2014 in retrospect

Throughout the year, both our team and product took a big step towards automating JVM performance monitoring.

This has been possible first and foremost thanks to you, our dear customers and to you, our team. The insights and lessons learned steered the way to major improvements by the team.

Our long-term vision in becoming the tool all Java server-side installations benefit from during the performance monitoring is now built on a lot stronger foundation.

Key facts and figures for the year:

Deeper understanding of the needs and pains of our users. Throughout the year we concluded:

• 100+ face-to-face interviews to understand the day to day problems related to performance monitoring and troubleshooting
• 500+ remote interviews with customers to understand how they discovered Plumbr, what was the actual problem triggering the need and what were the expectations to the product
• 500+ product demos. Exhibiting at conferences such as JavaOne and Devoxx allowed us perform more than 500 demos, again resulting in lots of feedback and insight into the actual problem domain.
• Loads of statistical conclusions on performance metrics collected by Plumbr Agents and thousands of use cases qualitatively analysed in our labs. The visibility via statistical data has built the much needed foundation to solve next set of performance problems our current customers face.

Expanding the product vision by detecting a lot more than just memory leaks. In 2014 we added automatic root cause detection for Garbage Collection pauses and locked threads. Both of these performance problems are notoriously difficult and time consuming to solve manually. Plumbr is the only software product that can detect the root cause automatically.

Consequently, evolving from “Plumbr is a memory leak detector” to a “Plumbr is a JVM monitoring solution” value offering. With this change Plumbr has become a good complement to the traditional APM offering – APMs are good at discovering performance issues, measuring the impact and locating the failing node. Plumbr adds essential information to the equation by deep-diving into the incident, discovering and explaining the actual root cause. After APM hands you the information about the incident severity and location in the infrastructure, Plumbr does the other half of the job by linking this information to the actual root cause in the (potentially huge) code base or configuration. This saves you on average 80 hours per performance incident encountered.

Growing the team from from 10 to 16 full time members. This included our first international hire and relocating him to Estonia. We can hire top talent and at the same time make the team more international. In the beginning of 2015 we already helped another engineer relocate here, and we are currently in talks with the next potential ones. If you are interested in joining us, check out our open positions.

Usage of Plumbr increased significantly. During the 2014 Plumbr monitored 4,000 different JVMs for more than 600,000 hours. Considering that for the most part of the year Plumbr was perceived as a troubleshooting product and only during the last two months we started receiving monitoring customers, we expect to see a significant change in this regard.

Throughout those 600,000 hours, we ended up discovering more than 150,000 performance incidents. Even more importantly, Plumbr linked those 150,000 incidents directly to 6,000 unique root causes, enabling our users on 6,000 cases to bypass the complex and tedious troubleshooting and immediately fix the underlying root cause.

Achieving 50% revenue growth compared to 2013. A deal that stood out for me was the first time a customer significantly increasing the amount of monitored JVMs when rolling from their first year of subscription to the second one.

Thank you for making 2014 such a successful and memorable year. Here is to the even greater 2015!