Airtel XLabs: Transformed by Data

Airtel is world’s third largest telco with ~400 million subscribers in 16 countries and provides mobile telecom, fixed line and WiFi broadband, voice, DTH, services to both consumers and businesses. It is building a growing portfolio of consumer internet and fintech properties; Airtel Payments Bank, Airtel TV, Wynk Music etc.

Today we generate 100’s of trillions of records annually from calls, network, apps, IoT devices, phones, GPS and more. Very few entities in India, or globally, come even within a (lower) orders of magnitude of this volume of data.

The approach to solving critical business problems for Airtel is similar to creating 3-layered cake. We need to understand, measure and maximize the exigency of personalization and engagement of large base of customers in an omnichannel environment using optimized intelligent network.

Figure 1: Solve for the Omnichannel Customer Engagement problem

Within the aegis of these business requirements, here are samples of specific problems we are solving:


  • Customer 360: Build single customer view combining all Lines of Business (LOB) — Mobile, Voice, Broadband, DTH, as well as apps. This requires solving one of the hardest “India Class” problems — disambiguating myriad forms of addresses that a person uses for same location or house. Once the customer views are built, create customer segmentations using collaborative filtering and deep learning techniques
  • Chatbot Customer Support: Solve for 100’s of types of customer queries and support in Hindi, English and “Hinglish” chat (existing). Extend to regional languages, both in chat and eventually, for voice calls to support centers

Customer Engagement:

  • “Live, Work and Play”: Creating a comprehensive Customer Satisfaction Index involves understanding a customer’s multi-product (mobile, DTH, broadband, voice and more), multi-device (TV, smart TV, phone, tablets), multi-location (home, work and in transit) usage (voice calls, watching Netflix). This is a complex problem as customer choices keep exploding and new devices are being added to the market at an ever-increasing rate
  • Recommendation Engines: Based on customer views, segments and past actions, build a recommendation engine that can suggest the next best actions (“pay bills” or “watch Bahubali 2”), or, products or services (“buy bundle that includes insurance for your new Samsung Galaxy 9 Plus)
Figure 2: Array of Customer, Engagement and Network problems we solve

Intelligent Network:

  • Network Optimization: Providing optimum network experience to 400 million customers making over tens of billions of calls, consuming tens of petabytes of content on over billion devices is one of the most complex challenges one can encounter
  • Smartphone Characterizations: Ultimately, the devices the consumers own form an integral part of the quality of the calls or data consumptions. We collect a few trillion data points every month to characterise the top 150 devices in the market. Understanding of the data and statistical modeling is key for arriving at Key Performance Indicators (KPIs). Eventually this information is shared internally and device manufacturers to improve the consumer experience.

Our Tech Stack:

Here, the picture is truly worth a thousand words. In short, the usual big data stack: and we really process “big” amount of data with this infra

  • 2 to 5 trillion records per day
  • > 10 billion calls per month
  • 10’s of petabytes of data per month
Figure 3: Data Tech Stack
How can you be a part of this journey?

An industry disrupted by Reliance Jio leading to a 3-player market for world’s largest subscriber base, plus the upcoming 5G networks and emergence of AI, IoT, AR/ VR will enable us to re-imagine and re-shape the world, and make significant business impacts in the next 18–24 months.

The time for getting involved in this historic journey is right now. If you love all things data and digital, ping us.