TelMe

In today’s telecom market, TelMe reduces churn with an AI chatbot for seamless onboarding and ML-driven plan optimization, using K-Means clustering to enhance retention and meet customer needs.

Comment 3

Inspiration

The inspiration for TelMe arose from the need to simplify telecom customer support for both new and existing customers. New customers often require onboarding guidance, while existing customers need easy access to information specific to their accounts and services. TelMe was designed to provide a personalized experience for both groups, with a chatbot for new customers and a data-driven portal for existing users, making telecom services more accessible and efficient.

What it does

TelMe has two main components:

  1. Chatbot for New Customers: A conversational AI that guides new customers through telecom services. The bot helps them explore mobile, internet, and TV options, answering questions on plans, upgrades, and troubleshooting.
  2. Portal for Existing Customers: For current users, TelMe offers a portal that leverages pre-existing data to provide personalized insights, account details, and easy access to service options based on their profile.

How we built it

We built TelMe using Python, HTML, Flask, C, and Wix to create an intuitive chatbot and data portal. The chatbot’s front end was developed on Wix to leverage its customizable design features, making the interface accessible and visually appealing. Flask serves as the backend framework, handling interactions and integrating with a mock database for existing customer data. The chatbot’s decision trees guide new customers through various options, while the portal provides tailored information for existing users. This combination of technologies allowed us to create a dynamic, user-friendly platform for both customer types.

Challenges we ran into

  • Creating a chatbot conversation flow that feels natural and helpful for onboarding new customers.
  • Managing database integration for the existing customer portal to ensure data retrieval and display were efficient and user-friendly.
  • Developing and Launching a Dynamic Flask API on Replit.
  • Meticulously identifying and fixing issues within the code to ensure the functions work as intended.

Accomplishments that we're proud of

  • Dual-Feature System: Successfully developing a system that caters to both new and existing customers with an integrated chatbot and data-driven portal.
  • Intuitive Chatbot Interface: Leveraging HTML and Wix for a user-friendly front end that simplifies complex service information for new users.
  • Data-Driven Portal: Using Python and Flask to handle dynamic data processing and customer-specific information, which improves usability for current customers by providing tailored insights.

What we learned

  • Structuring a Flexible Chatbot: Using Flask to handle decision-making for chatbot responses, we learned to structure conversation flows that are user-friendly and adaptable.
  • Database Optimization & Data Handling: Through Python and mock data integrations, we recognized the importance of efficient data retrieval for real-time interaction, ensuring personalized support.
  • Balancing Multiple User Types: Combining HTML for flexibility on the front end and Flask for backend control, we developed strategies to maintain an optimized experience for both new and existing users within a single application.

What's next for TelMe

  • Expand TelMe’s chatbot capabilities to include more conversational AI elements for improved interactions.
  • Integrate advanced analytics within the portal to better understand and serve existing customer needs.
  • Explore more robust data security and authentication features to protect user data in the existing customer portal.

Built With

Share this project:

Updates