From Zoom call to a signed deal: Automate your proposals in <21 minutes 🤝

EDITOR’S NOTE

Dear Nanobiters,

If you’ve ever wrapped up a client call on Zoom, you know the drill: frantic note-taking, scrambling to capture every detail, and then staring at a blank document, trying to turn that conversation into a polished proposal.

Rewinding the recording to double-check requirements, copy-pasting bullet points, and praying you didn’t miss a critical ask buried in the small talk. Hours later, you’ve got a draft—but is it actually aligned with what the client wants?

Sound familiar?

The truth is, turning meeting discussions into actionable proposals is a grind. It’s easy to lose nuance in translation or waste time formatting instead of strategizing.

But what if your Zoom/G-Meet transcript could automatically become a tailored proposal draft—complete with project scope, deliverables, and pricing—before you even finish your post-meeting coffee?

This week, I built an automation that does exactly that. It extracts key details from meeting transcripts, analyzes them with ChatGPT, and generates a client-ready proposal draft in minutes using Make.com.

No more manual note-sifting, missed deadlines, or copy-paste marathons.

The best part? It’s stupidly simple to set up. No coding, no complex tools—just your existing meeting recordings, ChatGPT, and a few Make.com modules.

Let me show you how.

ARE YOU NEW TO MAKE.COM?

It’s a super easy and intuitive tool to learn.

Here’s a beginner-friendly guide/primer I wrote a few weeks ago. It will help you get started with Make.com.

AI Agent That Turns Zoom Meeting Transcripts Into Client Proposals

This automation transforms raw Zoom meeting transcripts into client-ready proposal drafts—complete with scope, deliverables, and pricing—in minutes.

Let’s break it down.

Click “Create a new scenario” on the top right corner of the page.

This now drops us into the scenario editor or designer, where we can build our automation.

Step 1: Capture Your Zoom Transcript
Tools: Google Drive → Router → CloudConvert

Let’s unpack how these modules work together to grab and prep your Zoom transcript for ChatGPT.

1. Google Drive

• What it does: Automatically detects new Zoom transcripts saved to a specific folder.

• How to set it up:

  • Use the “Watch Plan in a Folder” module on Make.com.

  • Select the Google Drive folder where Zoom saves recordings (or your manual uploads).

• Why it matters: You won't have to manually trigger workflows anymore. The second a transcript lands in Drive, your automation kicks into gear.

• Pro Tip: Name your transcript files clearly (e.g., “ClientX_Project_Date”) to avoid mix-ups.

2. Router

• What it does: Splits your workflow into parallel paths (like a traffic cop for tasks).

• How to set it up:

  • Add a Router after the Google Drive module.

  • If the video's file size is greater than 100 MB, it will be converted into an audio file for further processing.

  • The objective of the router is to create branches for different actions (e.g., one path processes video files greater than 100 MB, converts them into MP3 files for further processing, and another processes the video file directly in the OpenAI modules).

• Why it matters: It keeps your workflow flexible. If you need to add a step later, you can just tack it onto a new branch.

• Pro Tip: Start simple—use the Router only if you need multitasking. For basic workflows, skip it!

3. CloudConvert

• What it does: Converts Zoom’s transcript file (usually .VTT or .TXT) into a ChatGPT-friendly format.

• How to set it up:

  • Add the CloudConvert module after Google Drive.

  • Select “Convert a File” and choose your input/output formats (e.g.,.VTT →.TXT).

• Why it matters: ChatGPT works best with clean, unformatted text. CloudConvert automatically strips timestamps and messy formatting.

• Pro Tip: Most transcripts are already .TXT—double-check before adding this step!

Done? Now ChatGPT can work its magic. ➡️

Step 2: Let ChatGPT Decode the Conversation
Tools: OpenAI (Whisper + GPT-4)

Here’s how the Whisper module transforms your raw audio or multilingual transcripts into ChatGPT-ready text:

Whisper Module: Transcription

What it does:

• Transcribes audio files (if your Zoom recording isn’t already text-based).

• Outputs clean, formatted text stripped of filler words, timestamps, or speaker labels.

How to set it up in Make.com:

To connect Make and ChatGPT, you need an OpenAI API key, which you can find here.

1. Add the OpenAI module and select “Create a Translation” (Whisper).

2. Input source:

   • If working with audio: Attach the Zoom recording file (MP3/WAV) from Google Drive.

Why it matters:

• Audio-to-text magic: Even if Zoom’s auto-transcript fails, Whisper transcribes speech with near-human accuracy.

Pro Tip:

• Use Whisper only if needed. If your Zoom transcript is already in English and text format, skip to the GPT-4 step.

• For large files (>25MB), split the audio into chunks first (use CloudConvert or a free tool like Audacity).

What’s next?
Once Whisper delivers a polished transcript, GPT-4 extracts key details (budgets, deliverables, timelines). No more lost-in-translation moments!

Step 3: Structure the Proposal Draft
Tools: Google Docs + ChatGPT

Now it’s time to turn ChatGPT’s extracted insights into a client-ready proposal. Here’s how to automate the structure, tone, and formatting:

Part 1: ChatGPT’s “Secret Sauce” Prompt

What it does: Transforms raw data (from Step 2) into a polished, structured proposal.
How to set it up:

1. Add another OpenAI module (GPT-4) after Google Docs.

2. Use a prompt like this:

Act as an experienced sales leader with 15 years of experience and generate a proposal for the client using the provided Zoom call transcription by summarizing the key points discussed during the call and outlining the proposed actions to assist the client. 

*Call Summary*

Provide a concise summary of the call, highlighting the client's primary concerns and issues, if any, mentioned during the conversation.

*Proposed Actions*

Outline the steps to be taken to address the client's needs, using bullet points where appropriate. Ensure that all actions are grounded in the information discussed during the call and do not introduce any new or unsubstantiated information.

Use appropriate formatting wherever needed; for instance make the heading bold. Refrain from excessive use of special characters like *** & ###.

*Timeframes and Important Details*

Include relevant timeframe estimates and other crucial details necessary for the proposal, while strictly adhering to the information discussed during the call and avoiding any fabrications.

Why it works:

• Clarity over fluff: ChatGPT cuts vague language and focuses on actionable items.

• Brand alignment: The prompt enforces your voice (e.g., “collaborative” vs. “formal”).

Part 2: Google Docs Setup

What it does: Creates a blank document in your Google Drive, ready to be filled with ChatGPT’s generated content.
How to set it up:

1. Add the Google Docs module in Make.com and select “Create a Document.”

2. Name the file dynamically (e.g., {{Client Name}} Proposal Draft {{Date}}).

3. Set the destination folder (e.g., “Proposals/Automated Drafts”).

Pro Tip: Pre-format your Google Doc with headers, fonts, and branding colors. Attach it as a template to ensure consistency.

Step 4: For Small Files
Tools: Whisper + GPT-4 + Google Docs (No CloudConvert!)

Not every file needs CloudConvert.

If your Zoom transcript or recording is under 100MB and already in a compatible format (e.g., TXT, or MP3), you can streamline the workflow. 

Here’s how to cut the conversion step and save time:

When to Skip CloudConvert:

• Small files: Zoom’s auto-generated transcripts (usually .TXT) under 100MB.

• Direct audio: Recordings in .MP3/.WAV format (under 100MB) that Whisper can process natively.

• Clean transcripts: Files without timestamps or speaker labels (e.g., edited manually).

How to Set It Up:

This is the easiest part: you have to duplicate everything in the Cloud-Convert flow in this part, only without the Cloud-Convert module.

Part - 1

Part - 2

Part - 3

Why This Works:

• Speed: Bypass CloudConvert’s processing time (2-5 minutes saved per file).

• Cost-Efficiency: Reduce Make.com operations by eliminating a module.

• Flexibility: Handle ad-hoc edits (e.g., manually cleaned files) without rework.

Let’s Set It & Forget It: How to Schedule Your Automation

The final touch? Making sure your workflow runs automatically every time a new meeting recording lands in Google Drive. No manual clicks, no babysitting.

Here’s how to schedule it:

1. Enable Automatic Triggers

• In your Make.com scenario, locate the Google Drive module (Step 1).

• Under the module’s settings, find the “Schedule” section. This determines how often Make.com checks your Drive folder for new transcripts.

• Choose an interval: Options include every 5 minutes, hourly, daily, or custom.

Example: If you host 2-3 client calls daily, set it to check hourly. For lighter workflows, daily is fine.

2. Balance Speed vs. Cost

• Remember: Frequent checks (e.g., every 5 minutes) consume more Make.com operations.

• Pro Tip: Match the schedule to your meeting rhythm. If you rarely get same-day follow-ups, a 12-hour interval saves resources.

3. Activate the Scenario

• Toggle the scenario “On” in Make.com’s dashboard.

• That’s it! Now, every new transcript will trigger the workflow without your input.

Why This Matters:

• Zero manual kicks: The automation hums along in the background, like a silent assistant.

• No missed deadlines: Even if you forget, the system never does.

Pro Tip: Test with a dummy transcript first. Rename an old file to trigger a dry run and spot tweaks!

And just like that, you’ve replaced hours of post-meeting busywork with a self-running proposal machine. Time to celebrate—or better yet, use that reclaimed time to win your next client. 🚀

Let’s examine the results from one iteration of the workflow:

For this newsletter, I have used a sample call recording from YouTube.

Why This Output Wins

• Speed: Drafts land in your Drive in 5-10 minutes vs. hours manually.

• Consistency: Every proposal follows your brand voice and structure.

• Accuracy: ChatGPT cross-references the transcript to avoid missed details.

Pro Tip: Save outputs as templates! Tweak prompts to match niche needs (e.g., add “ROI estimates” for enterprise clients or “testimonial highlights” for freelancers).

Your Turn: Run a test call transcript through the workflow.

Beyond Client Meetings: 4 Unconventional Ways to Automate Knowledge Building

This workflow isn’t just for client-facing tasks. Any meeting can become a searchable, actionable knowledge asset.

Here’s how different stakeholders can turn conversations into institutional gold:

1. For Product Teams: Capture Feature Requests from Internal Brainstorms

• What it does: Transcribe sprint planning or UX review meetings, then use ChatGPT to extract and categorize feature ideas.

• Why it works: Auto-tag requests by priority (e.g., “Quick Win” vs. “Long-Term”) and log them in Notion/Airtable.

• Example: “We need dark mode by Q4” → Logged as “Feature: Dark Mode | Priority: P1 | Votes: 8” 

• Knowledge base boost: Build a searchable backlog of user-centric ideas, ranked by team demand.

2. For HR Teams: Transform Exit Interviews into Retention Insights

• What it does: Analyze exit interview transcripts to identify patterns (e.g., “3 employees cited lack of growth opportunities”).

• Why it works: ChatGPT flags recurring themes and generates actionable recommendations.

• Example prompt: Summarize this transcript into 3 key takeaways. Suggest retention strategies. 

• Knowledge base boost: Create a living document of turnover drivers and fixes for leadership.

3. For Educators: Turn Lecture Q&A into Study Guides

• What it does: Convert student questions from class discussions into FAQs or exam prep materials.

• Why it works: ChatGPT identifies gaps in understanding (e.g., “10 students asked about glycolysis steps”).

• Example: Raw transcript: “Wait, why does ATP production drop without oxygen?” Output: Added to “Chapter 5: Common Misconceptions” in the course wiki.

• Knowledge base boost: Build a self-updating resource hub that evolves with student needs.

4. For Community Managers: Archive Event Debriefs into Engagement Playbooks

• What it does: Turn post-event team retrospectives into templated guides (e.g., “How to Run a Viral Twitter Spaces Session”).

• Why it works: ChatGPT distills tacit knowledge (e.g., “Starting with polls boosts attendance by 40%”).

• Example: Transcript: “We should’ve promoted the speaker earlier…”
  Output: Added to “Event Best Practices: Promotion Timeline” in the team’s playbook.

• Knowledge base boost: Turn one-off lessons into repeatable strategies for future volunteers.

The Bigger Picture
Every meeting is a data source. By automating summaries and structuring outputs, you’re not just saving time—you’re building a self-improving system where past conversations fuel future decisions.

Try This: Pick one non-client meeting this week. Run it through the workflow. What hidden gems did ChatGPT uncover? Hit reply and share your “aha” moment. 🔍

End Note

Turning meeting recordings into client-ready proposals doesn’t have to eat up your day. With this automation, you’re not just saving hours—you’re ensuring no detail gets lost, and every draft aligns perfectly with what your client actually said.

The best part? This workflow is just the start. Clone it, tweak it, and adapt it:

• Build custom templates for different industries (e.g., SaaS vs. consulting).

• Add approval steps for complex projects.

• Use ChatGPT to auto-generate follow-up emails after sending proposals.

The setup takes less than 21 minutes. The ROI? Hours reclaimed monthly, fewer missed deadlines, and more deals closed.

I’d love to hear how you use this!
Are you streamlining freelance gigs? Scaling agency workflows? Or finally saying goodbye to late-night proposal marathons? Hit reply and share your story.

Need Help?
Stuck setting up the ChatGPT prompts? Reply to this email! I’ll help you troubleshoot or level up your automation.

Until then, happy automating—and may your proposals always land before the coffee gets cold. ☕

P.S. Revisit your workflow quarterly. Update prompts to reflect new services, or adjust the schedule if your client load changes. Small tweaks = long-term gains.