Step 2: Connect Google Sheets

Connect Google Sheets so the workflow can append summary metrics and individual post records.

1. Open 📥 Append Metrics to Sheets and set Operation to append.
2. Set Document to [YOUR_ID] and Sheet to gid=0.
3. Map columns using the expressions: Total likes → {{ $json.totalLikes }}, Total posts → {{ $json.totalPosts }}, Average likes → {{ $json.averageLikes }}, Total comments → {{ $json.totalComments }}, Average comments → {{ $json.averageComments }}.
4. Credential Required: Connect your googleSheetsOAuth2Api credentials in 📥 Append Metrics to Sheets.
5. Open 📥 Log Posts in Sheets and set Operation to append, with Document [YOUR_ID] and Sheet gid=0.
6. Map columns using the expressions: date → {{ $json.date }}, likes → {{ $json.likes }}, video → {{ $json.videos }}, content → {{ $json.content }}, comments → {{ $json.comments }}, post link → {{ $json.post_link }}, Competitor → {{ $json.competitor }}, Post title → {{ $json.post_title }}.
7. Credential Required: Connect your googleSheetsOAuth2Api credentials in 📥 Log Posts in Sheets.
⚠️ Common Pitfall: Replace [YOUR_ID] with your actual Google Sheets document ID or the append operations will fail.

Step 3: Set Up Processing and AI Nodes

Configure the LinkedIn profile source, AI agent, tools, and parsing to extract structured post data.

1. In 🔗 Define LinkedIn Profile URL, set URL to https://www.linkedin.com/company/hubspot/.
2. Open 🤖 LinkedIn Post Fetcher Agent and set Prompt to =Scrape the below user profile on LinkedIn and get the latest 5 post data:{{ $json.URL }}.
3. 🧠 OpenAI Chat Engine is connected as the language model for 🤖 LinkedIn Post Fetcher Agent — Credential Required: Connect your openAiApi credentials in 🧠 OpenAI Chat Engine.
4. 🌐 Bright Data MCP Connector is connected as a tool for 🤖 LinkedIn Post Fetcher Agent — ensure credentials are added to the parent flow and not the tool itself. Credential Required: Connect your mcpClientApi credentials.
5. 🛠️ Auto-Correct Output Parser and 🧩 Structured Output Parser are connected as AI parsers for 🤖 LinkedIn Post Fetcher Agent — configure parsing schemas there and add credentials on the parent language model.
6. 🧠 OpenAI Chat Engine 2 is connected as the language model for 🛠️ Auto-Correct Output Parser — Credential Required: Connect your openAiApi credentials in 🧠 OpenAI Chat Engine 2.
7. In 🧩 Structured Output Parser, keep the JSON Schema Example as provided to ensure consistent output shape.
8. In 📈 Compute Engagement Stats, keep the JavaScript Code as-is to compute totals and averages from $json.output.
9. In 🧾 Split Post Items, keep the JavaScript Code that maps each post into a separate item using $('🤖 LinkedIn Post Fetcher Agent').first().json.output.
Tip: If the agent returns malformed JSON, the 🛠️ Auto-Correct Output Parser will attempt to repair it before the 🧩 Structured Output Parser validates the schema.

Step 4: Configure Output and Logging Flow

Confirm the workflow’s execution sequence and ensure metrics and post logs are appended in the correct order.

1. Verify the execution order follows: 🔘 Manual Launch Trigger → 🔗 Define LinkedIn Profile URL → 🤖 LinkedIn Post Fetcher Agent → 📈 Compute Engagement Stats → 📥 Append Metrics to Sheets → 🧾 Split Post Items → 📥 Log Posts in Sheets.
2. Ensure 📥 Append Metrics to Sheets receives totals and averages from 📈 Compute Engagement Stats.
3. Ensure 🧾 Split Post Items runs after 📥 Append Metrics to Sheets so individual posts are logged afterward.
4. Confirm 📥 Log Posts in Sheets appends each split item with date, likes, comments, content, and link fields.

Step 5: Test and Activate Your Workflow

Run a manual test to confirm LinkedIn data collection, stats calculations, and Sheet appends all succeed.

1. Click Execute Workflow to run 🔘 Manual Launch Trigger.
2. Verify 🤖 LinkedIn Post Fetcher Agent outputs a list of 5 posts in output.
3. Check that 📥 Append Metrics to Sheets creates a new row with totals and averages in your Google Sheet.
4. Check that 📥 Log Posts in Sheets appends multiple rows—one per LinkedIn post.
5. When satisfied, toggle the workflow to Active for production use.