THE CHALLENGE
Transitioning away from SQL to unlock scale and flexibility
The ability to efficiently scale and manage an immense quantity of data has been central to digitizing Hindustan Times’s operations and expanding its services to the fintech world.
Before working with MongoDB, Hindustan Times was using a legacy SQL database to manage its Blank Paper CMS and other parts of its media content and publishing ecosystem. Modernizing the organization’s data infrastructure presented five key challenges:
- Data complexity and scale: Hindustan Times’s vast datasets—comprising tens of thousands of articles generated every day, Personal Identifiable Information (PII), and clickstream data from 15 news sites—had become extremely complex and difficult to manage with a SQL database. As a result, inefficiencies were hindering business operations and user experience.
- Flexibility: Hindustan Times required a database with high flexibility and adaptability to accommodate the daily addition of new content and dynamic metadata. While some of its articles were mostly text-based, others could have up to 700 different fields, seven indexes, and a considerable amount of embedded metadata.
- Time to market: The burden of manual management—including manual scaling of servers, frequent database downtimes, and intricate back-up procedures— took a toll on Hindustan Times’s developer velocity.
- Data security compliance: The use of PII data for both media and personal finance services mandated that Hindustan Times had to comply with stringent security and privacy requirements. Security could not be an optional add-on, it was a prerequisite.
- AI-readiness: Hindustan Times wanted to explore the potential of large language models (LLMs) and artificial intelligence (AI). However, its legacy SQL platform was not built to support the AI capabilities required to become a modern digital media and fintech powerhouse.
“Over the years, data has become core to our business and operations,” said Amit Verma, Chief Technology Officer, Hindustan Times. “We needed to move away from the relational database model and find a modern, agile database capable of addressing high concurrency needs, simplifying maintenance, and offering seamless scale.”
THE SOLUTION
MongoDB Atlas removes data complexity and simplifies operations
In 2016, Hindustan Times adopted MongoDB Community Edition. Moving from SQL to the document database model immediately helped the organization address some of the challenges it was facing related to data complexity and flexibility.
As the digital ecosystem matured and more scale was needed, Hindustan Times decided to move to an alternative managed NoSQL database in late 2023. However, the migration was difficult and the database did not deliver the same functionality and performance that MongoDB had delivered. This led Hindustan Times back to MongoDB, which the team loved using and trusted from past experience. Hindustan Times successfully migrated to MongoDB's native managed database MongoDB Atlas—hosted on AWS—in August 2024.
MongoDB Atlas’s flexible schema design and agility enabled Hindustan Times to meet the demands of its complex CMS and data-rich editorial content. Blank Paper was one of the core applications migrated to the cloud, which unlocked seamless publishing and editing across all 15 websites.
“With MongoDB, we can add and remove fields based on different requirements. We can also effortlessly manage and query a multitude of fields to support our data-rich articles,” said Vivek Kumar, Head of DevOps at Hindustan Times. “This is easy and doesn’t impact user experience as we don’t need any downtime to change configurations in Blank Paper.”
MongoDB has also enabled Hindustan Times to rapidly develop and launch financial services like loan recommendation engines and credit card affiliation programs. Being able to adapt its data architecture quickly to meet partner requirements has enhanced collaboration and optimized time-to-market for these services.
“Personalized personal loan offerings are all based on data. Being able to see clearly what data pools we could draw from, and use that data dynamically, was instrumental in launching into the personal finance space,” said Kumar.
To feed its personalization workloads, Hindustan Times is currently using data from both Elasticsearch and MongoDB. But as the company looks to further reap the benefits of LLMs and AI, Hindustan Times is considering moving more of its workloads onto MongoDB Atlas Search.
“Because MongoDB Atlas is built with AI in mind, our data is ready to be fed to a wide range of LLM models, as well as used within RAG-based use-cases,” said Kumar.
As of 2025, Hindustan Times is operating ten MongoDB clusters across a wide range of applications and workloads, with more to come.