Open standards in 2026: The backbone of modern observability

Open source software and open standards are now an essential part of how organizations maintain their systems.

That's not to say they haven't always been important, but the fourth annual Observability Survey, brought to you by Grafana Labs, shows just how deeply the shift to open has taken hold, with 77% of respondents saying open source and open standards are important¹ to their observability strategy. We're also seeing open source have an impact on how people prioritize and select new tools, while OpenTelemetry continues to become a unified standard across all telemetry signals.

Taken collectively, this points to a broader change: Organizations are shifting from strategies anchored to a single vendor to strategies that prioritize open, portable standards for moving data across tools and platforms to better understand the state of their systems—and even their businesses.

In this blog post, we'll take a closer look at the role of open source and open standards in observability today, based on the results from the 2026 edition of the Observability Survey, the largest community-driven survey on the state of the industry.

Note: Check out the full 2026 Observability Survey to see what more than 1,300 respondents have to say about AI, open source, the expanding scope of observability, and the practices organizations are adopting today. And watch the video below for a recap on some of the main themes coming out of this year's report. 

You can also read our in-depth analysis on the role of AI in observability, or dig into the stats yourself in our Grafana dashboard, which includes filters for industry, role, region, and company size so your team compares to its peers.

Open source and open standards are central to observability strategies today

Companies today are collecting so much telemetry from so many disparate sources, and they need a way to unify that data in one central location. As a result, open source and open standards have become mainstream. In fact, 61% of respondents say open source/open standards are either “very important” or “essential” to their observability strategy. Only 3% say they’re not important at all.

Open is more important to companies that prioritize observability as part of a broader business strategy, too. Among respondents who say observability is business-critical at the highest levels of their company ("CTO/C-level"), 64% cite open source/open standards as "very important" or "essential." In contrast, among those who say observability is "not at all" considered critical at their company, just 51% cite open source/open standards are "very important" or "essential."

Teams that place greater value on open source/open standards are also more likely to save time or money through centralized observability, with 84% of those who say it's "essential" reporting time/money savings, compared to 77% of all respondents.

And while teams that use self-managed setups are much more likely to see open as important, a majority of those using SaaS say it's important, too, including 61% of those using "SaaS only" and 73% of those using "mostly SaaS." Most SaaS platforms that dominated the early days of observability all used proprietary collectors and backends, so this broad support for open is yet another sign of the shifting dynamic in this space today.

Open source, open standards, and the focus on outcomes

When asked about selecting new observability tools, a quarter of respondents say being "based on open source software/technologies" is one of their most important criteria, making it the fourth most common consideration behind cost, ease of use, and interoperability with other tools. 

That puts "open source" in the top third of selection criteria (respondents could pick up to three), but it's also well short of the percentage of people who say "open source/open standards" are important¹ to their observability strategy (77%). There are many possible reasons for this, but part of it likely comes down to how people separate open source from open standards, with the former typically associated with backends and visualization layers and the latter associated with data collection frameworks. 

The two de facto open standards in observability today are Prometheus and OpenTelemetry. Yes, they're both open source projects, but they're perceived differently than the components that make up an open source observability stack. 

Note: It's worth mentioning here that OpenTelemetry, as a project, aspires to be a much broader standard to work across all telemetry types, while Prometheus is focused squarely on metrics. You don't have to standardize across your entire architecture to get the full value out of Prometheus the same way you would with OpenTelemetry. Check out this recent "Big Tent" podcast to hear more on this topic.

And while we didn't explicitly ask about open standards as part of the tool-selection criteria, it's clear practitioners value the benefits that come from using them. For example, "interoperability with other tools at my organization" (26%), "cross-observability signal correlation" (15%), and "ease of switching to another tool in the future" (12%), can all be tied back to the benefits of open standards. And when you combine those answers with "based on open source software/technologies," 58% of all respondents—and 62% of those who say open is important¹ to their observability strategy—cite at least one of them as a top criteria.

Open source and AI

You may have noticed the chart in the previous section that stated only 15% of respondents cite "AI capabilities" as a top priority when selecting new observability tools. This survey was conducted between October and January, so it's likely that opinions have shifted since then, given the rapid pace of development in this space.

It's also worth calling out the emerging relationship between open source, observability, and AI. Grafana Labs CTO Tom Wilkie recently called open source the "cheat code" for AI. He noted that LLMs are trained on vast amounts of publicly available information about open source projects, so they're essentially "pre-wired with knowledge of how to interact with these open ecosystems"—something that helped accelerate the development of our own LLM, Grafana Assistant. 

We're also seeing a lot of interest in our open source MCP server, which gives AI agents and LLMs comprehensive access to Grafana and its observability ecosystem. Clearly AI and open source will continue to evolve and influence one another, so this will definitely be a space to keep an eye on.

Prometheus is deeply embedded, and OpenTelemetry is quickly catching up 

If open standards are the foundation of modern observability, then Prometheus and OpenTelemetry are the main pillars of that foundation. Both projects are seeing widespread investments across organizations of all sizes and industries. In fact, 65% of respondents report investing (investigating, building POCs, or using in production) in both Prometheus and OpenTelemetry. 

Prometheus holds a slight lead in overall adoption (77% vs. 76%), but there are notable differences in how people are deploying the two technologies. Prometheus is used much more in production² (59% vs. 41%), which makes sense since it's a more mature technology. OpenTelemetry has a greater share of people investigating or building POCs (35% vs. 18%), meaning it has a larger pipeline of organizations preparing to scale their OpenTelemetry deployments.

There's also a split in the degree to which teams use OpenTelemetry and Prometheus in production. Both are used for "some production workloads" at the same rate (20%), but Prometheus is used for the "majority of" or "all" workloads at almost twice the rate. This could be attributed to Prometheus being more established, but it also points to a potential missed opportunity. Certain parts of OpenTelemetry—more specifically, traces—are less useful when you don't have everything fully instrumented, so it's possible people aren't yet seeing the full benefits of moving to OpenTelemetry. 

Regardless, each of the projects is seeing positive momentum, as organizations were more likely to have increased their usage of both over the past year. The percentage of respondents who say they increased their usage of OpenTelemetry was slightly higher (47% vs. 42%), highlighting how it's at an earlier stage of adoption, with a higher trajectory for potential growth. Only a small minority decreased their use of Prometheus (6%) and OpenTelemetry (4%) compared to the previous year.

OpenTelemetry is delivering on its potential for metrics, logs, and traces

While Prometheus was designed exclusively for metrics, OpenTelemetry has been built to do more as a single, vendor-neutral standard for metrics, logs, traces, and even profiles (though profiles are still an emerging space for the project and observability in general). And that potential is quickly becoming reality, as respondents say they use OpenTelemetry broadly across all three major telemetry types—metrics (57%), traces (50%), and logs (48%).

Interestingly, organizations using SaaS are more likely to deploy OpenTelemetry today. For example, when you compare "SaaS only" to "self-managed only" users, they're more likely to use it for metrics (65% vs. 51%), traces (52% vs. 46%), and logs (52% vs. 45%). However, self-managed users are generally more likely to have increased their investments across all three telemetry types over the past year. This points to two possible explanations—either SaaS enables teams to move quicker, or teams that want to move quicker are inclined to opt for SaaS.

Teams want OpenTelemetry to deliver simpler observability—and more freedom

So why are organizations embracing OpenTelemetry? Yes, the promise of support for metrics, logs and traces is important—as is the need for semantic conventions—but it's clear that the technical specs are less important than the intended outcomes. In fact, it really came down to two things: simplicity and flexibility.

"Ease of adoption and management" (41%) and "ease of switching vendors and backend technologies" (37%) were the clear leaders in terms of what respondents want to solve with OpenTelemetry—and for good reason. Moving to a new observability platform often means re-instrumenting applications, rebuilding telemetry pipelines, and retraining employees, but with OpenTelemetry you can instrument once and send data anywhere.

That type of functionality will become more important  as the scope of observability expands. For example, half of all respondents report using observability tools to track business-related metrics. That type of data is traditionally analyzed in BI platforms, so standardized pipelines will be essential if teams want to correlate that data with other telemetry signals.

And the more teams pin their future to OpenTelemetry, the more likely they are to see it as a potential solution to these two challenges. Taken together, the results point to a growing desire for observability systems that are easier to use, more flexible, and less tied to any single vendor.

Final thoughts on the Observability Survey and open source

We want to thank everyone who participated in this year's survey, which was our biggest yet. Between October and January, more than 1,300 observability practitioners and leaders from 76 different countries participated in our survey. We collected responses online and at events around the world, including our own shows and third-party conferences such as AWS re:Invent and KubeCON.

It speaks to the collective power of the open source community that so many of you were willing to share your thoughts, and we take pride in making the Observability Survey an ungated, public asset for anyone who wants to learn from and share the data.

We've built some of the most popular open source technologies available today, including Grafana and Loki, and we've been major contributors to both Prometheus and OpenTelemetry. Listening to your feedback is a core to our understanding of Grafana Labs as an open source company.  We learn from you how to do better work. Thank you for sharing your knowledge and opinions with us.

^{1 "Somewhat important," "very important," and "essential" answers combined}

^{2  "We are using it across a few production workloads," "we are using it across the majority of production workloads," and "we are using it in all production workloads" answers combined}

Grafana Cloud is the easiest way to get started with metrics, logs, traces, dashboards, and more. We have a generous forever-free tier and plans for every use case. Sign up for free now!