Automate Personalized Email Campaigns with n8n, Twitter & LinkedIn

Opening Problem Statement

Meet Justin, a busy marketing manager at Pollup Data Services. Every week, Justin spends hours manually researching leads by digging through their LinkedIn profiles and Twitter feeds. After gathering a handful of posts, he writes personalized emails to propose relevant services. It’s tedious, time-consuming, and prone to errors—sometimes emails don’t connect well because Justin can’t keep up with every lead’s unique interests. If only there was a way to automate this tedious process while keeping the communication deeply personal and relevant.

Justin typically wastes over 5 hours weekly on this task, delaying follow-ups and missing out on potential deals. Mistakes in personalization lower his email response rate, directly impacting Pollup Data Services’ sales pipeline.

What This Automation Does

This specific n8n workflow automates the entire process Justin struggles with by integrating social media data extraction with AI-driven email creation, saving him precious time and improving outreach precision.

• Automatically trigger lead processing from a Google Sheets list filled with leads’ LinkedIn URLs, Twitter handles, and emails.
• Fetch up to 10 recent LinkedIn posts for each lead using RapidAPI’s Fresh LinkedIn Profile Data service.
• Retrieve up to 10 recent tweets for each lead using RapidAPI’s Twitter API.
• Analyze social media content with OpenAI’s GPT-4o chat model to identify common interests or activities that align with Justin’s company profile.
• Generate a personalized email subject and HTML cover letter customized for each lead based on their social media activity.
• Send the tailored email directly to the lead, carbon-copying Justin for tracking purposes, and update the Google Sheets to mark the lead as processed.

By automating these steps, Justin saves at least 4-5 hours per week, reduces manual errors, and significantly increases the relevance and impact of each outreach.

Prerequisites ⚙️

• Google Sheets account 📊 – to store and trigger lead processing from a spreadsheet with columns: linkedin_url, name, twitter_handler, email, and a “done” status column.
• RapidAPI account 🔑 – subscribed to Fresh LinkedIn Profile Data API and Twitter API for social media extraction (free tiers available with usage limits).
• OpenAI account 🔐 – with API access for GPT-4o chat model integration.
• SMTP email service account 📧 – configured for sending emails automatically (could be SMTP, Gmail, or other supported email service).
• n8n account ⚙️ – to create, run, and manage the automation workflow. Self-hosting options are also available if desired.

Step-by-Step Guide to Build This Workflow

Step 1: Create Your Lead Google Sheet

Go to Google Sheets and create a new spreadsheet named “Analyze social media of a lead”. Add columns: linkedin_url, name, twitter_handler, email, and done.

Fill the sheet with sample lead data but leave the “done” column empty. This column controls which leads get processed.

Step 2: Set Up the Google Sheets Trigger Node in n8n

In your n8n editor, add the Google Sheets Trigger node. Configure it to monitor the created sheet by selecting your document ID and sheet name (gid=0). Set it to poll every minute.

This node kicks off the workflow whenever new or unprocessed leads are detected.

Step 3: Add a Conditional “If” Node to Filter Processed Leads

Add an If node connected to the trigger. Set the condition to check if the “done” field is empty (meaning the lead hasn’t been processed yet). Only proceed if this field is empty.

This avoids duplicate processing and emails.

Step 4: Configure the “Set Your Company’s Variables” Node

Add a Set node where you define your company variables such as:

• your_company_name = “Pollup Data Services”
• your_company_activity = “Whether it’s automating recurring tasks, analysing data faster, or personalising customer interactions…”
• your_email = “[email protected]”
• your_name = “Justin”

These values will be used for AI personalization.

Step 5: Get Twitter User ID via RapidAPI

Add an HTTP Request node named “Get twitter ID”. Set the HTTP method to GET with these configurations:

• URL: https://twitter-api47.p.rapidapi.com/v2/user/by-username
• Query Parameter: username={{ $json.twitter_handler }} (from the sheet)
• Headers include x-rapidapi-host and your RapidAPI key in the x-rapidapi-key header.

This fetches the Twitter numeric user ID needed for subsequent tweet retrieval.

Step 6: Retrieve Recent Tweets

Add another HTTP Request node called “Get tweets”. Configure it as a GET request to:

• URL: https://twitter-api47.p.rapidapi.com/v2/user/tweets
• Query Parameter: userId={{ $json.rest_id }} (from previous node)
• Headers as above with RapidAPI auth.

This gets the user’s recent tweets.

Step 7: Extract and Limit Tweets Using a Code Node

Add a Code node named “Exract and limit X” to parse the tweets JSON and limit results to the most recent 10 tweets. Use this JavaScript code:

output = []
max_posts = 10
let counter = 0
for (const item of $input.all()[0].json.tweets) {
 if(!item.content.hasOwnProperty('itemContent')) continue
 let post = {
 text: item.content.itemContent?.tweet_results?.result.legacy?.full_text
 }
 output.push(post)
 if(counter++ >= max_posts) break;
}
return {"Twitter tweets": output};

This ensures your AI prompt only gets relevant recent tweets.

Step 8: Fetch LinkedIn Posts via RapidAPI

Add an HTTP Request node named “Get linkedin Posts” to retrieve LinkedIn posts. Configure it as GET to:

• URL: https://fresh-linkedin-profile-data.p.rapidapi.com/get-profile-posts
• Query Parameter: linkedin_url={{ linkedin_url from Google Sheets }}
• Headers with RapidAPI credentials.

Step 9: Extract and Limit LinkedIn Posts

Add another Code node named “Extract and limit Linkedin” to clean and limit LinkedIn posts to 10:

output = []
max_posts = 10
let counter = 0
for (const item of $input.all()[0].json.data) {
 let post = {
 title: item.article_title,
 text: item.text
 }
 output.push(post)
 if(counter++ >= max_posts) break;
}
return {"linkedIn posts": output};

Step 10: Generate Personalized Subject and Cover Letter with AI

Add the Generate Subject and cover letter based on match node (LangChain LLM Chain). Provide the AI prompt incorporating your company info, lead’s name, and the extracted social posts in JSON format.

The instruction asks the AI to identify a common topic from the posts and generate an HTML formatted cover letter plus a subject line in JSON.

Step 11: Parse the AI Output

Add Structured Output Parser (LangChain Output Parser Structured) configured with a JSON schema example:

{
 "subject": "",
 "cover_letter": ""
}

This ensures the AI output is correctly parsed for email use.

Step 12: Use OpenAI Chat Model Node

Connect the Chat Model (GPT-4o) to power the AI generation in the previous step.

Step 13: Send Email with Generated Cover Letter and CC

Add an Email Send node configured to:

• Send HTML email body from AI output
• Subject as AI-generated subject
• Recipients: lead’s email and your own email (CC)
• From your configured SMTP “[email protected]”

This dispatches the personalized outreach automatically.

Step 14: Mark Lead as Processed in Google Sheets

Finally, add a Google Sheets node in update mode to set the “done” column to “X” for completed leads to avoid repeated emailing.

Customizations ✏️

1. Modify AI Prompt for Tone or Style
In the “Generate Subject and cover letter based on match” node, customize the prompt text to match your company’s voice or add specific frameworks like AIDA or PAS to improve persuasiveness.

2. Expand Social Platforms
Add additional HTTP Request nodes to incorporate social data from platforms like Instagram or Facebook if APIs are accessible, enriching lead insights.

3. Add More Lead Data Fields
Extend your Google Sheet with additional fields like job title or company size and pull them into the AI prompt for deeper personalization.

4. Adjust API Limits and Pagination
Modify HTTP Request nodes to paginate API results or respect rate limits based on your RapidAPI subscription tier.

5. Customize Email Templates
Change the HTML structure in the email send node to better fit branding or include dynamic data placeholders.

Troubleshooting 🔧

Problem: “No tweets found or empty response from Twitter API”
Cause: Incorrect Twitter handle or API quota exceeded.
Solution: Verify the Twitter username in Google Sheets is accurate and check your RapidAPI usage dashboard. Renew quota if needed by upgrading plans.

Problem: “LinkedIn posts not retrieved”
Cause: Invalid LinkedIn URL or API key issues.
Solution: Confirm LinkedIn URLs are complete (including https://) and your RapidAPI LinkedIn key is properly set in the HTTP request headers.

Problem: “AI output parsing errors”
Cause: AI might generate unexpected output format.
Solution: Refine your JSON schema in the Structured Output Parser node and test prompt modifications to ensure AI compliance.

Pre-Production Checklist ✅

• Validate Google Sheets data integrity and “done” column logic.
• Test API credentials and ensure RapidAPI keys have correct scopes and limits.
• Run test leads through the workflow to confirm data flows correctly from sheet to email.
• Backup your Google Sheet and n8n workflow definitions.

Deployment Guide

Activate your workflow by switching the Google Sheets Trigger node to active mode. Ensure all API keys and credentials are securely stored in n8n credential manager.

Monitor your workflow runs within n8n’s interface for errors or throttling issues.

Schedule periodic audits of API usage and update credentials as needed for seamless operation.

FAQs

Q: Can I replace RapidAPI with direct API calls?
A: While possible, RapidAPI simplifies authentication and usage. Direct calls require managing OAuth tokens and endpoints yourself.

Q: Is it safe to handle lead data this way?
A: Ensure your n8n instance is secure and access restricted. Data is transferred securely through HTTPS APIs.

Q: What are the API call limits?
A: Free RapidAPI plans offer limited calls monthly; monitor usage to avoid hitting limits.

Q: Can I use other AI models?
A: Yes, you can switch the OpenAI node credentials or select different models if you want to adjust cost or performance.

Conclusion

By following this guide, you created a powerful n8n workflow that automates personalized email outreach by analyzing Twitter and LinkedIn activities of potential leads. Justin at Pollup Data Services now spends far less time on manual research, ensuring more relevant and engaging communication.

This workflow saves at least 4 hours per week while improving lead response rates, boosting business growth. Next, consider integrating a CRM system for tracking or adding more social channels like Instagram for richer lead profiles.

Keep experimenting with AI prompts and personalization to make your outreach even smarter and more impactful.

Happy automating!