Automate Sales Call Research and Follow-Ups with GPT-4, Tavily, Google Sheets

Description

This n8n template automates prospect research and personalized follow-up writing right after a sales call is booked.

It gathers company background, tech stack, and updates; suggests relevant solutions; and then drafts a handcrafted email, subject line, and SMS, all saved back into Google Sheets.

Benefits

Always show up to calls with prepared context. Ensure no follow-up slips through the cracks. Keep all notes, research, and outreach in a single system of record. Scale personalized outreach without losing the human touch.

Use cases

Prep for sales calls with richer prospect context. Save hours writing personal follow-ups. Keep sales records neatly in one shared sheet.

Requirements

Google Sheets with:

Meeting Data → prospect info + enrichment fields Success Stories → testimonials (company, industry, results, solution used) Tavily API key OpenAI API key (GPT-4.1 or later recommended) n8n account with Google Sheets + LangChain enabled

Google Sheets Template

You’ll need two sheets inside one Google Spreadsheet:

1. Meeting Data (input + output)

Columns:

Name Email Company Website Job Title Meeting Date Company Overview (AI-filled) Tech Stack (AI-filled) Company Updates (AI-filled) Primary Solution (AI-filled) Solution 2 (AI-filled) Solution 3 (AI-filled) Email Subject (AI-filled) Follow-up Email (AI-filled) Follow-up SMS (AI-filled)

2. Success Stories (for testimonials)

Columns:

Company Industry Results Solution Used

You can copy this structure into a fresh Google Spreadsheet before starting.

How it works

Trigger → Pulls new rows from Google Sheets (or connects to your booking system). Research Agent → Uses Tavily for company overview, tech stack, and updates. Product Suggestions → Pulls from Product List sheet to propose 3 tailored solutions. Save Research → Updates the row with the six research fields. Sales Writing Assistant → Drafts subject, email, and SMS using context + testimonials. Update Sheet → Writes final outputs (subject, email, SMS) into Google Sheets.

Setup Steps

Prepare Google Sheets

Create a new spreadsheet with the Meeting Data and Success Stories tabs. Add the required columns (see template above). Fill in at least one success story for the workflow to use.

Connect Google Sheets in n8n

Add Google Sheets credentials under Settings → Credentials. Make sure n8n has access to your spreadsheet.

Add API Keys

In n8n, create credentials for OpenAI (GPT-4.1 or later). Create credentials for Tavily.

Import the workflow

Load the JSON file (this template) into n8n. Replace any placeholder spreadsheet IDs with your own. Map the correct sheet/tab names.

Test the Research Agent

Run the workflow manually. Check that Company Overview, Tech Stack, and Company Updates populate.

Test the Writing Assistant

Ensure the workflow uses at least one testimonial from Success Stories. Run it again and confirm Email Subject, Follow-up Email, and SMS are written.

Automate

Replace the Manual Trigger with your preferred trigger:

 Google Calendar / Calendly → Google Sheets
 CRM → Google Sheets
 Directly from n8n integrations

Customization

This workflow is designed as a flexible foundation. Here are ways to adapt it:

Change AI prompts**

Update the Research Agent prompt to focus on specific industries or data points. Edit the Sales Writing Assistant to match your tone of voice.

Swap success stories**

Add more detailed case studies to the Success Stories sheet for stronger personalization.

Add integrations**

Send follow-up emails automatically with Gmail or Outlook nodes. Push SMS directly via Twilio. Sync enriched data into HubSpot, Salesforce, or Notion.

Error handling**

Add retry logic for API calls. Include a notification node (Slack/Email) if the workflow fails.

Tips & Troubleshooting

If research doesn’t update → confirm Email is the matching column in your Update Sheet nodes. If follow-up copy is blank → ensure at least one testimonial exists in Success Stories. Use Debug Mode to inspect AI Agent or Sales Writing Assistant outputs.