This AI System Scrapes Amazon Best Sellers (NO-CODE Tutorial)

Key Concepts

• No-code automation: Building automated workflows without writing code.
• Web scraping: Extracting data from websites.
• AI summarization: Using AI to condense and extract key information from large datasets.
• Bright Data: A web data platform used for proxy and web scraping services.
• NAND: A no-code platform for building automations.
• OpenAI: An AI research and deployment company, used here for text extraction and summarization.
• Google Sheets integration: Connecting the automation to Google Sheets for data storage and organization.
• HTTP Request Node: Used to make requests to external APIs, in this case, Bright Data.
• Set Node: Used to define variables and parameters within the workflow.
• Code Node: Used to execute custom JavaScript code for data manipulation.
• Split Item Node: Used to split a single data item into multiple items.

Automation Workflow for Identifying Top Selling Amazon Products

1. Introduction

The video demonstrates how to automate the process of identifying top-selling products on Amazon using AI and no-code tools. The goal is to create a system that automatically scrapes Amazon's bestseller lists, summarizes the data, and delivers it to the user via email or Google Sheets on a daily basis. This automation can be used by businesses to stay competitive or by individuals to identify profitable product niches.

2. Setting up the Workflow in NAND

• Creating a New Workflow: The video starts by creating a new workflow in NAND, a no-code automation platform. A link to NAND is provided in the description for a 14-day free trial.
• Importing a Pre-built Workflow (Optional): The video demonstrates how to import a pre-built workflow from the creator's community using a JSON file. This is presented as a shortcut for community members. The JSON file is downloaded from the community and imported into NAND using the "Import from file" option.
• Manual Trigger: For demonstration purposes, a manual trigger is used. In a production environment, a schedule trigger would be used to run the automation daily.

3. Scraping Amazon Data with Bright Data

• Creating a Bright Data Account: The video explains how to create a Bright Data account for web scraping. A link to Bright Data is provided in the description.
• Setting up Web Unlocker API: The video guides the user through setting up the Web Unlocker API in Bright Data, which is used to bypass anti-scraping measures.
• Creating a Zone: A zone is created in Bright Data to manage the web scraping requests. The zone name is used in the NAND workflow.
• Adding a Set Node: A set node is added to the NAND workflow to define the Bright Data zone and the Amazon URL to scrape.
  • The "zone" variable is set to the name of the Bright Data zone.
  • The "URL" variable is set to the Amazon bestseller URL for a specific category (e.g., "Home & Kitchen").
• Adding an HTTP Request Node: An HTTP Request node is added to send a request to the Bright Data API.
  • The method is set to "POST".
  • The URL is set to api.brightdata.com/request.
  • Authentication is set to "Generic Credential Types" with "Authorization" as the header and "Bearer [API Token]" as the value. The API token is obtained from the Bright Data account.
  • The body is set to include the "zone" and "URL" variables from the set node, as well as format: raw.

4. Extracting and Summarizing Data with AI

• Adding a Code Node: A code node is added to extract a sample of the HTML content from the scraped data. This is done to reduce the amount of data processed by the AI. The code used is provided in the video description or community.
• Adding an Information Extractor Node (OpenAI): An Information Extractor node is added to extract the top 10 Amazon bestsellers from the HTML sample.
  • The text input is set to the output of the code node ("partial HTML").
  • A schema is defined to specify the data to extract (e.g., title, stars, total ratings, offer, product URL, image URL). The schema is provided in the video description or community.
  • A chat model (e.g., GPT-4.0 Mini) is selected for the OpenAI node.

5. Outputting Data to Google Sheets

• Adding a Split Item Node: A Split Item node is added to split the output of the Information Extractor node into individual items. This is necessary for adding the data to Google Sheets.
• Adding a Google Sheets Node: A Google Sheets node is added to append the extracted data to a Google Sheet.
  • The Google Sheet credentials are connected.
  • The resource is set to "Sheet within document".
  • The operation is set to "Append or update row from list".
  • The spreadsheet and sheet are selected.
  • The columns are mapped to the corresponding data fields from the Split Item node.
• Google Sheets Template: A Google Sheets template is provided in the video description or community to ensure that the columns match the data fields in the workflow.

6. Optional: Sending Data to Slack

• Adding an OpenAI Node for Summarization: An OpenAI node is added to summarize the top 10 products for sending to Slack.
  • A prompt is used to instruct the AI to summarize the data.
  • A chat model (e.g., Nano) is selected.
• Adding a Slack Node: A Slack node is added to send the summarized data to a Slack channel.

7. Testing and Deployment

• Testing the Workflow: The workflow is tested to ensure that it is working correctly.
• Scheduling the Workflow: The workflow can be scheduled to run automatically on a daily basis.

8. Changing the Amazon Category

• To change the Amazon category, the user simply needs to update the URL in the set node with the URL of the desired category's bestseller list.

9. Conclusion

The video provides a step-by-step guide to automating the process of identifying top-selling products on Amazon using AI and no-code tools. The automation can be used by businesses to stay competitive or by individuals to identify profitable product niches. The video emphasizes the importance of using Bright Data for web scraping and OpenAI for data extraction and summarization. The video also promotes the creator's community, which provides additional resources and support for building automations.

Notable Quotes

• "This is going to be a no code solution regardless of your background even if you have never done anything we this will make sense to you because I'm going to walk through step by step and show you how to build this thing."
• "All you have to do is just go and change the URL inside uh this set node over here and it will scrape that."

Technical Terms and Concepts

• API (Application Programming Interface): A set of rules and specifications that software programs can follow to communicate with each other.
• JSON (JavaScript Object Notation): A lightweight data-interchange format that is easy for humans to read and write and easy for machines to parse and generate.
• HTML (HyperText Markup Language): The standard markup language for creating web pages.
• Node: A building block in NAND workflows that performs a specific task.
• Workflow: A series of connected nodes that automate a process.
• Schema: A structure that defines the format and data types of data.
• Prompt: A text input that instructs an AI model to perform a specific task.

Logical Connections

The workflow is structured logically, starting with the trigger, then setting up the data source (Amazon URL and Bright Data), scraping the data, extracting and summarizing the relevant information using AI, and finally outputting the data to Google Sheets or Slack. Each step builds upon the previous one, creating a seamless automation process.

Data and Statistics

The video does not provide specific data or statistics, but it emphasizes the importance of using data to make informed business decisions. The automation provides users with access to real-time data on top-selling products, which can be used to identify trends and opportunities.