At the beginning of 2020, I joined a team of 4 software engineers (later grew to 6) to influence the way an organization of a little over 1000 people

3 considerations from building a platform for Observability

submited by
Style Pass
2021-06-13 12:30:06

At the beginning of 2020, I joined a team of 4 software engineers (later grew to 6) to influence the way an organization of a little over 1000 people (~40 teams) observed their business. Our vision was to give them the tools they needed to understand the state of different aspects of the business so that they could use those observations to inform decisions they were making in a timely way.

We were able to provide an observability platform to our first customers in 4 months, with bi-weekly releases of new features following that initial release. Within a 10 month period, we had about 40% of the organization using the platform and a number of customers providing feedback on how we could continue evolving the platform.

A lot has been written about Observability so I’ll skip over that context and share specifically about 3 things that I think were pivotal for this successful effort.

The ability to visualize a metric on a dashboard (for example) is what I am calling a capability. This capability is provided by a tool like grafana. It is critically important to identify the capabilities you want your platform to provide to customers before selecting or building the tools that can provide those capabilities. Most tools available today will attempt to provide several capabilities that you may not need or may prefer to get from other tools. Without a firm foundation of what these core platform capabilities are, you run the risk of making your customer’s experience worse buy yet adding another tool that provides the same capability and you miss out on the opportunity to maximize your investment by targeting your efforts towards developing a specific capability holistically.

Leave a Comment