現在、このページは英語版のみです。
サブスクライバー限定
サインアップして限定コンテンツにアクセスしましょう。
Success! Here it comes.
Content automatically in 3...

0
サインアップすると、 利用規約​ ​および サービスプライバシー通知.

Published October 2024

Overview

The 2024 Observability Forecast in Asia Pacific provides insights into the state of observability, identifying key areas of growth and stagnation, and uncovering how external forces are shaping adoption and investment strategies. With input from 800 technology professionals across nine countries, it is one of the largest and most comprehensive studies of the observability industry in the region. 

Despite the culturally diverse nature of Asia Pacific, similar themes emerged. For example, those surveyed in Asia Pacific experienced much more frequent high-business-impact outages than any other region, with 53% reporting at least once a week (compared to 31% in Europe and 26% in the Americas).

map of APAC region FPO

The high number of outages could be due to the fact that those surveyed in Asia Pacific were most likely to use more than five tools (55% compared to 43% for those in Europe and 35% for those in the Americas), which is leading to increased silos of information across both data and teams. In fact, many countries like Australia/New Zealand and South Korea reported challenges with siloed data. 

The adoption of AI technologies was the top strategy or trend driving the need for observability (36%)— reflecting the growing importance of observability in supporting innovation. Respondents surveyed in Asia Pacific were more likely to deploy these compared to Europe. Organizations that deploy AI-driven observability reported higher business value and return on investment (ROI) overall.

Respondents surveyed in Asia Pacific were most likely to say the primary benefit of observability is business and/or revenue growth (31%). This year’s data also revealed that respondents in Asia Pacific had a higher median annual value from observability investments than other regions, receiving $10.08 million in value compared to $7.05 million in Europe and $5.40 million in the Americas.

Key findings

The median hourly cost from an outage is $2.3 million.

The median annual downtime from a high-impact outage is 109.8 hours (approximately 4.5 days), with an hourly cost of up to $2.3 million. This is double the figure in Europe and in the Americas. Asia Pacific also experienced much more frequent high-impact outages than any other region, with 53% reporting a frequency of at least once a week.

Capacity constraints and unexpected traffic surges are causing outages.

Survey takers in the Asia-Pacific region were more likely to contend with capacity constraints (21% compared to 18% for those in the Americas and 15% for those in Europe) and unexpected traffic surges (22% compared to 18% for those in the Americas and 16% for those in Europe) as common causes for outages.

32% of respondents said they plan to consolidate tools in the next year.

More than half (55%) were using 5+ observability tools—more than any other region. Those in the Asia-Pacific region are less likely than other regions to plan to consolidate tools to get more value out of their observability spend (32% compared to 50% for the Americas and 43% for Europe).

Organizations are embracing observability to capitalize on AI technologies and business applications.

The adoption of AI technologies was the top strategy or trend driving the need for observability (36%), followed by an increased focus on security, governance, risk, and compliance (34%), and the integration of business apps like enterprise resource planning (ERP) and customer relationship management (CRM) into workflows (34%).

Asia-Pacific highlights

When we looked at the results for each country, the diversity in cultures and business practices across the Asia-Pacific region is apparent:

Association of Southeast Asian Nations 🇮🇩 🇲🇾 🇸🇬 🇹🇭 finds its complex tech stacks to be a substantial barrier to achieving full-stack observability. Tool consolidation and data integration remain priorities, and those consolidating to just one observability tool grew over last year.

Australia and New Zealand 🇦🇺 🇳🇿 struggled with tool sprawl and siloed data, and their artificial intelligence for IT operations (AIOps) deployment is low. However, future plans are ambitious, with those in these countries pursuing tool consolidation, data integration, and AIOps capabilities.

India 🇮🇳 experienced some of the most frequent and costly outages compared to other countries. Tool sprawl is common, but many respondents in India seem resolved to consolidate tools in the future.

Japan 🇯🇵 experienced data data integration challenges. A lack of skills is seen as a top barrier to achieving full-stack observability, but plans to better train staff are a priority.

South Korea 🇰🇷 experienced frequent outages, and its mean time to detection (MTTD) and mean time to resolution (MTTR) were among the slowest of all countries. Outages are more costly compared to other countries, but they receive considerable business value from observability and enjoy a strong return on their investments.

State of observability in the Association of Southeast Asian Nations 🇮🇩 🇲🇾 🇸🇬 🇹🇭

In the Association of Southeast Asian Nations (ASEAN) where we surveyed respondents in Indonesia, Malaysia, Singapore, and Thailand, tool consolidation was common, observability was seen as an enabler of business growth, and machine learning (ML) model monitoring was highly deployed. Adopting AI technologies (38%), the integration of business apps (34%), and the migration to a multi-cloud environment (34%) were the key drivers of observability.

78%

in ASEAN said high-business-impact outages cost at least $1 million an hour

Tool consolidation and data integration are priorities

Nearly half (49%) of respondents use five or more tools for observability, compared to 45% overall. Notably, 14% said they were using just one tool—a significant increase from 3% in 2023. Furthermore, 63% preferred a single, consolidated platform, while only 18% favored multiple point solutions. As organizations shift toward fewer tools, ASEAN stands out as the least likely region in Asia Pacific to report that too many monitoring tools and siloed data prevented full-stack observability (28%).

ASEAN had more unified telemetry data (38%) compared to siloed data (30%). With the exception of communications and logistics data, ASEAN was more likely than average to integrate business-related data types with telemetry. For example, they were more likely to integrate operations data (50% compared to 43% overall), inventory data (46% compared to 36% overall), and customer data (44% compared to 41% overall).

Full-stack observability is on the rise, along with high deployments of ML model monitoring

Notably, 40% in Indonesia had achieved full-stack observability by the report’s definition—making it the top country in the region (compared to just 20% across ASEAN). In Singapore, only 10% had achieved full-stack observability. Similarly, 65% in Indonesia had deployed 10 or more capabilities, while only 20% in Singapore had reached that level. Those in Singapore cited a complex tech stack (60%) and resistance to change (50%) as primary barriers to achieving full-stack observability. Across ASEAN, a complex tech stack (36%) and lack of budget (30%) were the top challenges preventing full-stack observability.

Security monitoring was the most deployed capability in ASEAN (55%), followed closely by infrastructure monitoring (54%). ML model monitoring was the least deployed (21%). However, in the next one to three years, ASEAN respondents planned to deploy ML model monitoring the most (73%), followed by distributed tracing (69%).

Outages are frequent and expensive

About a third (33%) of ASEAN respondents experienced high-business-impact outages once a week or more (Thailand had the highest outage frequency, and Malaysia the lowest); 75% take at least 30 minutes to detect them, and 72% take at least 30 minutes to resolve them. Most (87%) estimated that business-critical application outages cost their organization at least $500,000 per hour of downtime, and 78% estimated they cost $1 million or more. The median outage cost for high-business-impact outages in ASEAN was $2.5 million per hour, 32% higher than the median $1.9 million per hour outage cost across all respondents. Notably, 80% in Singapore said their MTTR improved to some extent since adopting observability (more than any other country), compared to 56% for Indonesia.

More than a quarter of respondents in ASEAN (27%) learned about these outages with multiple monitoring tools, and 22% with manual checks, tests, or complaints. Only 18% learned about them with just one observability platform, with respondents in Singapore being the most likely to do so at 30%.

ROI and business value for observability are high

Notably, 80% of ASEAN respondents spent $1 million or more on observability per year, with the median annual spend on observability at $2.70 million (compared to $1.95 million overall). 

Malaysia had a median annual ROI of 302%, the highest among ASEAN countries and second-highest in the Asia-Pacific region. Thailand followed closely with a median annual ROI of 300%, while Singapore achieved 258%. In addition, 87% of respondents said their organization recorded at least $1 million in total value per year from their observability investment. More than two-fifths (42%) said observability reduced security risks, and 37% said it improved system uptime and reliability. Those in ASEAN countries were more likely than respondents from other countries to view observability as a key enabler for achieving core business goals to some extent (62% compared to 50% overall).

State of observability in Australia and New Zealand 🇦🇺 🇳🇿

The most popular strategies or trends driving the need for observability in Australia and New Zealand were the integration of business apps such as ERPs and CRMs into workflows (31%), increased focus on security, governance, risk, and compliance (30%), adoption of open-source technologies (29%), and the development of cloud-native application architectures (28%).

63%

in Australia and New Zealand preferred a single, consolidated observability platform over multiple point solutions

Observability adoption plans are ambitious, and being led by AIOps

Only 18% of respondents in New Zealand and 17% of respondents in Australia had achieved full-stack observability (by the report’s definition), lower than the average across all countries. More than one-third (38%) in New Zealand, but only 17% in Australia, had deployed 10 or more capabilities.

Nearly half (46%) had implemented network monitoring, followed by security monitoring (45%), infrastructure monitoring (43%), and browser monitoring (43%). AIOps was the least deployed capability, at just 17%. Plans for next year were ambitious, with 53% aiming to implement AIOps, 44% planning for Kubernetes monitoring, and 43% for dashboards. All 19 observability capabilities included in the report were expected to be deployed in the next one to three years by at least half of respondents in Australia and New Zealand.

Frequent outages are costly, but observability helps

High-business-impact outages occurred fairly frequently: 56% experienced them once per week or more (compared to 38% overall), including 11% who experienced them at least once per day. More than three-fifths (64%) took at least 30 minutes to detect them, and almost three-quarters (74%) took more than 30 minutes to resolve them.

This downtime has a substantial impact on the bottom line, as 86% reported that critical business application outages cost their organization $500,000 or more per hour of downtime, including 79% that said such outages cost $1 million or more per hour. The median outage cost for high-business-impact outages for Australia and New Zealand was $2.2 million per hour, which is 16% higher than the median $1.9 million per hour cost across all respondents. However, observability helped mitigate these costs—65% of respondents said their MTTR improved to some extent since adopting observability, including 29% who reported a 25% or more improvement. Nearly half (48%) said they monitored DevOps Research and Assessment (DORA) metrics to reduce their downtime.

Siloed data and tool sprawl remain a challenge

Australian and New Zealand respondents claimed more siloed telemetry data (40%) than unified (32%). They were also less likely to integrate five or more types of business-related data into their telemetry data (24% compared to 35% overall).

More than half (56%) were using five or more tools for observability, while only 1% used just one tool this year compared to 6% last year. However, 63% preferred a single, consolidated observability platform when compared to multiple point solutions (20%). About a third (31%) planned to consolidate tools over the next year to get the most value from their observability spend. In addition, 39% said too many monitoring tools and siloed data are primary challenges to achieving full-stack observability.

ITDMs see the business value of observability

Nearly half (49%) of IT decision makers (ITDMs) in Australia and New Zealand said observability improved their life most by enabling data visualization from a single dashboard. Many also reported that observability helped achieve technical (42%) and business (38%) key performance indicators (KPIs). Practitioners’ top responses for how observability improved their lives included increased productivity (38%), reduced guesswork when managing complex and distributed tech stacks (36%), and better time prioritization (35%).

Most (89%) said their organization received a total annual value of $1 million or more from their observability investment, with 79% reporting $5 million or more. Australian survey takers realized a 3.9x return on their observability investment, while those in New Zealand saw 4x ROI.

State of observability in India 🇮🇳

While Indian organizations had strong observability adoption and received considerable value from their investments, they struggled with tool fragmentation, low uptime, and high outage costs

57%

in India used 5+ tools for observability

AI, security, and business apps are driving the need for observability

In India, the biggest technology strategy or trend driving the need for observability was the adoption of AI technologies (39%), followed by an increased focus on security, governance, risk, and compliance (37%), and the integration of business apps like ERP and CRM into workflows (36%). The least popular trends were the adoption of serverless computing (24%), followed by the containerization of applications and workflows (25%). A significantly larger portion (60%) viewed observability as enabling core business goals rather than for incident response or insurance (14%).

Tool sprawl is widespread, but consolidation is a goal

More than half (57%) of respondents used five or more tools, with 3% using 10 or more. Just 5% used only one tool, and India was the least likely of all countries to prefer a single, consolidated platform to some degree (43% compared to 53% overall). However, 40% reported that too many monitoring tools and siloed data posed a challenge to achieving full-stack observability, and more than a third (35%) planned to consolidate tools to maximize the value of their observability investment.

Downtime and outage costs are high

Nearly two-thirds (65%) of respondents in India experienced high-business-impact outages at least once per week, with 31% stating they experienced them once a day or more, the highest of all countries. Additionally, detection and response were relatively slow. More than two-thirds (68%) reported it took more than 30 minutes to detect outages, including 32% who indicated it took more than 60 minutes. Furthermore, 71% noted it took more than 30 minutes to resolve outages. However, 67% said their MTTR improved to some extent since adopting observability. Nearly half (44%) of respondents indicated that they tracked, reported, and incentivized MTTx metrics to reduce downtime.

More than four-fifths (84%) said that the cost of significant downtime for critical business app outages was $500,000 or more per hour, including 77% who said it cost more than $1 million per hour. The median hourly cost for high-business-impact outages in India was $2.0 million, slightly higher than the median $1.9 million per hour outage cost across all respondents. The median hourly cost for high-business-impact outages in India was $2.0 million, slightly higher than the median $1.9 million per hour outage cost across all respondents.

Observability adoption is extensive and provides business value

Almost a quarter (22%) achieved full-stack observability by the report’s definition, which was on par with the average of 25% overall. Additionally, 38% had 10 or more observability capabilities deployed. AI monitoring was the most deployed capability (57%), followed by dashboards (53%) and application performance monitoring (APM; 50%). Synthetic monitoring was the least deployed (29%), followed by Kubernetes monitoring (30%) and AIOps (30%). However, 65% planned to deploy AIOps in the next one to three years.

More than three-quarters (77%) spend $1 million or more on observability per year, with a median annual spend of $2.43 million. Most respondents in India (85%) indicated $1 million or more in total value per year from that investment, with 70% receiving $5 million or more in value. The median annual ROI was 294%, or 3.9x. About a third (36%) reported that observability improved business and/or revenue growth. Other top benefits of observability included improved system uptime and reliability (42%), increased operational efficiency (42%), and improved developer productivity (40%).

State of observability in Japan 🇯🇵

In Japan, the top strategies or trends driving the need for observability were the adoption of AI technologies (46%), an increased focus on security, governance, risk, and compliance (38%), and cost management (34%).

22%

in Japan said they had achieved full-stack observability

Outages can be costly and MTTR is improving with observability

A third (33%) of respondents in Japan experienced high-business-impact outages once per week or more (compared to 38% overall). Three-fifths (61%) said it took at least 30 minutes to detect outages, and 72% took more than 30 minutes to resolve them, including 48% that required at least 60 minutes. More than half (56%) indicated that their MTTR improved to some extent since adopting observability, and 46% of practitioners said observability increases productivity. Additionally, more than half (51%) noted that improved system uptime and reliability was a primary benefit of observability.

Almost two-thirds (66%) said that critical business app outages cost $1 million or more per hour of downtime; however, this was a lower proportion than any other country in the Asia-Pacific region. Curiously, Japan’s median outage cost for high-business-impact outages was $2.8 million per hour, the highest of any country. This suggests that respondents in Japan who experienced the costliest outages endured far higher than $1 million per hour impacts.

Observability provides business value and ROI

Most respondents (73%) reported spending more than $1 million annually on observability. Japan’s median annual observability spend of $2.18 million is higher than the median of $1.95 million across all countries. 

Most respondents (84%) said their organization receives $1 million or more total value per year from that investment, including 68% who said they receive more than $5 million per year. Japan’s median annual ROI is 313%, or 4.1x, slightly higher than the 295%, or 4x, median ROI across all countries. In addition, 58% said observability is a key enabler to achieving their organization’s core business goals to some extent.

Unified data and consolidated tools are preferred

Almost half (48%) of respondents in Japan used five or more tools for observability, while only 2% used just one tool. More than half (59%) learn about interruptions via observability, including 43% who used multiple monitoring tools—the highest in Asia Pacific. Nearly half (46%) expressed a preference for a single, consolidated platform for observability, while 36% favored multiple point solutions. 

Respondents were also more likely to report that their telemetry data is more unified (44%) rather than more siloed (26%). A quarter (25%) integrated five or more business-related data types into their telemetry data, but Japan is below average in each type. For example, Japan integrated only 17% of production data (compared to 38% overall) and 19% of logistics data (compared to 34% overall).

Ambitious observability deployment and training are future priorities

More than a fifth (22%) had achieved full-stack observability by the report’s definition. More than three-fifths (62%) had deployed five or more capabilities, and about a quarter (23%) had deployed 10 or more. About half had deployed log management (53%) and network monitoring (50%), and 45% had deployed application performance monitoring. AIOps (18%) and ML model monitoring (20%) were among the least deployed capabilities for Japan, but future deployment plans were ambitious. More than half (55%) plan to deploy AIOps and 61% plan to deploy ML model monitoring within the next one to three years.

State of observabilit in South Korea 🇰🇷

Integration of business applications into workflows such as ERPs and CRMs is the top strategy or trend driving observability in South Korea (35%). Development of cloud-native architectures (34%), increased focus on security, governance, risk, and compliance (32%), and adoption of AI technologies (30%) are also prominent adoption trends.

83%

in South Korea took 30+ minutes to detect high-business-impact outages—more than any other country

Downtime is high and costly

Three-fifths (61%) of respondents in South Korea experienced high-business-impact outages once or more per week—83% reported that it took at least 30 minutes to detect outages (more than any other country), and 85% said it took more than 30 minutes to resolve an outage (more than any other country). 

Most (92%) estimated that critical business app outages cost at least $1 million per hour of downtime, the highest rate among all countries. South Korea’s median outage cost was $2.6 million per hour, the second-highest among all countries and 37% higher than the median $1.9 million per hour outage cost across all respondents. Those in South Korea also indicated that more than half (57%) of their engineering team’s time is spent addressing disruptions, which was higher than all other countries.

Observability provides significant business value

Almost all respondents in South Korea (98%) reported that their organization receives at least $1 million in total value per year from its investment, with 70% receiving $10 million or more. 

The median annual ROI for South Korea was 298%, or 4x. In addition, business and/or revenue growth (37%) were cited as the second-highest observability benefit, just behind the ability to handle traffic surges (39%). Nearly half (48%) indicated that observability was a key enabler for achieving their core business goals to some extent.

Observability deployment is low, but plans to increase are underway

Only 6% had achieved full-stack observability by the report’s definition, and only 9% had deployed 10 or more observability capabilities—considerably less than every other country. AI monitoring was the most deployed capability in South Korea (44%), followed by security monitoring (31%), log management (30%), and error tracking (30%). However, South Korea had ambitious deployment plans for the next three years, mobile monitoring (73%), distributed tracing (70%), serverless monitoring (69%), and business observability (68%).

Tool sprawl is widespread, and data integration is low

While respondents surveyed in South Korea preferred a single, consolidated observability platform (46%) to multiple point solutions (39%), only 22% said they plan to consolidate tools next year to get the most value from their observability investment—the lowest among all countries. More than three-fifths (63%) used five or more tools for observability, including 22% who used eight or more, while just 6% used a single tool. Notably, a third (33%) reported that having too many tools was a challenge to achieving full-stack observability, indicating an awareness of the limitations associated with tool sprawl.

Telemetry data in South Korea was both siloed and unified, with 42% saying it was more siloed, while 45% said it was more unified. Only 16% integrated five or more business-related data types into their telemetry data, with only Singapore integrating fewer (10%).

Summary

The 2024 State of Observability in Asia Pacific report results show that data, team, and tool fragmentation persists, observability provides significant business value, and organizations plan to invest more in the future.

About this report

All data in this report are derived from a survey, which was in the field from April to May 2024 as part of our research and work in publishing the 2024 Observability Forecast report. Asia Pacific comprised 800 of the total respondents, or 44%.

ETR, a technology market research firm, qualified survey respondents based on relevant expertise. ETR performed a non-probability sampling type called quota sampling to target sample sizes of respondents based on their country of residence and role type in their organizations (in other words, practitioners and IT decision makers). Geographic representation quotas targeted 16 key countries.

To avoid skewing results by industry, subsamples of n<10 are excluded from some analysis in this report.

All quotes were derived from interviews conducted by ETR with IT professionals who use observability.

All dollar amounts in this report are in USD.

About New Relic

The New Relic intelligent observability platform helps businesses eliminate interruptions in digital experiences. New Relic is the only platform to unify and pair telemetry data to provide clarity over the entire digital estate. We move problem solving past proactive to predictive by processing the right data at the right time to maximize value and control costs. That’s why businesses around the world—including Adidas Runtastic, American Red Cross, Domino’s, GoTo Group, Ryanair, Topgolf, and William Hill—run on New Relic to drive innovation, improve reliability, and deliver exceptional customer experiences to fuel growth.

About ETR

ETR is a technology market research firm that leverages proprietary data from its targeted IT decision makers (ITDMs) community to deliver actionable insights about spending intentions and industry trends. Since 2010, ETR has worked diligently at achieving one goal: eliminating the need for opinions in enterprise research, which are typically formed from incomplete, biased, and statistically insignificant data.

The ETR community of ITDMs is uniquely positioned to provide best-in-class customer/evaluator perspectives. Its proprietary data and insights from this community empower institutional investors, technology companies, and ITDMs to navigate the complex enterprise technology landscape amid an expanding marketplace.