Last month, I began a market research project to interview Wavefront customers to capture why they selected us and how they’re getting value from our ‘unified cloud monitoring with real-time analytics’ platform.
Many Wavefront users are DevOps teams in revenue-generating SaaS businesses. Developers, DevOps, Site Reliability Engineering, Tech Ops– all need a collaborative monitoring, analytics, and visualization platform that helps produce true insights across the full application and cloud stack, and keep their applications running with the highest performance and availability.
Here are a few observations so far. For one, Wavefront customers are generally a high achieving lot. Most have built a data culture into their business.
They understand the transformative power of measuring everything, discovering how data analytics can be used to improve the business and create wins. And as for their monitoring strategy, they leverage the advantage of having “one system to rule all metrics”.
Initial In-House Monitoring Built with Open Source
But prior to standardizing on Wavefront, many had started with an in-house monitoring and graphing platform, often assembled with open source components. While the components of each in-house system vary a bit across customers, the more popular elements at each layer include:
Most of these monitoring systems met expectations initially; but, as time went on and as the business grew, application and infrastructure systems grew in complexity too. The organization grew too. This then led to expanding requirements for what the monitoring system needed to do, eventually driving the impetus for finding a better-suited platform like Wavefront.
Five Signs That Compel a Replacement
Across the many customer interviews that I’ve completed so far, I’ve heard a recurring pattern – I call them the “5 signs” – summarizing what compels a SaaS business to realize it’s time to replace their in-house monitoring platform.
Here’s how customers described what they experienced:
1. Increasing needs for a more flexible visualization UI.
Needs are moving quickly beyond static displays of metric dashboards, to improving operational anomaly detection and diagnosis. But being more proactive on anomalies isn’t so easy given the complexities across their application systems.
To do this, customers see the need for a more powerful and intuitive UI to manipulate metrics, apply analytics and render charts that can be easily shared across teams. Most in-house systems are based on 5-10 year old technology, and weren’t optimized for rapid query and drill downs/up of multiple data streams in real-time.
2. Growing dissatisfaction with system response times.
Starting to hear a lot of, “It takes too long for charts to load.” Customers acknowledge that their in-house monitoring systems started to slow as the number of users, query loads, and data ingestion rates increased.
Wait times around data availability and query response now exceed a “speed of thought” response time, negative impacting users’ productivity.
3. Expanding requirements to define more intelligent alerts.
Experiencing lots of alert noise, often generated by many overlapping tools. Customers expressing need for a tool to easily design more intelligent (analytics-driven) alerts that reduce alert storms.
Many note that their current in-house system lacks a query engine that is flexible and advanced enough to improve design of their own KPIs from all of the metric data, as well as a good way to pre-test alerts before they are put in the wild.
4. Rising cost for ongoing system maintenance and support.
Costs to support the in-house system are starting rise significantly as complexity and needs increase. More resources are being spent on maintaining the monitoring system itself, versus on getting insights on the metrics and analytics provided.
This leads to interest in migrating at a SaaS, cloud-hosted alternative, eliminating the need to maintain the software altogether.
5. Emerging concerns around system scalability and data reliability.
Limitations are starting to appear in current systems’ abilities to scale horizontally while providing access to all of the granular (not aggregated) data collected for all time. Many embrace the Etsy axiom, “if it moves, we track it”, and while there’s great value in doing so, it exposes scaling issues with in-house systems.
The ability to ingest and apply analytics to 1 million points per second streams becomes relevant. Similarly, as the in-house system itself become more complex, it experiences downtime more often and data gaps start to impact operational effectiveness.
Can You Afford to Ignore these Signs?
Are you seeing any of these signs in your organization? If so, perhaps your monitoring system is holding your business back. DevOps leaders I’ve spoken to saw these as telltale signs to upgrade their in-house monitoring.
Wavefront has made a huge difference for them; we can likely do the same for you. Get in contact with Wavefront today, and learn how unified cloud monitoring with real-time analytics can help your teams win. Click here to get started with the Wavefront solution.