現在、このページは英語版のみです。

Future of observability

Organizations see the business value of observability—and expect to invest more in it.

Reducing MTTR

We asked respondents what would do the most to help reduce MTTR, a critical question considering 44–60% said they take 30 minutes or more to resolve outages. Overall, their top choices were:

  1. Better DevOps practices (39%)
  2. Automated incident response workflows (38%)
  3. Staff training for observability tools (36%)

Those with full-stack observability were notably more likely to say automated incident response workflows to reduce MTTR (42%) than those without full-stack observability (36%).

Ways to most help reduce MTTR for outages

  • Regional insight: Unlike respondents surveyed in Asia Pacific and Europe, those surveyed in North America were the most likely to select staff training for observability tools as the top way to reduce MTTR (40%), and their fourth choice was better DevOps practices. While those surveyed in Asia Pacific were the most likely to select better DevOps practices (42%). Download the full report to view more regional highlights.
  • Role insight: Practitioners were the most likely to select staff training for observability tools (37%). Executives were the least likely to select more staff (22%). And non-executive managers were the least likely to select a centralized view of available data (32%) and predictive anomaly detection (30%).
  • Organization size insight: The top choices for respondents from small organizations included staff training for observability tools (40%), more thorough instrumentation (36%), and predictive anomaly detection (35%). While the top choices for midsize and large organizations aligned with the overall results.
  • Industry insight: Government and healthcare/pharma industry respondents were the only ones to select more staff as their top choice. Those from the education, energy/utilities, financial/insurance, and retail/consumer industries all selected automated incident response workflows as their top choice. Nonprofit/unspecified and services/consulting respondents chose staff training for observability tools. While industrials/materials/manufacturing and IT/telco chose better DevOps practices. Download the full report to view more industry highlights.

Deployment plans for next year

This will be a big year for deploying additional observability capabilities—by the end of 2023, respondents expected that 72–86% of capabilities will be deployed:

  • Almost a third (32%) expected to deploy 1–5 new capabilities
  • More than half (56%) expected to deploy 6–10 new capabilities
  • Only 5% expected to deploy 11–14 new capabilities
  • Just 8% did not expect to deploy new capabilities

Notably, more than 40% expected to deploy ML model performance monitoring and AIOps in the next year.

When we look one year out, capability deployment is in the mid-80% range for capabilities like network monitoring, security monitoring, log management, database monitoring, alerts, and infrastructure monitoring.

Even with some of the capabilities that respondents were less likely to expect to deploy next year (like Kubernetes monitoring, synthetic monitoring, distributed tracing, and the like), we see numbers in the mid-70% range.

 

Capabilities deployment expectations for the next year

Deployment plans for next three years

Looking out to 2025, we see that nearly all respondents expected to deploy observability capabilities like network monitoring, security monitoring, log management, and more.

The majority of respondents indicated that they would have most capabilities (88–97%) deployed by 2025. Over the next two to three years:

  • Most (60%) expected to deploy 1–5 new capabilities
  • Just 8% expected to deploy 6 or more new capabilities
  • About a third (32%) did not expect to deploy new capabilities (presumably because they will have already been deployed)
  • One respondent indicated that they expect to deploy all 17 capabilities

And even among some of the least commonly deployed capabilities, like Kubernetes monitoring, we saw that an overwhelming 88% of respondents said they had deployed it or expected to have it deployed in three years.

88–97%

of 17 different observability capabilities should be deployed by 2025

Deployment plans summary

Forward-looking enterprise leaders are implementing observability as a business imperative. It’s interesting to see how aggressively respondents expect to have most capabilities deployed in the next year and the next three years.

By 2025, 88–97% of 17 different observability capabilities are expected to be deployed. Very few of our survey respondents did not expect to deploy these observability capabilities (2–7%).

This stated intent to deploy a large number of observability capabilities is one of the most eye-opening results from this study as it suggests that most organizations may have robust observability practices in place by 2025. The finding highlights the current state of observability and growth potential in the near future.

Capabilities deployment summary for 2022 through 2025

With remote work, there is the need for more monitoring and automatic alerting. There is a need for tools that fully monitor all the aspects and alert them quickly. I think it’s going to move to a much higher percentage. You’re probably going to end up easily closing in on 90% of them being monitored.

Budget plans

So, how do respondents’ budget plans align with their aforementioned aggressive observability capability deployment plans? We asked respondents about their observability budget plans for the next year and found that:

  • More than half (52%) expect observability budgets to increase (38% somewhat and 14% significantly or extensively)
  • One out of five respondents expect to maintain their observability budgets (within + or - 5%)
  • Only 27% expect to decrease their observability budgets (12% somewhat and 15% significantly or extensively)

Surprisingly, respondents who expected to deploy the least number of capabilities (0–3) in the next year were the most likely to say their observability budgets will increase or stay the same (80% range). While those who expected to deploy the most capabilities (4–14) were, for the most part, more likely to say their observability budgets will decrease. This could indicate that those who make decisions about what to deploy are not aligned with those who make budget decisions—or that decision-makers are using or expect to use an observability vendor that doesn’t charge extra for each capability (like New Relic).

Those who had achieved full-stack observability (based on our definition) were more likely to say their observability budgets will increase or stay the same over the next year (79%), compared to those who hadn’t achieved full-stack observability (69%).

And those who had a mature observability practice (based on our definition) were also more likely to say their observability budgets will increase or stay the same over the next year (86%), compared to those who didn’t have a mature observability practice (71%).

Interestingly, those who saw observability as completely for incident response/insurance were also the most likely to expect their observability budgets to increase or stay the same over the next year (83%). While those who saw observability as completely for core business goals were less likely to expect to increase their observability budgets or keep them the same (70%).

All-in-all, observability continues to be a budget priority for organizations.

Observability tools budget change expectations over the next year

  • Regional insight: Respondents in North America were more likely to say they expect to increase their budgets in the next year than those from other regions (63%, compared to 51% for those surveyed in Asia Pacific and 45% for those surveyed in Europe). Download the full report to view more regional highlights.
  • Role insight: Over the next year, 57% of executives expected to increase their budgets for observability tools, including 16% who expected to increase them significantly or extensively.
  • Organization size insight: Respondents from large organizations were more likely to say their observability budgets will increase over the next year (57%), midsize to say they will decrease (30%), and small to say they will stay the same (28%).

Market opportunities

We also wanted to know what other types of technology respondents foresee their organizations most needing observability for in the next three years. The survey results show that:

  • More established technologies that are likely to be on executive priority roadmaps—including artificial intelligence (AI) and the Internet of Things (IoT)—stood out as the top choice (mid-40% range)
  • Second-wave technologies or those on the way out—including business applications, 5G, blockchain, and edge computing—were all in the low- to mid-30% range
  • Emerging technologies or those on the way up—including cloud gaming, indiscriminate personalization, super apps, Web3, and the metaverse—were all in the low-20% range or slightly less

As observability can help make newer technologies like AI, 5G, and blockchain more manageable to deploy and leverage as a competitive advantage, their prioritization isn’t surprising.

Download the full report to learn more about each of these 11 technologies.

Technologies that respondents foresaw most needing observability for in the next three years

  • Regional insight: Respondents surveyed in North America were more likely to foresee needing observability for AI in the next three years (52%, compared to 46% for those surveyed in Asia Pacific and 43% for those surveyed in Europe). Those surveyed in Asia Pacific were slightly more likely to select indiscriminate personalization such as real-time digital experience (26%, compared to 19% for those surveyed in Europe and 22% for those surveyed in North America). While those surveyed in Europe were less likely to select blockchain (29%, compared to 35% for those surveyed in Asia Pacific and 34% for those surveyed in North America). Download the full report to view more regional highlights.
  • Role insight: In the next three years, executives were more likely to foresee needing observability for AI (51%, compared to 41% for non-executive managers and 46% for practitioners) and edge computing (38%, compared to 31% for non-executive managers and 29% for practitioners).
  • Industry insight: IoT was the top choice for respondents in several industries, including education (51%), energy/utilities (61%), financial/insurance (42%), healthcare/pharma (47%), and industrials/materials/manufacturing (43%). Respondents from the industrials/materials/manufacturing industry were the only ones to select business apps such as enterprise resource planning (ERP) and customer relationship management (CRM) as their top choice (43%, tied with IoT), while it was the second choice for retail/consumer respondents (40%). Respondents from the energy/utilities and services/consulting industries were more likely to select edge computing than those from other industries (42% and 45% respectively). Those from energy/utilities were also more likely to select metaverse than those from other industries (31%). While 5G was the third choice for IT/telco respondents (36%). Download the full report to view more industry highlights.