Fanatiz is one of the fastest-growing sports streaming platforms in the world. It has tens of thousands of subscribers and is particularly popular with expats, who can stream their team’s games in real time or catch up on-demand wherever they are. That wouldn’t be possible without 1190 Sports, which manages and commercializes sports rights, or Nunchee, the white-label technology behind direct-to-consumer streaming services.

The businesses are owned and run by Fz Sports, which has 85 staff in offices across Latin America, Europe, and the United States. It streams around 10,000 games each year, which amounts to around 20 per day.

“We want to be the largest digital stadium in the world. Our subscribers mainly follow men’s and women’s soccer, but we also stream basketball, handball, golf, and other sports,” reveals Damian Krynveniuk, vice president of engineering and technical operations at Fz Sports.

To connect sports fans with the content they love, Fz Sports is a fast adopter of new technologies, and all three of its businesses are underpinned by cutting-edge technology and a culture of innovation. “We’re optimizing the whole sports value chain, reinventing the way rights holders, brands, and fans connect with each other, and plan on becoming the leading digital destination for enjoying and monetizing sporting content,” adds Krynveniuk.

When it comes to live events, scalability is a universal challenge. In soccer alone, there can be up to 30 matches happening simultaneously. Systems need to be ready to handle exponential surges in demand without a hitch.

“If your system goes down in the middle of a game, you risk losing subscribers. Their loyalty is with their team, not their streaming platform,” says Krynveniuk. “We need to be able to stream uninterrupted in real time and manage peaks in traffic.” But having spare capacity isn’t a luxury that many can afford, so it was imperative for Fz Sports to find a high performing database platform to support its business.

The company built a minimum viable product with the technology it knew best — relational databases. However, it quickly ran into challenges around scalability and versatility, and the developer team was losing too much time on manual maintenance.

Fz Sports makes content available in multiple regions and across any device — from wearables and smartphones to web browsers and connected TVs. This level of complexity requires a flexible database that can handle unstructured data and make it easy to navigate. Subscribers should be able to find the sports they want to watch quickly and easily, whether they’re watching live or browsing the archives.

Fz Sports began looking for a non-relational database that supports unstructured data and can grow and adapt in line with its business. “We were looking for a light, user-friendly database that was compatible with Javascript, microservices, and AWS. When we went to market, the best database to handle high volumes of traffic was MongoDB,” says Krynveniuk.

The team adopted a MERN stack; a web development framework made up of MongoDB, Express.js, React.js, and Node.js. MongoDB acts as the database layer, with Express and Node drivers in the middle as the application layer, and React on top to create the presentation layer. The advantage of this approach is that developers can create full-stack solutions using a standard methodology and core digital foundation.

Initially, Fz Sports implemented the MongoDB Community edition and had a dedicated team managing 12 databases. While this was a pivotal step on the journey from a relational to non-relational way of working, it wasn’t the final destination. As Krynveniuk explains; ”MongoDB Community was an excellent tool to start with, but we felt limited by the hands-on management. We had one full-time employee solely focused on database management when we’d rather have staff working on improving the subscriber experience and interface.”

The team decided to move to MongoDB Atlas, a fully managed database service that includes support and best practice guidance, as well as better dashboards, metrics, and tools to optimize performance. “We take a strategic view of optimization, focusing on projects that will have the biggest impact. For example, we worked with MongoDB on a month-long project to reorganize our authorization database and increased performance by 100% as a result,” recalls Krynveniuk. “That was down to having an unstructured database and great support from MongoDB engineers.”

Damian Krynveniuk, Vice President of Engineering and Technical Operations, Fz Sports

For the day-to-day running of the business, MongoDB Atlas comes with all the features Fz Sports needs to handle peaks in demand and help subscribers find the right content quickly. Its capability to scale a database cluster both vertically and horizontally, without downtime, and even supporting automatic scale-up and down, proves to be particularly powerful and cost-effective.

The team uses snapshots and dashboards in Atlas to track pay-per-view customers and report sales to the rest of the company. These real-time capabilities help Fz Sports plan activities, such as marketing campaigns and the activation of other channels, 25% faster than before.

MongoDB Atlas Device SDKs is used to manage Fanatiz triggers and move business logic events into the database for queries. It keeps profile information synchronized to the millisecond and maintains consistency between eight nodes working in parallel with MongoDB. This is also useful for generating dashboards.

To meet its priority of improving user experience, Fz Sports relies on Atlas' impressive search and archiving capabilities.

“We have more than 10,000 games per year available on our platform, and that needs to be available on demand to subscribers,” says Krynveniuk. “MongoDB Atlas Search helps people find the games they want to watch quickly, and Online Archive keeps the database fast, responsive, and cost-efficient, with content being archived after two years.” Online Archive has reduced storage costs by 40%.

The team is currently running a proof of concept with a recommendation engine built on Atlas Search. It will use AI and the MoreLikeThis feature to proactively recommend content that subscribers might enjoy based on their preferences and viewing history.

As early adopters of new technology, Fz Sports attends annual workshops and training sessions with MongoDB to keep staff upskilled and aware of the latest developments. The team chooses which new Atlas features to test based on what will have the biggest impact on the customer experience.

“We’re more productive than ever with MongoDB Atlas. It has a fast learning curve and it’s easy to onboard new staff because it’s so user-friendly. We don’t need a dedicated database guru anymore because we’ve got MongoDB support. We can focus on improving our services knowing we’ve got the agility to make those changes,” says Krynveniuk.

And Fz Sports has the metrics to back up its productivity gains. It saw a 50% performance increase when it restructured a service using MongoDB indexes, Online Archive, and best practices learned from the support team. Where once managing a huge number of concurrent viewers was a challenge, the company could now handle this without any additional scaling.

“MongoDB Atlas is an exceptional tool for everything we do across mobile, web, and cloud. It’s vital to have a fast, transactional database, and as the leader in its field, I highly recommend MongoDB,” concludes Krynveniuk.

Damian Krynveniuk, Vice President of Engineering and Technical Operations, Fz Sports