Step 2: Connect Airtable

The workflow searches, creates, and updates Airtable records for each LinkedIn profile.

1. Open Lookup Existing Record and select your Airtable Base and Table.
2. Set Operation to search and keep Filter By Formula as =({Name} = "{{ $json.profile_name }}").
3. Open Create Airtable Entry and select the same Base and Table.
4. Confirm the column mappings use expressions such as {{ $('Normalize Profile Data').item.json.profile_name }} for Name and {{ $('Normalize Profile Data').item.json.profile_linkedin_url }} for Profile URL.
5. Open Update Airtable Entry and select the same Base and Table, then confirm id is set to {{ $('Create Airtable Entry').item.json.id }}.

Credential Required: Connect your airtableOAuth2Api credentials to Lookup Existing Record, Create Airtable Entry, and Update Airtable Entry.

Step 3: Configure LinkedIn Data Fetching and Throttling

This section pulls reactions from Apify and throttles requests before iterating through profiles.

1. Open Fetch Post Reactions and set URL to https://api.apify.com/v2/acts/apimaestro~linkedin-post-reactions/run-sync-get-dataset-items?token=[CONFIGURE_YOUR_TOKEN].
2. Set Method to POST, enable Send Body, and set JSON Body to { "post_url": "[YOUR_ID]", "page_number": 1 }.
3. Keep the connection from Fetch Post Reactions to Rate Limit Pause to avoid API throttling.
4. Ensure Rate Limit Pause feeds into Iterate Records so each reaction is processed in batches.

Step 4: Normalize and Enrich Profile Data

This step formats reaction data, checks for existing records, and enriches profile details.

1. In Normalize Profile Data, confirm each field uses the correct expressions, such as {{ $json.reactor.name }} for profile_name.
2. Connect Normalize Profile Data to Lookup Existing Record, then to Presence Check for record existence logic.
3. In Enrich Profile Data, set the URL to https://api.apify.com/v2/acts/dev_fusion~linkedin-profile-scraper/run-sync-get-dataset-items?token=[CONFIGURE_YOUR_TOKEN].
4. Set JSON Body to { "profileUrls": ["{{ $('Normalize Profile Data').item.json.profile_linkedin_url }}"] } to enrich each profile.
5. Keep the route from No-Op Placeholder to Iterate Records as it acts as a passthrough for existing records.

Step 5: Build the Profile Aggregation and Delay Gates

These nodes consolidate profile fields and insert random delays to respect external API limits.

1. In Combine Profile Fields, keep all Fields to Aggregate such as fullName, jobTitle, and experiences[0].title.
2. In Generate Delay Value, keep the JavaScript code as provided to return a random delay in seconds.
3. Set Pause for Delay Value Amount to {{ $('Generate Delay Value').item.json.seconds }}.
4. Ensure Validity Gate 2 checks for non-empty JSON using {{ $json && Object.keys($json).length > 0 }} before data is mapped.

Step 6: Configure the AI Qualification and Mapping

Here, the workflow composes scraper output, evaluates ICP fit, and prepares structured results.

1. In Map Scraper Output, set linkedin_scraper_out to the multi-line expression that joins fields from Combine Profile Fields, starting with {{ $('Combine Profile Fields').item.json.fullName[0] }}.
2. Open AI ICP Evaluation and keep the Text prompt as defined; it references {{ $('Map Scraper Output').item.json.linkedin_scraper_out }} for input.
3. Attach Structured Result Parser as the output parser for AI ICP Evaluation and keep the JSON schema example.

Credential Required: Connect your LLM credentials in AI ICP Evaluation (the model is configured on the parent node, not on Structured Result Parser).

Step 7: Configure Creation and Updates After AI Scoring

New profiles are created in Airtable, delayed, enriched, scored by AI, and then updated.

1. From Create Airtable Entry, keep the flow to Generate Random Delay to add randomized wait time after a new record.
2. In Delay by Random Time, set Amount to {{ $('Generate Random Delay').item.json.seconds }}.
3. Confirm Update Airtable Entry sets MSP? to {{ $('AI ICP Evaluation').item.json.output.msp }} and Reason to {{ $('AI ICP Evaluation').item.json.output.reasoning }}.
4. Keep the update of Email Address to {{ $('Enrich Profile Data').first().json.email }}.

Step 8: Test and Activate Your Workflow

Run a manual test to verify the API calls, Airtable operations, and AI evaluation results.

1. Click Execute Workflow starting from Manual Start Trigger.
2. Confirm Fetch Post Reactions returns a list of reactors and that Iterate Records processes them.
3. Verify new entries appear in Airtable from Create Airtable Entry and that Update Airtable Entry fills MSP?, Reason, and Email Address.
4. Once successful, toggle the workflow to Active for production use.