Raygun is a company that I first got involved in as an investor (and, for a brief period, as a board member) a few years ago. Raygun was formed out of an earlier developer tools company, Mindscape. But while Mindscape had seen pretty good success, it was seeing some even better promise with a new tool, a crash reporting offering.
Based on that success, the company raised a bunch of money and focused its attention on that tool and the potential around it. They added Real User Monitoring (RUM) to build out the value of the initial tool. Fast forward to today and the New Zealand headquartered company now has an extensive North American presence (in Seattle, demonstrating that Raygun’s CEO, JD Trask, likes rainy, windy cities.) They’ve also picked up some impressive customers – Domino’s Pizza, Coca-Cola, Nordstrom and Microsoft among them.
That product and customer momentum are culminating today in the announcement that Raygun is adding to its existing offerings with an application monitoring solution. More on why that is important below.
It’s all about convergence
Once upon a time (but not in a galaxy far, far away), there were distinct tools for application monitoring, crash reporting, and real user monitoring. Different vendors offered these different components and left it up to their customers to tie them together. But those days were also different from a development and operations perspective, and often the development and production teams didn’t really communicate. In addition, software wasn’t seen as particularly dynamic so there was no pressing need for holistic monitoring solutions.
Fast forward to today and the rise of development leveraging modular developer services, microservices architecture and the “deploy often” (actually, constantly) status quo and the needs that organizations have when it comes to monitoring are also very different – the different components need to be delivered by a single vendor and ideally under a single dashboard and management paradigm.
It is for this reason that other vendors in the monitoring space have converged to offer complete solutions – Datadog, for example, started off as an infrastructure monitoring solution, but moved into application monitoring, while Silicon Valley darling New Relic went the other way – formerly an APM vendor exclusively, they moved into the infrastructure monitoring space.
So whereas Raygun’s existing capabilities alert teams to software errors in their application’s code and report on poor quality end-user experiences (that’d be the error reporting and real user monitoring parts of the puzzle) with this release, customers can also monitor server-side performance too. To be clear, this announcement sees Raygun compete head-on with these other vendor’s Application Performance monitoring (APM) product, but not directly with their infrastructure monitoring products, however, Trask tells me that the company has some near-term intentions including:
- Additional integrations beyond those that Raygun already supports
- Lightweight infrastructure monitoring
- Expanded programming language support
- OpenTracing support
A big opportunity
The APM space is big. And growing. Estimates suggest that the market will, within only a year or two, be worth around $5 billion. That’s a massive opportunity and what Raygun (alongside its competitors mentioned above) is banking on is the fact that more traditional monitoring companies are so slow to react, and so tied up with monolithic, on-premises and slow development and deployment approaches, that they won’t be able to sufficiently serve the emerging market.
In addition to the traditional monitoring vendors, these companies are banking on the fact that traditional logging approaches are too slow, by orders of magnitude, for modern requirements.
The value add is, as I have stated, the breadth of the offering. Raygun’s APM solution is tightly integrated as a part of the Raygun Platform. Meaning that customers can see performance traces against crash reports, or drill from an error in a trace report to the Raygun Crash Reporting view to get the full context. Customers can also combine Raygun Real User Monitoring with Raygun APM meaning they can see that a user waited on the server for a while to load a page, and drill in and see what the server was doing that caused it to be that slow.
Closing the workflow loop
I’ve long suggested that monitoring and management should go hand in hand. To clarify, it seems to me silly that a solution can highlight that an issue exists, without helping teams to actually resolve the issue itself. Raygun’s solution to this conundrum is a built-in issue detection system. The system creates issues around problems impacting end-user experiences and builds a workflow around resolving them. Teams are then able to prioritize, assign, discuss and solve issues as a team. It’s a glimpse of a new way of doing things, one in which the continuous loop of issue identification and remediation is just that, a continuous loop and not two different worlds.
In addition to the workflow aspects, Raygun has also built integration with repository solutions such as GitHub and BitBucket which automatically links the diagnostic details captured to the source code in a customer’s repository. What this means in action is that a developer just needs to inspect the code causing the issue and take the appropriate action without having to switch between tools. Trask is, unsurprisingly, very bullish about his company’s move into APM, saying:
The addition of APM to the Raygun Platform delivers on our vision of an integrated monitoring solution. Errors, crashes, real user performance and server/application monitoring can now be all in one place, not siloed or fragmented between teams. We’ve been building towards the most integrated monitoring platform for years, re-thinking what is achievable and support all modern solution architectures.
MyPOV
Personal bias aside, there is no option, in my eyes but to deliver a consistent and coherent platform that covers the various different monitoring requirements. It’s great to see Raygun finally get to launch a product which has been a long time in the making.