Track Expenses from Telegram to Google Sheets with AI Analysis

This n8n workflow automates the process of tracking expenses by capturing messages from Telegram, analyzing them with an AI agent (GPT-41-mini), and then recording the structured expense data into a Google Sheet. It also provides feedback to the user via Telegram.

What it does

This workflow streamlines expense tracking through the following steps:

1. Listens for Telegram Messages: It acts as a Telegram bot, waiting for incoming messages from a user.
2. Analyzes Message Content with AI: The received message is fed into an AI agent (configured with an OpenAI Chat Model and a Structured Output Parser). The AI is tasked with extracting key expense details like amount, currency, category, and a description.
3. Parses AI Output: The AI's response, which is expected to be in a structured format (e.g., JSON), is then parsed to make the data usable.
4. Conditional Processing: It checks if the AI successfully extracted a valid amount.
   • If an amount is found: The extracted expense details (date, amount, currency, category, description) are appended as a new row to a specified Google Sheet. A confirmation message is sent back to the user on Telegram.
   • If no amount is found: A message is sent back to the user on Telegram, asking them to provide a valid expense amount.
5. Provides User Feedback: In both success and failure scenarios, the workflow communicates back to the user via Telegram, confirming the expense recording or requesting more information.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running instance of n8n.
• Telegram Bot Token: A Telegram bot set up with a BotFather token.
• OpenAI API Key: An API key for OpenAI (or a compatible service) to power the AI Chat Model.
• Google Account: Access to a Google account with a Google Sheet prepared for expense tracking.
  • The Google Sheet should have columns corresponding to the data you expect to extract (e.g., Date, Amount, Currency, Category, Description).

Setup/Usage

1. Import the Workflow: Import the provided JSON into your n8n instance.
2. Configure Credentials:
   • Telegram Trigger & Telegram Node: Set up your Telegram Bot API credential.
   • OpenAI Chat Model Node: Set up your OpenAI API Key credential.
   • Google Sheets Node: Set up your Google Sheets API credential (OAuth 2.0 is recommended).
3. Configure Nodes:
   • Telegram Trigger: Ensure it's configured to listen for messages from your desired chat(s).
   • AI Agent:
     • The "AI Agent" node will likely need instructions (a "prompt") on how to parse expense information from user messages. For example: "Extract the amount, currency, category, and a brief description from the following expense message. Output in JSON format with keys: 'amount', 'currency', 'category', 'description'."
     • The "OpenAI Chat Model" node should be configured with your preferred model (e.g., gpt-4-mini if available, or gpt-3.5-turbo).
     • The "Structured Output Parser" node should be configured with the expected JSON schema that the AI agent will output.
   • Code Node: This node will contain JavaScript logic to check if the AI successfully extracted an amount and to format the data for Google Sheets. You might need to adjust the logic to match the exact output of your AI agent.
   • Google Sheets Node:
     • Specify the Spreadsheet ID and Sheet Name where you want to record expenses.
     • Ensure the "Operation" is set to "Append Row" and map the incoming data fields (e.g., {{ $json.amount }}, {{ $json.category }}) to the correct columns in your Google Sheet.
   • Telegram Nodes (for responses): Customize the success and failure messages sent back to the user.
4. Activate the Workflow: Once configured, activate the workflow.
5. Start Tracking: Send expense messages to your Telegram bot (e.g., "Spent $50 on groceries", "Lunch for 15 EUR at cafe"). The workflow will process them and update your Google Sheet.