Turn web search into clean insights you can use right away. The flow collects results, converts them into structured JSON, and creates a short summary that you can send to any endpoint. It fits teams that need quick market research or content briefs without manual copy and paste.

The run starts on manual click, calls a Bing search through a Bright Data scraper request, and stores the snapshot id. It checks snapshot progress on a loop with a wait step, then downloads the data when ready. A Google Gemini model formats the content into a defined JSON schema using an output parser, while a summarization chain splits long text and writes a concise recap. Two HTTP requests send both the structured dataset and the summary to your chosen webhooks.

Set up needs a Bright Data API key, a Google Gemini API key, and your webhook URLs. Update the notifier URLs, confirm header credentials on the Bright Data calls, and adjust the schema if your fields differ. Expect faster research cycles and fewer errors, especially for list building, location analysis, and content planning. Most teams see hours of manual work drop to minutes per run.

• Manual start for controlled research runs and testing
• HTTP request triggers a Bright Data scraper search and returns a snapshot id
• Status polling with If checks and a 30 second Wait to avoid timeouts
• Snapshot download returns JSON data ready for processing
• Structured Output Parser enforces a clear JSON schema for fields like city and hotels
• Google Gemini model formats and extracts structured data
• Summarization chain with text splitter builds a concise summary from long content
• Two webhook notifiers send the structured dataset and the summary to external endpoints
• Error path and basic checks keep runs stable during network issues

• Reduce manual research from 3 hours to 10 minutes
• Automate 80% of copy and paste tasks in web research
• Improve data quality by 90% with structured parsing
• Handle 10 times more queries with snapshot polling
• Connect Bright Data, Bing, Gemini and your webhook in one flow

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with Bright Data, Google Gemini, Microsoft Bing and Webhook.site. See the Tools Required section above for links to create accounts with these services.
3. In the n8n credentials manager, create a new HTTP Header Auth credential for Bright Data. Use a header name like Authorization and set the value to your token as provided by Bright Data. Save and name it clearly.
4. Open the Perform a Bing Copilot Request node and select your Bright Data header credential. Confirm the request URL and headers match your Bright Data scraper configuration.
5. Open the Check Snapshot Status and Download Snapshot nodes. Choose the same Bright Data header credential. Confirm the format parameter is set to json and keep the timeout values reasonable.
6. In the credentials manager, create a Google Gemini credential. Choose Google Gemini PaLM API and paste your API key from Google AI Studio. Save and name it for easy reuse.
7. Open the Google Gemini Chat Model and Google Gemini Chat Model1 nodes and select your Gemini credential. Keep the provided models or switch to a model your account supports.
8. Adjust the Structured Output Parser schema if you need different fields. Match the schema to the data your team wants, such as city, hotel name, address, and website.
9. Set the webhook URLs in the Structured Data Webhook Notifier and Summary Webhook Notifier nodes. For quick tests, create a temporary endpoint in Webhook.site and paste the URL.
10. Click Test workflow to run. Watch the execution map. The If and Wait nodes will poll until the snapshot is ready, then the data will be parsed and summarized.
11. Verify both webhook endpoints receive the JSON and the summary. If nothing arrives, check the Check on the errors branch, review headers, and increase the wait time if your snapshot needs longer.