Auto-Enrich Your Lead List with LinkedIn Data
Enrich LinkedIn profiles from a Google Sheet. Pulls detailed profile data including job title, company, location, and more.
Last updated
February 4, 2026
Connectors used
Tags
Key Takeaways
- AI-powered profile matching - Uses web search, LinkedIn search, and website browsing to find the right profile
- Confidence scoring - Each match gets a 0 to 1 score indicating how certain the AI is
- Only writes high-confidence matches - Profiles below 70% confidence are filtered out
- Updates your existing spreadsheet - Writes linkedin_url, website, and confidence score back to your Google Sheet
- Processes only missing profiles - Skips rows that already have a LinkedIn URL
What This Workflow Does
This workflow reads a lead list from Google Sheets, filters for contacts that are missing a LinkedIn URL, uses AI to search the web and LinkedIn for each person, scores match confidence, and writes the results back to your spreadsheet. It is designed for enriching lead lists where you have names and companies but no LinkedIn profiles.
Use cases:
- Fill in missing LinkedIn URLs in your CRM export
- Enrich event attendee lists with profile links
- Research prospects before outreach
- Qualify inbound leads with LinkedIn data
How It Works
| Step | What Happens |
|---|---|
| 1. Manual trigger | You start the workflow |
| 2. Read Google Sheet | Pulls all rows from your lead list |
| 3. Filter missing profiles | Keeps only rows where linkedin_url is empty |
| 4. AI search for each lead | GPT-4.1 searches the web and LinkedIn using the person's name and context |
| 5. Filter by confidence | Only profiles with confidence above 70% proceed |
| 6. Update Google Sheet | GPT-4.1 finds the correct row and updates linkedin_url, website, and confidence score |
There is a 5-second delay between each profile update to avoid rate limits.
AI Models Used
| Model | Provider | Role |
|---|---|---|
| GPT-4.1 | OpenAI | Searches web and LinkedIn to find matching profiles |
| GPT-4.1 | OpenAI | Writes enriched data back to Google Sheets |
Tools Used in Search
The AI search node has access to multiple tools for finding profiles:
| Tool | What It Does |
|---|---|
| search_web | Searches the web for "[Name] [Company] LinkedIn" |
| browse_web | Visits websites to check footers, about pages, etc. |
| search_organization | Searches LinkedIn for the person's company |
| get_member_profile | Retrieves LinkedIn profile details |
Setup Instructions
- Click "Use template" on this page
- Copy the template Google Sheet linked inside the workflow
- Add your lead data to the sheet (name, company, and any other context you have)
- Paste your sheet URL into the Google Sheets read node and the update node
- Connect your Google Sheets account
- Connect your LinkedIn account
- Run the workflow
Google Sheets Columns Updated
| Column | Field |
|---|---|
| N | website |
| O | linkedin_url |
| P | confidence_score_for_profile |
The workflow reads from columns A through M for context (name, company, job title, etc.) and writes to columns N, O, and P.
Customization
| What You Can Change | How |
|---|---|
| Confidence threshold | Edit the filter code node (default filters below 0.69) |
| Search strategy | Modify the AI search prompt to adjust how it looks for profiles |
| Output columns | Change the column references in the update AI node |
| Delay between searches | Edit the wait node duration (default is 5 seconds) |
| Input sheet structure | Adjust the code node that parses rows to match your column layout |
FAQ
Q: How accurate is the profile matching? A: It depends on how unique the person's name is and how much context you provide. Adding company and job title significantly improves accuracy.
Q: Why is there a 70% confidence threshold? A: This balances automation with accuracy. Matches below 70% are more likely to be wrong. You can adjust this threshold in the filter code node.
Q: Can I adjust which columns are read and written? A: Yes. The code node that parses the spreadsheet and the AI node that writes back can both be modified to match your column layout.
Q: Does this work with non-English names? A: Yes, the AI search handles international names and characters.
Q: What if a person has a very common name? A: The AI uses all available context (company, job title, location) to narrow the search. Common names will generally produce lower confidence scores.
Want to showcase your own workflows?
Become a Needle workflow partner and turn your expertise into recurring revenue.
