talking Points

If you're attending a meeting rather than running it, AI can help you prepare contributions.

Prompt:

"Here's the agenda for a budget review meeting I'm attending. I'm responsible for the marketing budget section. Based on these figures [paste], what are the three most important points I should be ready to make, and what objections should I anticipate?"

Key takeaway: Five minutes of AI-assisted preparation can make the difference between a meeting that drifts and one that delivers a decision.

During the Meeting: Transcription Tools

AI-powered transcription tools have transformed what's possible during meetings. Instead of someone frantically taking notes while trying to participate, the whole conversation can be captured automatically, then summarised and processed by AI.

Someone on a video call with a laptop and headphones

The Main Tools

Otter.ai Otter joins your Zoom, Google Meet, or Teams call and transcribes in real time. After the meeting, it produces a full transcript, an AI summary, and identified action items. Free plan covers 300 minutes/month. Works well for smaller teams.

Fireflies.ai Similar to Otter but with more integrations (Salesforce, HubSpot, Slack) and stronger search across past meetings. Useful if you want to search "what did we decide about the pricing model in last month's meeting?" across your whole history. Free plan available.

Microsoft Copilot in Teams If your organisation uses Microsoft 365, Copilot in Teams is built directly in. It transcribes, summarises, and can answer questions like "What did I miss in the first 10 minutes?" after the fact. Requires a Copilot for Microsoft 365 licence.

Notion AI / Google Meet transcripts Google Meet now offers built-in transcription (Workspace plans). Notion AI can process meeting notes if you paste them in.

How to Use These Tools Well

  • Tell participants at the start that the meeting is being recorded and transcribed. This is both polite and, in many jurisdictions, legally required.
  • Let the tool run without worrying about capturing everything yourself. You can be more present in the conversation.
  • Review the AI summary before it goes anywhere — these tools are good but not perfect, and misattributed quotes or missed nuance does happen.

After the Meeting: Action Items and Follow-Up

This is where the real value is. The meeting ends, everyone disperses, and nothing happens — until the follow-up email lands. AI can generate that email in about two minutes.

Generating Action Items from a Transcript

If you have a transcript (from any of the tools above, or from notes you took), paste it into ChatGPT, Claude, or Grok.

Prompt:

"Here is a transcript from our team meeting. Please: 1) list all action items with the owner's name and any mentioned deadline, 2) summarise the key decisions made, 3) note any open questions that weren't resolved."

Writing the Follow-Up Email

Prompt:

"Based on this meeting summary, write a follow-up email to the team. Tone: friendly but professional. Include: a brief summary of what we agreed, the action items as a numbered list with owners and deadlines, and a note about when we'll next meet to check progress."

This takes two minutes and means the team has a clear record before they've even left the building.

A Real Workflow in Practice

Here's how one small business uses this end-to-end:

  1. Fireflies automatically joins all external client calls
  2. After the call, Fireflies emails a summary and transcript
  3. The account manager pastes the summary into Claude and asks for a draft follow-up email
  4. They review and personalise it, then send within 15 minutes of the call ending
  5. The action items go into their project management tool

Total additional time: about 5 minutes. Total time saved: probably 30–45 minutes of note-writing and email drafting.

Practical Tips

  • Create meeting templates. Once you've built a good prompt for a particular type of meeting (weekly team sync, client review, project kick-off), save it. Reuse it every time.
  • Don't send raw AI output. Always read and edit before sending. The AI doesn't know your tone, your relationship with the recipient, or the subtext of what was discussed.
  • Use AI to prepare for difficult conversations. "I need to give difficult feedback to a team member in a meeting this week. Help me structure what I want to say clearly but compassionately." AI is surprisingly good at this.
  • Try 'catch me up' prompts. Paste a meeting transcript and ask: "I'm new to this project. Summarise the most important things I need to know from this meeting."

Key takeaway: AI doesn't make meetings shorter. But it makes the time before and after them dramatically more efficient — and a well-prepared meeting often is a shorter meeting.

What to Try This Week

Pick one meeting you have this week. Before it: use AI to write an agenda or prepare talking points. After it: take your notes (or a transcript if you use a transcription tool) and use AI to generate action items and a follow-up email. Notice how much time it saves and whether the quality of follow-through improves.

Previous lessonNext Lesson