Auto-Scrape & Enrich LinkedIn Leads with n8n (Apollo + RapidAPI + GPT-3.5)

Input job title & location → Apollo search → extract LinkedIn username → verify email → RapidAPI profile & posts → GPT-3.5 summaries → write to Google Sheets

Who

Sales teams, founders, B2B marketers who need bulk LinkedIn lead enrichment

You get

Automated lead list with verified emails, LinkedIn profile summaries, and recent post summaries ready for personalized cold outreach

You need

n8n · google-sheets

Steps

1. Search Apollo and ingest leads

   Triggered by form or webhook, call Apollo.io API to search leads by job title and location. Use HTTP Request node POST to https://api.apollo.io/api/v1/mixed_people/search with person_locations, person_titles, per_page. Split Out node splits people array, Clean Data extracts id, name, title, linkedin_url, organization, then append to Google Sheets with status columns initialized to pending/unscraped. [First-test safety] Before the first run, disconnect or retarget the Google Sheet write nodes (Add Leads / Update Profile Summary / Update Posts Summary / Append to Enriched Leads Database) to a test sheet, or use a test API key; only reconnect production writes after extraction and summaries look correct.

   Execution prompt · Source text。HTTP Request node 'Generate Leads with Apollo.io' JSON Body

   Copy prompt

   ={
     "person_locations": ["{{ $json.Location }}"],
     "person_titles": ["{{ $json['Job Title'] }}"],
     "page": 1,
     "per_page": {{$json['Number of Leads']}},
     "projection": ["id", "name", "linkedin_url", "title"]
   }

2. Extract LinkedIn username

   Google Sheets Trigger fires on new row, fetches row with extract_username_status=pending. Workflow OpenAI node "OpenAI1" strips the http(s)://www.linkedin.com/in/ prefix from linkedin_url to yield the username (the listing description says "script node", but the attached workflow.json uses OpenAI1). Writes username back to sheet and sets extract_username_status to finished.

   Execution prompt · Source text。OpenAI node 'OpenAI1' messages.content

   Copy prompt

   =remove the http or https://www.linkedin.com/in/ from this  {{ $json.linkedin_url }}

3. Fetch and validate email

   Google Sheets Trigger detects contacts_scrape_status=pending. HTTP Request node "Get Email from Apollo" POSTs to https://api.apollo.io/api/v1/people/match (body id={{ $json.apollo_id }}, query reveal_personal_emails=true) for email; then HTTP Request node "Confirm Email Validity" calls https://api.mails.so/v1/validate?email={{ $json.person.email }} with header x-mails-api-key, filtering on undeliverable and mx_record. Valid emails are written to the sheet with status finished; invalid marked invalid_email.

4. Enrich profile summary (RapidAPI primary)

   Google Sheets Trigger detects profile_summary_scrape=pending. HTTP Request node calls RapidAPI LinkedIn Data API for profile details (summary, headline, education, position). Code node cleans data, Stringify node serializes to JSON, then OpenAI generates personalized profile summary. Writes back to sheet as completed. Failures marked failed, retried by Schedule Trigger.

   Execution prompt · Source text。OpenAI node 'AI Profile Summarizer' messages.content

   Copy prompt

   =Please summarize the following linkedin profile data {{ $json.profileString }} for a lead I want to cold email. I want to combine this summary with another information about them to send personalied emails, so please make sure you include relevant bits in the summary

5. Summarize recent posts (RapidAPI primary)

   Google Sheets Trigger detects posts_scrape_status=unscraped. HTTP Request node calls RapidAPI for recent posts/reposts. Code node "Clean Posts Data" extracts post text and dates, Stringify serializes, then OpenAI generates a thematic summary (max 2 paragraphs). Writes back to sheet as scraped; failures marked failed and retried by Schedule Trigger.

   Execution prompt · Source text。OpenAI node 'Posts AI Summarizer' messages.content

   Copy prompt

   =Below are the most recent posts and reposts from a LinkedIn user. Summarize them collectively in no more than two short paragraphs. Focus on capturing the main themes, tone, and any recurring interests or professional concerns.
   
   Avoid listing each post separately — instead, synthesize the information into a narrative that gives a clear idea of what this person is currently focused on or passionate about.
   
   Posts: {{ $json.postsString }}
   
   Keep it insightful but brief — no more than 2 concise paragraphs.

6. Consolidate to enriched database

   When a row satisfies contacts_scrape_status=finished AND profile_summary_scrape=completed AND posts_scrape_status=scraped, Google Sheets Trigger fires and appends/updates the lead's name, email, title, company, profile summary, and posts summary to the separate Enriched Leads Database sheet.

You are an n8n workflow setup assistant. The user wants to replicate a LinkedIn lead scraping and enrichment workflow. Assist in this order: 1) Confirm user has n8n instance, Google Sheets OAuth2, Apollo.io Master API Key, RapidAPI LinkedIn Data API Key, Mail.so API Key, OpenAI API Key; 2) Guide importing the workflow JSON or manual setup; 3) Explain Google Sheets template structure (status columns: extract_username_status, contacts_scrape_status, profile_summary_scrape, posts_scrape_status); 4) Remind to configure API keys as Generic Credentials, not hardcoded; 5) If Apify alternative is needed, guide configuring the actor endpoint.