N8N Financial Tracker: Telegram Invoices to Notion with AI Summaries & Reports

Workflow Name

Key Features and Highlights

This workflow implements an automated financial tracking system that receives invoice images via Telegram, utilizes the Google Gemini AI model for text recognition and data extraction, automatically parses detailed expense information from invoices, and saves the structured transaction data into a Notion database. Additionally, the workflow supports periodic aggregation of transaction data, generates visual expenditure reports and charts, and automatically sends these reports through Telegram groups or private chats, completing a fully closed-loop automation process from data collection to report generation and distribution.

Core Problems Addressed

Manual entry of invoices and expense details is tedious and error-prone
Financial data from multiple channels is difficult to unify and analyze
Lack of automated tools for data aggregation and report generation
Need for convenient financial status viewing through familiar chat tools

Use Cases

Daily expense and invoice management for individuals or small teams
Automated financial tracking for entrepreneurs and freelancers
Expense reimbursement and spending aggregation for remote or distributed teams
User groups requiring regular financial report push notifications

Main Workflow Steps

Invoice Image Reception: Monitor and receive invoice or receipt photos sent by users via Telegram Bot.
Image Information Extraction: Retrieve image file metadata to prepare for subsequent processing.
AI Data Extraction: Invoke the Google Gemini language model with custom prompts to parse invoice content, identifying product names, quantities, prices, taxes, categories, and other details.
Structured Conversion: Transform the AI-generated text output into objects conforming to a predefined JSON schema for easier downstream processing.
Data Splitting and Storage: Split transaction records into individual line items and write each entry into the Notion database.
User Feedback: Send brief transaction summary confirmation messages back to users via Telegram.
Scheduled Triggering: Periodically fetch recent transaction data from Notion according to preset time rules.
Data Aggregation: Summarize and total expenses by category.
Chart Generation: Create expenditure bar charts based on aggregated data.
Report Delivery: Automatically send the generated chart images to designated Telegram groups or private chats.

Involved Systems and Services

Telegram: Serves as the input interface for receiving invoice images and sending messages.
Google Gemini (PaLM API): Provides AI-powered invoice text recognition and structured parsing.
Notion: Stores and manages the financial transactions database.
n8n Built-in Nodes: Including time triggers, code nodes, and chart generation nodes to implement automation control and data processing.

Target Users and Value Proposition

Individual Users and Freelancers: Simplify daily financial record-keeping and reduce manual input workload.
Small Businesses and Teams: Centralize team expense management and enhance financial transparency.
Financial Managers: Quickly obtain summarized expenses and visual reports to support decision-making.
Automation Enthusiasts and Developers: Customize AI prompts and database structures for flexible feature expansion.

By integrating AI-powered intelligent recognition with robust automation services, this workflow creates an efficient, intelligent, and user-friendly financial tracking and reporting system that significantly improves the efficiency and accuracy of financial data processing.