Step 2: Set Up AI City and Category Generation

Two AI steps generate city lists and classify the service type. Each agent uses a connected Anthropic model.

1. Open City LLM Provider and select your Anthropic model. Credential Required: Connect your anthropicApi credentials.
2. Open Generate City Roster and set Prompt to the existing expression: =Provide a list of cities or sub areas located inside the Location of {{ $json.State }}, within the country {{ $json.Country }}. This data will be used to search on Google maps for businesses in that region. The output should be a plain text list, without bullets, numbering, or any special characters. Do not include any introduction, explanation, or concluding text—only the list of city names.
3. Open Category LLM Provider and select your Anthropic model. Credential Required: Connect your anthropicApi credentials.
4. Open Classify Service Type and set Prompt to: =Determine the appropriate category based on the following input: "{{ $('Intake Form Trigger').item.json.Service }}" The output must contain only the category name that best describes the given service. Examples of categories include but are not limited to: Electronics, Healthcare, Education, Food & Beverage, Automotive, Finance, Real Estate, etc.

City LLM Provider is connected as the language model for Generate City Roster and Category LLM Provider is connected to Classify Service Type — ensure credentials are added to those provider nodes (not the agent nodes).

Step 3: Split and Batch City Records

The workflow turns the AI city list into structured items and batches them for downstream processing.

1. Open Split Records by City and keep the existing JavaScript that reads Service, State, Country, and the AI city list to build per-city items.
2. Ensure Batch Cities Iterator is connected after Split Records by City to process cities in manageable chunks.
3. Confirm the execution path: Classify Service Type → Split Records by City → Batch Cities Iterator.

Batch Cities Iterator outputs to both Retrieve Sheet Records and Maps Data Scraper in parallel.

Step 4: Configure Google Sheets Deduplication

This section checks for duplicates and deletes existing rows before adding new leads.

1. Open Retrieve Sheet Records and set Document ID to YOUR_GOOGLE_SHEET_ID and Sheet Name to gid=0. Credential Required: Connect your googleSheetsOAuth2Api credentials.
2. In Detect Duplicate Records, keep the JavaScript that compares Business Name, Phone Number, Address, and Google Maps URL to output duplicates.
3. Open Locate Duplicate Row and confirm filters: Address uses ={{ $json.Address }} and Google Maps URL uses ={{ $json['Google Maps URL'] }}. Credential Required: Connect your googleSheetsOAuth2Api credentials.
4. Open Remove Duplicate Row and set Operation to delete with Start Index ={{ $json.row_number }}. Credential Required: Connect your googleSheetsOAuth2Api credentials.

Step 5: Configure Scraping and Status Polling

The workflow scrapes Google Maps via Bright Data and polls until results are ready.

1. Open Maps Data Scraper and keep the URLs expression: =[ { "url": "https://www.google.com/maps/search/{{ $json.name }}+in+{{ $json.city }} {{ $json.state }}", "category": "{{ $json.category }}", "country_name": "{{ $json.country }}", "sample": false } ] . Credential Required: Connect your brightdataApi credentials.
2. Open Verify Scrape Response and keep the condition checking ={{ $json.message }} exists.
3. Open Query Collection Progress and set URL to =https://api.brightdata.com/datasets/v3/progress/{{ $json.snapshot_id }}. Credential Required: Connect your httpHeaderAuth credentials.
4. Open Throttle Pause and set Amount to 25 seconds to control polling frequency.
5. Open Confirm Data Readiness and keep the status check ={{ $json.status }} equals ready.
6. Open Download Scrape Results and set URL to =https://api.brightdata.com/datasets/v3/snapshot/{{ $json.snapshot_id }} with query parameter format=json. Credential Required: Connect your httpHeaderAuth credentials.

Step 6: Append Results to Google Sheets

Scraped results are mapped into your spreadsheet fields.

1. Open Append Leads to Sheet and confirm Operation is append. Credential Required: Connect your googleSheetsOAuth2Api credentials.
2. Verify the column mappings use expressions like City ={{ $('Split Records by City').item.json.city }}, Business Name ={{ $json.name }}, and Google Maps URL ={{ $json.url }}.
3. Confirm the flow Download Scrape Results → Append Leads to Sheet → Batch Cities Iterator to continue processing remaining cities.

Step 7: Test and Activate Your Workflow

Run a manual test to validate form input, AI output, scraping, and sheet updates before activating.

1. Click Execute Workflow and submit the Intake Form Trigger with a known service (e.g., Laptop Store) and location.
2. Watch for successful data passing from Generate City Roster and Classify Service Type into Split Records by City.
3. Confirm Download Scrape Results returns JSON and Append Leads to Sheet appends rows with city, business name, address, and URLs.
4. Once verified, toggle the workflow to Active to accept live form submissions.