This system activates the moment a client books a call and prepares everything automatically.
or, Explore 7000+ Free Workflows
What this system does (end result)
When someone books a call:
- You receive a personalized meeting brief
- The client receives a custom proposal
- Your team gets the brief (email + optional audio)
- Everything is saved in Google Sheets
- No manual research. No copy-paste. No prep stress.
Tools you’ll need
- n8n (core automation)
- Cal.com (or Calendly / Google Calendar)
- Perplexity API (deep internet research)
- OpenAI / OpenRouter (analysis, proposal, audio)
- Gmail (sending emails)
- Google Sheets (storage + tracking)
Step-by-step setup
1. Trigger: Calendar booking
Use:
- Cal.com → Booking Created trigger (or Calendly / Google Calendar equivalent)
Setup:
- Connect API key
- Trigger fires on every booking
Important:
- Pin sample booking data after first test This lets you build the workflow without re-booking calls.
2. Filter the correct event
Why:
You may have multiple event types.
How:
- Add an IF node
- Check event_id === your_target_event_id
- Only proceed if it matches
All other bookings get ignored or routed elsewhere.
3. Extract booking details
Use a Set node to cleanly extract:
- Attendee name
- Attendee email
- Company name
- Meeting type
- Notes / answers from booking form
- Meeting link
This becomes the input for all research.
4. Centralized research prompts (important)
Create one Set node that stores all prompts:
- Attendee research prompt
- Company research prompt
- Competitor research prompt
- Meeting goal analysis prompt
Why:
You can tweak prompts in one place
instead of editing 10 different nodes.
5. Research with Perplexity
Use HTTP Request / AI Agent with Perplexity API.
Recommended setup:
- Model: Sonar / Pro
- Depth:
- Competitors: deeper (10+ pages)
- Attendee + company: lighter (5 pages)
Run separate research calls for:
- Competitors
- Attendee
- Company
- Meeting goal
6. Merge all research into one object
Use a Merge node with:
- Mode: Combine
- By: Position
- Inputs: all research outputs
Why this matters:
- Append = multiple items (bad here)
- Combine + position = one clean object
This is critical for downstream reliability.
7. Proposal + internal summary generation
Use an LLM Chain / AI Agent.
Input:
- Merged research
- Your company knowledge base (services, strengths, use cases)
Output must be strict JSON:
- html_summary
- confidence_score (0–100)
- proposal_good (true/false)
Tip:
Enable required output format + auto-fix
This prevents broken JSON.
8. Quality gate (very important)
Add an IF node:
If:
proposal_good === true
Then:
→ Send proposal to client
Else:
→ Do nothing (no half-baked proposals sent)
9. Send proposal to client
Use Gmail node:
- HTML body = proposal HTML
- Meeting link = pulled directly from booking data (never let AI generate links)
This avoids broken URLs.
10. Internal meeting brief (team-facing)
Generate a separate internal summary:
- Clear attendee profile
- Company context
- Talking points
- Risks / opportunities
Format:
- HTML (for email)
- Plain text (for audio)
11. Optional: Audio briefing
Use OpenAI Audio Generation:
- Input: internal meeting brief
- Output: MP3 file
Attach this audio to the internal email so:
- You (or your team) can listen instead of read
12. Email internal team
Use Gmail node:
- To: Host email
- CC: Team members (optional)
- BCC: Ops / sales (optional)
- Attach: MP3 briefing
- Include HTML summary + meeting details
13. Save everything to Google Sheets
Use Google Sheets → Append Row.
Store:
- Booking details
- Attendee + company
- Proposal confidence score
- Proposal status (sent / not sent)
- Meeting date
- Notes
This becomes your:
CRM + audit log + performance tracker.
Final result
You now have a system where:
- Calls trigger prep automatically
- Research is consistent and deep
- Proposals are personalized and gated
- Team alignment happens instantly
- Everything is documented
No scrambling.
No last-minute prep.
No forgetting context.
When this setup makes sense
Best for:
- Consultants
- Agencies
- Freelancers
- Sales-led businesses
- High-value discovery calls
If calls matter, this system pays for itself fast.
Follow Vikash Kumar for more helpful content.