Telegram Business Assistant
Telegram bot for order tracking, status updates, and delivery scheduling
Business Challenge
A logistics company was operating a 12-person call center just to handle order tracking inquiries. Customers would call to ask “Where is my package?”, “When will it arrive?”, and “Can I change the delivery time?” — questions that had answers in the system but required a human to look up and relay.
The call center was expensive to operate, couldn’t scale during peak shipping seasons, and was only available during business hours. Customers were frustrated when they couldn’t get updates on weekends or evenings, and the company was losing B2B clients who needed 24/7 tracking visibility for their own customers.
Solution
We built a Telegram bot that gives customers instant access to their order information. Customers can track packages by order number, get real-time status updates, schedule delivery windows, and even reroute packages to a different address — all through a conversational interface in Telegram.
The bot proactively sends notifications at key milestones: order confirmed, out for delivery, delivered, delivery exception. Customers can subscribe to specific orders and receive push notifications. For B2B clients, the bot supports batch queries — a warehouse manager can check the status of 50 orders in a single message.
Technologies
The bot uses the Telegram Bot API for messaging, LangChain for natural language understanding, FastAPI for the backend service, and Redis for session management and caching. The FastAPI service connects to the company’s existing order management system via internal APIs.
Redis plays a critical role in handling concurrent users and caching frequently accessed order data. The bot supports both Russian and English, with language detection and automatic switching. A webhook-based architecture ensures messages are processed in real-time, and a queue system handles traffic spikes during peak shipping periods.
Results
Call center volume dropped significantly within the first month as customers discovered they could get faster answers via Telegram. The bot is available 24/7, which eliminated the weekend and evening gap that was costing the company B2B clients.
Customer satisfaction improved notably in post-interaction surveys, with customers specifically praising the speed and convenience. The company was able to reduce the call center substantially, reallocating the savings to a dedicated customer success team for high-value B2B accounts. Two enterprise clients cited the Telegram bot as a key reason for renewing their contracts.
Screenshots
Customer tracking an order via Telegram bot
Bot usage analytics and call deflection metrics
Architecture
The system is built on a FastAPI backend that receives Telegram webhooks and processes messages through a LangChain pipeline. The pipeline classifies the message intent, queries the order management system, and generates a natural language response. Redis manages user sessions, caches order data with a 5-minute TTL, and handles rate limiting.
A notification service runs separately, listening to order management events and pushing proactive updates to subscribed Telegram users. The system is containerized and deployed with auto-scaling to handle peak shipping season traffic. A fallback mechanism routes complex queries to human agents through a shared dashboard.
Technologies used
Let's build something great together
Ready to automate your business processes and save hundreds of hours every month? Let's talk about your project.