New Relic APM, Logs + Infrastructure Monitoring (PAYG with Free Tier)

Since I'm a performance engineer, I typically use New Relic  day-to-day for investigating any performance bottlenecks identified during our performance testing of any application. I look at the results, identify the root cause behind performance issues, and create dashboards to monitor the observability perspective of the tool for production as well as the QA environment.

New Relic helps me every day because whenever we face any issue, the very first thing we do is check the dashboard we have created. We browse through the dashboard to look into different aspects including CPU, memory, logs, and different transactions. Once we identify what might be the reason, we either directly go into those metrics in detail to look for the root cause or look into transactions to identify if a performance bottleneck exists, then at what layer or step it has occurred and what is causing that performance issue. Once we identify that, it is very easy to look into the code for that particular method or area to pinpoint the root cause, inform the developers about the issue, and identify possible solutions for it.

I have been using New Relic extensively for performance engineering troubleshooting and root cause analysis for our large-scale SaaS environment, which has become one of the primary tools we heavily rely on during our performance testing and production issues investigations. I appreciate New Relic's ability to provide end-to-end visibility across the entire application stack, and its APM  capabilities are particularly useful for our day-to-day tasks by identifying slow transactions, bottleneck methods, external service dependencies, database performance issues, and error hotspots. We frequently use transaction traces, distributed tracing, thread profiling, database analysis, and external call breakdowns to pinpoint where the response time is being spent. With our Kubernetes  workloads moving to containerized environments, the Kubernetes  monitoring capabilities that New Relic provides are valuable as they allow us to correlate application performance with pod-level metrics, resource utilization, deployments, and infrastructure behavior, helping reduce the time required to identify performance bottlenecks. The integration we have seen between APM , Kubernetes monitoring, logs, and distributed tracing provides a much more complete picture than using disconnected tools. The NRQL feature allows us and our team tremendous flexibility to build custom dashboards, perform deep analysis, and answer specific questions that may not be available through standard views. The dashboarding capabilities we rely on are powerful, allowing both engineers and leadership to view performance trends and operational metrics meaningfully. Leadership is not technical, so being able to create that dashboard helps, and for engineers, alerts can be automatically triggered without needing developers or QA engineers to investigate further. If there's nothing wrong, they are happy that all is well, and if there is an issue, alerts are triggered, prompting them to go to New Relic to identify the cause and analyze the issue.

The best feature of New Relic is the effectiveness of the observability platform. The combination of APM, distributed tracing, Kubernetes monitoring, logs, database visibility, profiling, and analytics capability significantly reduces our troubleshooting time, and those are the features I appreciate about New Relic.

I think we utilize all of the features available due to our architecture, where we must go through everything. There is a sequence in how we go through these features: first, we look at the dashboard for a holistic overview, then check the transactions to identify degraded ones, move to traces to find the cause of slowness, and might have to go into logs, database queries, external calls, or distributed tracing.

While I appreciate many aspects of New Relic, I believe the product could improve in some areas—specifically, some advanced capabilities can have a learning curve for new users, and the licensing and consumption model can be difficult to predict, particularly in environments generating large volumes of telemetry data. Organizations may need to invest time in proper instrumentation and dashboard design, as we did, to reach the current stage we are now.

The licensing and consumption model can be unpredictable due to dependency on telemetry data, making it challenging for environments generating large volumes to estimate costs. Organizations need to invest time to explore New Relic's extensive functionality and properly instrument features to realize their full impact. Additionally, designing dashboards is not straightforward; users need to create their NRQL queries before they can fully understand the value.

Regarding the user interface, I have noticed changes in past years, with some features we appreciated in previous versions and others in new ones. One issue we have reported is the text overlaying on dashboards when editing, which makes it challenging to navigate. Another issue is that some users experience blurred text, requiring a system restart and login to rectify.

I have been using New Relic  for more than eight years now.

Before using New Relic, we used Grafana , Splunk, VividCortex, and Opster to monitor and observe each layer, requiring different tools for each, which slowed us down. With New Relic, we can visualize all underlying systems architecture and metrics in one place, which helps us correlate issues faster and easier, allowing us to save almost seven to eight hours of analysis time previously spent on multiple systems and tools.

New Relic is deployed in our organization as a private cloud or possibly hybrid cloud, as it can function both ways.

I advise that New Relic is the best APM tool available, providing everything you need to pinpoint root causes for performance bottlenecks and serving as an effective observability tool for production telemetry metrics. There are open-source tools that are easier to use but have limitations, whereas New Relic offers a wide variety of useful features for performance engineers, SREs, and leadership. While it is costly, the valuable data and time saved justify the investment, so I encourage others to give it a try.

I would rate New Relic an eight out of ten. We utilize nearly 70 to 80 percent of its features, so eight seems appropriate given the ease with which users can work and instrument New Relic.

I choose eight out of ten as we are yet to explore 20 percent of New Relic features, and the ease of creating alerts, dashboards, and instrumentation often requires expertise. Live tracing of a transaction would enhance the platform, allowing us to trace transaction data flow with an architectural diagram to visualize time spent, which would elevate my rating to a 9 or 10.

The time savings initially amounted to seven to eight hours per person per week, and increased to almost 24 hours after creating dashboards with automated alerting systems in place. Now, we can directly get alerts during our tests and have a holistic view of the architecture's performance, enabling us to quickly identify any bottlenecks. Since we save time, we can create better dashboards and invest more time in harnessing other areas of the applications we test. My overall rating for New Relic is eight out of ten.