"Developers are very happy because New Relic has enabled them to be proactive and patch a particular issue prior to it escalating and affecting users. Leadership is also happy just seeing that they are not the first to have to come back to us with an error in the first place. And the user experience is greatly enhanced as a result, which was our primary aim in deploying an observability solution."

Amit Dsouza
Digital DevOps Engineer, Southern Cross Austereo.

Southern Cross Austereo (SCA) is one of Australia’s leading media companies reaching more than 95% of the Australian population through its radio, television, and digital assets. Under the Triple M and Hit network brands, SCA owns 99 stations across FM, AM, and DAB+ radio. SCA provides national sales representation for 34 regional radio stations and broadcasts 93 free to air TV signals across regional Australia, reaching 2.7 million people a week. It also operates LiSTNR, Australia’s free audio destination for consumers, housing radio, podcasts, music, and news.

Searching for a sound observability solution

Like many industries, Australia’s radio broadcasting sector is going increasingly digital. Faced with increasing competition from subscription services like Spotify and Apple Music, radio broadcasters are looking for ways to innovate and increase audience share. As part of a digital-first initiative, SCA created a major new application, dubbed LiSTNR, for its audience. Developed by Australians for Australians, the app is designed to consolidate all of SCA’s shows, podcasts, and apps under one umbrella.

SCA knew it required an understanding of what users were experiencing, but this was hampered by a lack of observability. Application visibility was lacking in root cause analysis, leaving engineers to manually troubleshoot, with tracing being more complicated when performing manual tasks.

SCA had previously looked to open source tools for monitoring but these were becoming increasingly difficult to use.

"Over a period of time, the tools themselves became quite complex to manage and maintain, to keep the system up and running, and ensure that there were no issues. The main problem was that if we ever had an outage on the operational front of our monitoring tool, and then an outage on production software such as LiSTNR, we wouldn't have any visibility over the issues going on," explains Amit Dsouza, Digital DevOps Engineer, Southern Cross Austereo.

Monitoring the network with instant alerts

To overcome the lack of visibility, SCA thoroughly researched the market and looked at multiple providers. It finally identified New Relic as ideal for its needs.

"New Relic has the best support and technical prowess. We have tried other tools from similar companies and had problems with either support or with the toolset itself which is why we made the switch," Amit says.

SCA found implementation very easy, especially with the support received from the New Relic team. While only managed as a side project due to different priorities in the business, “everything was up and running within a month”, he recalls.

"New Relic was very helpful and they were constantly present for any of our questions. We would have weekly or biweekly calls based on our requirements, and they would either help us physically or point us in the right direction [of who could help]. So, it was quite smooth sailing overall," Amit says.

SCA set up a standard dashboard in its network centre that shows the current status of all its applications. “It’s a simple display that shows whether an app is up or down, and isn’t overwhelming in terms of detail so the first level support team can quickly trigger an on-call,” Amit explains.

More specific dashboards have also been set up, such as alerts for certificate expirations if auto-refresh stops working. Further dashboards are planned for the leadership team in the future to provide a clear view of the business impact of technology initiatives and to enable real-time decision making at the executive level.

Alerts also meant managing peak periods was easier. “If we have additional users jumping on, auto-scalers are in place to automatically scale up the entire service. But if New Relic alerts show that particular machines are overwhelmed, the team can add servers to the entire Kubernetes setup, and that way prevent any outage during a peak,” he says.

Full visibility: clear skies ahead

Five months after LiSTNR went live, radio streaming had grown significantly with podcast listening up 200%. The volume of content consumed per signed-in user has also grown by over 60%, which is reflective of users seeing value in the app.

New Relic alerts have also led to significant productivity gains. The team has set up alerts for certain shows so it knows of any problems with data syncing the day before the show is due to air. Engineers can immediately jump on an issue and fix or patch the problem. This has greatly improved customer experience, as the team is now the first, instead of the last, to know about any problems.

"Putting alerts into place has really helped a lot. We get instantly alerted about issues and can fix them straight away, rather than waiting until end users are impacted and registering complaints.

"Developers are very happy because New Relic has enabled them to be proactive and patch a particular issue prior to it escalating and affecting users. Leadership is also happy just seeing that they are not the first to have to come back to us with an error in the first place. And the user experience is greatly enhanced as a result, which was our primary aim in deploying an observability solution," Amit says.

Overall, SCA has seen a 20-40% reduction in the time spent to detect, understand and repair issues. Open source hosting, maintenance, and support costs have also been cut by 50%.

The solution is also driving metrics around business strategy. Having full visibility has enabled SCA to report on engineering effort and efficiency against application availability and up-time.

"New Relic was very helpful and they were constantly present for any of our questions. We would have weekly or biweekly calls based on our requirements, and they would either help us physically or point us in the right direction (of who could help). So, it was quite smooth sailing overall."

Get started today for free.