"With respect to monitoring and troubleshooting, New Relic has been our single source of truth. Swiggy’s user base is growing into the millions and thanks to New Relic, we can track the digital experiences of our end users while analysing performance. The insight we get from New Relic is enabling us to expand rapidly while remaining confident that we’re providing a reliable, flawless customer experience."
Tushar Tayal
Senior Engineering Manager at Swiggy

Swiggy, backed by Softbank Group, is India's largest online food ordering and delivery platform. Founded in July 2014, Swiggy is based in Bangalore, India, and currently operates in 500 Indian cities. In early 2019, Swiggy expanded into general product deliveries under the name Swiggy Stores. In September 2019, Swiggy introduced an instant pickup/drop-off service, Swiggy Go. The service is used for a diverse array of items, including laundry and document or parcel deliveries to business and retail customers. Swiggy is operated by Bundl Technologies Private Limited.

Observability: a recipe for success

India’s online Food & Beverage (F&B) industry has undergone explosive growth over the past few years, driven by the pandemic and growing consumer demand for convenience and speed, alongside more confidence with online ordering. According to ResearchandMarkets.com, India’s online food delivery market alone will hit US$20 billion-plus by 2026, at a Compounded Annual Growth Rate of 30% during the 2021-2026 period.

Swiggy, the market leader with more than 30 million users across 500 cities in India, is a digital organisation growing at lightning speed. It has been expanding exponentially throughout cities and states and other lines of business.

"With such a buoyant market we’re seeing a growing number of new competitors enter the FoodTech space. There’s a need to innovate faster in order to outpace the competition. That means optimising operational efficiency and ensuring our platform is stable, reliable, and scalable to keep up with our rapid growth," explains Tushar Tayal, Senior Engineering Manager, Swiggy.

As the business grew, the Swiggy tech stack became increasingly complex. The company had to ensure true observability was in place to address problems before they escalated and impacted customer experience.

"With respect to monitoring and troubleshooting, New Relic has been our single source of truth. Swiggy’s user base is growing into the millions and thanks to New Relic, we can track the digital experiences of our end users while analysing performance. The insight we get from New Relic is enabling us to expand rapidly while remaining confident that we’re providing a reliable, flawless customer experience," Tushar says.

Key ingredients for reliability, transparency

Swiggy began using New Relic in 2018 for basic monitoring via Google Firebase, which worked well with Android but was less effective on Apple iOS. Since then, New Relic has demonstrated further value as a complete observability partner–the relationship with Swiggy has deepened and investment has increased.

With ‘custom events,’ New Relic enabled Swiggy engineers to target and monitor a specific user base. Insights were available within 15 minutes and there was increased flexibility in integrating New Relic with all mobile devices and platforms.

This has been critical as Swiggy has expanded into new verticals such as grocery delivery and general-purpose delivery. For example, Swiggy has a navigation widget with entry points to its new verticals. Multiple errors were reported by users, but not measured by any of Swiggy’s dashboards, frontend and backend.

"We wanted to understand what was going wrong so we added custom metrics. We started monitoring how many users were missing the entry point to the new verticals. 

“We had a dashboard on New Relic for this, and each week we could go over all the dashboards, determine the instances and times when issues occurred, and what was actually going wrong in the backend," Tushar explains.

Through New Relic, Swiggy was able to identify multiple connected systems with configuration issues, and engineers were able to make changes for a better user experience.

According to Tushar, Swiggy relies on user-level metrics from New Relic to "create a segregation of our user cohort–what OS they’re using, their network (LTE, WiFi, or 3G), and the type of devices.”

“With more than 30 million users transacting with us, understanding the device they’re using and to test our solutions (across all devices) is very important. All the traffic (data) that comes through–even the different level of orders between different cities–we gather from New Relic.

"New Relic really helped us understand what the user has faced, and we were able to debug any issues. Overall, New Relic has improved our productivity by up to 10 percent," Tushar says.

Swiggy’s teams have also benefited from New Relic University, including dedicated sessions with New Relic experts on Cloud Optimisation, and how to enable front and back-end teams to use observability to realise positive performance outcomes.

The food delivery giant uses New Relic for design optimisation so Swiggy engineers estimate how many users are facing UI issues, for instance, and assess the impact on customers.

Always on, instant commerce

Having true observability has enabled Swiggy to respond swiftly as alerts are set up with New Relic for approximate errors of more than 5%. Swiggy has dedicated, 24/7 teams on-call to fix any critical issues, and depending on the type of severity, an issue can be addressed in as little as five minutes.

New Relic has also been vital for peak periods such as major festivals and sporting events which can see a 20% to 30% spike in traffic. Swiggy uses historical data to gauge the type of requests it could receive on a particular day.

"For example, on Diwali, how many consumers are expecting, how many orders, and what are the probable requests per minute (RPM)? How much do we want to scale? What about additional buffer? Using historical data we can scale our systems better and prepare for any spikes," Tushar says.

Uptime has also improved. Tushar cites Christmas as an example when Swiggy was "getting different types of errors due to traffic increase, and thanks to New Relic we could still serve up to 100 percent of our potential in the lead up to this festive period.”

Tushar’s team turned to New Relic to resolve the issues, and by the time the new year came around, Swiggy was able to scale its systems to mitigate and prevent any outages from traffic surges.

New Relic has truly become part of Swiggy’s DNA.

“In general, if we see that one of our APIs, which is part of the critical order flow, is not working well or throwing up errors we can figure out the root cause. New Relic is the way to understand if there is a deviation in standard API latency.

“And if we start having issues with one of the payment endpoints, New Relic is the single source of truth again for us to understand, monitor, and rectify or mitigate those errors,” Tushar says.

 

Get started today for free.