Is it Still Early Days for Observability?

Editor’s Note: This is the first installment of a series of blog posts previewing our State of Observability 2024 survey report

For nearly a decade now, observability has been one of the hottest topics in IT operations. Borrowed from control theory in engineering, observability in IT operations means measuring a system’s performance by the data it generates, typically metrics, events, logs and traces. As log file analysis tools found their way from security operations into IT operations and microservices-based applications started to replace monolithic code bases—generating demand for distributed tracing-- the concept of observability in IT operations took on a life of its own in the early to mid-2010s. The migrations of workloads from on-premises environments to the cloud, which started around the same time, also generated more demand for observability data as traditional monitoring tools provided less insights into cloud environments.

So by now, you would figure most IT organizations were pretty far along in their uses of observability tools. However in our new State of Observability survey of more than 600 enterprise IT organizations and managed service providers, we found just the opposite. Most enterprises and MSPs are just getting started with observability. Just 24% of our total survey respondents told us they had achieved full-stack observability across 90% of their organizations. More than half told us they were either at the pilot stage (26%) or still exploring suitable use cases and potential solutions (30%). Another 19% claimed only “partial” observability implementations in a few business units.

Blog_Report_Observability_2024_Chart_01@2x-100

While vendor and analyst hype may be leading observability forward, both enterprises and MSPs (there was little difference in responses between the two in our survey) are taking a much more cautious approach to observability, working on pilot implementations, starting small at the department or BU level or still investigating whether and where observability is the right fit for them.

That doesn’t mean observability isn’t working. Most respondents—96%—told us that their current observability solution was delivering the value they expected, even though we infer from the responses that nearly a third of respondents aren’t far enough along in their observability deployments to expect much value yet.

So where are ITOps teams and MSPs looking for value first in observability? The clear winner according to our survey was Cloud/Cloud-Native Observability, with 61% of respondents saying they were exploring this type of observability solution, easily outpacing Hybrid Observability and Security Observability (both 50%), Network Observability (47%) and Full Stack/Unified Observability (34%).

Blog_Report_Observability__2024_Chart_02@2x-100

The results weren’t surprising. Like any new technology adoption, the “start small, look for easy wins” advice applies to observability. Organizations look to cloud/cloud native first as that tends to be where the most obvious need is for new tools; there really aren’t any legacy cloud/cloud-native monitoring tools to replace. Instead, this is an area where most organizations are eager to gain new visibility.

Full-stack/Unified Observability drew the fewest responses. Organizations may work their way up to this point—and 24% of our survey respondents are already there—but they’ll get there in steps. Cloud/cloud-native observability is the first step they’ll take.

Next week, we’ll look at the benefits enterprises and MSPs are realizing from their observability deployments and the ongoing concerns they have and challenges they face as they deploy the technology. You can download the entire report at the link below.