2024 State of Observability for Media and Entertainment

Compared to all other sectors surveyed, the media and entertainment industry has unique technology adoption priorities. These organizations are 34% more likely to cite containerization of applications and workloads as a key driver for observability adoption (30% compared to 23% overall and the second highest of all industries) and were 23% more likely to say that adopting Internet of Things (IoT) technologies was a catalyst for observability adoption (33% versus 27% overall).

Conversely, cost management holds less significance, with a 42% lower likelihood of being a driver of observability adoption (19% versus 33%, which is the lowest of all industries surveyed), and being 21% less likely to integrate business applications like Enterprise Resource Planning or Customer Relationship Management into workflows (28% versus 35% overall). This forward-looking approach highlights that innovation drives the industry, not cost constraints.

Media and entertainment organizations are looking to observability to support them in delivering on many strategies and trends, including an increased focus on security, governance, risk, and compliance (39%), the adoption of artificial intelligence (AI) technologies (35%), migration to a multi-cloud environment (35%), and the adoption of the IoT technologies (33%). The observability drivers for the media and entertainment industry were notably different from the rest of the sectors surveyed.

In terms of how AI can support observability adoption, more than a third (35%) of media and entertainment respondents believe the AI-assisted generation of runbooks would improve their organization’s observability practice most, followed by AI-assisted remediation actions like rollbacks or configuration updates (33%), automatic root cause analysis (32%), and forecasting and predictive analytics (32%).

A fifth (20%) of media and entertainment organizations have achieved full-stack observability, lower than the 25% rate overall. The top challenge preventing these organizations from achieving full-stack observability was having a complex tech stack (39%). They also felt hampered by a lack of strategy (34%) and budget (28%), having too many monitoring tools and siloed data (28% each). Media and entertainment respondents were the most likely industry to cite a lack of strategy as a challenge preventing full-stack observability (34%); 36% more likely than the average of all other respondents.

When it came to the number of observability tools adopted, 59% of media and entertainment organizations were more likely to use five or more tools out of the 19 observability capabilities included in this study - the highest of all industries. The average number of tools used by media and entertainment organizations is 5, higher than the 4.5 across all respondents. Just 4% of media and entertainment respondents used a single tool for observability, indicating that tool proliferation is widespread.

When asked how unified their telemetry data – metrics, events, logs, traces (MELT) – was, these organizations report the third-most siloed telemetry data (47%) and second-lowest proportion of unified data (30%) compared to other industries.

Despite having siloed data, respondents were more likely than average to integrate more business-related data types into their telemetry data. They integrated nine of 10 data types at higher rates than average.

Operations data (52%), marketing data (49%), and customer data (48%) were the most integrated business-related data types for these organizations. Unified telemetry data and more integrated business-related data types tend to have several benefits. Across all respondents in the broader survey, those with more unified telemetry data on average:

• Experienced 78% less annual downtime (107 hours per year compared to 488 hours per year)
• Spent 11% less engineering time addressing disruptions (28% compared to 32%)
• Had a 4% higher median ROI (302% compared to 290%).

And compared to those who had less than five business-related data types currently integrated with their telemetry data, those who had integrated five or more:

• Spent 32% less on hourly outage costs ($1.5 million per hour for high-business-impact outages compared to $2.2 million per hour)
• Experienced 63% less annual downtime (139 hours compared to 370 hours)
• Spent 27% less engineering time addressing disruptions (11 hours compared to 15 hours based on a 40-hour work week).

These findings strongly suggest that media and entertainment organizations benefit from integrating more types of business-related data, but may miss out on the benefits associated with more unified telemetry. Moving away from siloed telemetry could help them better benefit from these gains.

Many IT teams in media and entertainment organizations learn about software and system interruptions without observability tools. Some 57% reported they unearthed these interruptions via manual checks and tests using monitoring tools, or through complaint and incident tickets. This figure was the highest of all industries surveyed. Likewise, only 43% relied on observability tools, the lowest of all industries. Across all respondents, 54% learned about interruptions from observability tools, and 46% without observability tools. This highlights that media and entertainment organizations are much more dependent on incident tickets for detecting problems, and manual methods for resolution.

In today's rapidly evolving digital landscape, media and entertainment companies are constantly seeking ways to stay ahead of the competition. One key aspect of this is regular use of an observability tool, with about three-quarters (76%) of media and entertainment respondents saying they personally use their organization’s observability solution once a week or more.

While the preference among media and entertainment respondents was for a single, consolidated observability platform (56%), just 27% planned to consolidate tools in the next year to get the most value out of their observability spend—the lowest of all industries. Scalability—the volume, frequency, cardinality, and retention of data (34%)—and the breadth of supported capabilities and features (34%) are the most important features media and entertainment organizations look for in an observability solution, followed by onboarding time and ease of implementation (32%).

Out of all industries, media and entertainment organizations experienced the most high-business-impact outages. 63% said these took place at least once a week, compared to the average of 38% among all respondents. In the past two years, network failures (32%) and power failures (32%) were the most common cause of unplanned outages, followed by the deployment of software changes (29%) and hardware failures (29%).

Organizations in this sector also take longer to detect and solve issues surrounding outages. More than two-thirds (69%) of respondents said it takes at least 30 minutes to detect high-business-impact outages, the highest of any industry and much higher than the average of 54%. The median mean time to detect (MTTD) high-business-impact outages was 56 minutes, which was 51% higher than the median and the highest of any industry.

When it came to resolving outages, more than two-thirds (69%) said it takes at least 30 minutes to resolve high-business-impact outages, and a third (33%) said it takes at least an hour. For mean time to resolution (MTTR) of high-business-impact outages, the median time for the media and entertainment industry was 51 minutes, the same as the median MTTR across all respondents.

Nearly three-quarters (73%) said MTTR has improved to some extent since adopting an observability solution. Tracking, reporting, and incentivizing MTTD and MTTR improvements (49%), monitoring DevOps Research and Assessment (DORA) metrics (48%), conducting root cause analysis (RCA) and post-incident reviews (37%), and monitoring golden signals, such as latency, utilization, errors, and saturation (35%) were the top observability practices that media and entertainment organizations found helpful in reducing downtime and improving MTTD and MTTR.

Notably, media and entertainment organizations with full-stack observability were more likely to report a faster MTTD and MTTR for high-business-impact outages than organizations that had not achieved full-stack observability. For instance, more than half (57%) of those who took less than 30 minutes to detect high-business-impact outages had full-stack observability (compared to 17% without full-stack observability), with this figure the same for those resolving issues in less than 30 minutes (compared to 14% without full-stack observability).

About three-quarters (74%) of media and entertainment respondents said critical business app outages cost at least $1 million per hour of downtime. The median outage cost of high-business-impact outages for media and entertainment organizations was $2.2 million per hour, 16% higher than the median hourly outage cost across all respondents.

Engineering time is also an important factor in understanding the overall costs of outages for organizations. Two-thirds (68%) of media and entertainment respondents said their engineering team spends at least 25% of their time addressing outages, with 48% saying they spend at least half of their time on this. The median estimated percentage of engineering team time spent addressing disruptions was 49%, which is the highest of any industry and 63% higher than the overall median of 30%.

With streaming services looking to subscriptions and advertising to drive revenue, uptime is a crucial consideration for media and entertainment organizations to address. Every minute of downtime costs money, and if a company website or app goes down for even a few minutes on a peak-traffic day, like a major sporting event or high-budget show premiere, it could cost these providers millions of dollars and negatively influence brand perception. The stakes are high for media and entertainment organizations, but observability can reduce the risk of downtime.

Media and entertainment organizations spent more on observability than average, with 82% saying they spent $1 million or more on observability, and no respondents spending less than $100,000 per year on observability. These organizations spent a median of $2.6 million per year on observability, which is the highest of any industry, and 33% higher than the overall median of $1.95 million.

The survey asked how IT decision makers (ITDMs) and practitioners benefit from observability adoption. ITDMs say it makes their job easier (37%), helps them achieve business key performance indicators (KPIs) (37%), drives business strategy (37%), and helps prioritize environment updates and new service rollouts (37%).

For practitioners, observability creates less guesswork when managing complicated and distributed tech stacks (44%), increases productivity (39%), and increases innovation (35%). More than any other industry, practitioners in the media and entertainment industry cited observability as a source of innovation, with three times more media and entertainment respondents saying it was used more for core business goals (62%) compared to incident response or insurance (19%).

In terms of the benefits of observability, nearly half (46%) of media and entertainment respondents said observability increases operational efficiency, 39% said it improves system uptime and reliability, and 38% said it results in business and/or revenue growth. At least a third also noted that observability improves customer experience (37%) and developer productivity (33%).

When asked how much total value media and entertainment organizations receive from their observability investment per year, 90% said the total annual value was $1 million or more while 51% said the total value was $10 million or more. Media and entertainment organizations reported a median $10 million total annual value from observability. This was 23% higher than the $8.15 million value received across all respondents and the third highest by industry.

Based on annual spend and annual value received estimates, media and entertainment organizations receive a 296% median annual return on investment (ROI), or 4x.

Media and entertainment organizations had ambitious observability deployment plans for the next one to three years. For example, by mid-2027, nearly all expect to have deployed network monitoring (99%), security monitoring (99%), database monitoring (99%), browser monitoring (98%), and AI monitoring (98%).

DEM is also an important focus. In the next one to three years, more than half (52%) expected to deploy synthetic monitoring, 47% expected to deploy browser monitoring, and 44% expected to deploy mobile monitoring. These findings indicate that by mid-2027, the vast majority expected to have deployed browser monitoring (98%), mobile monitoring (96%), and synthetic monitoring (94%).

To get the most value out of their observability investments in the next year, 42% planned to reduce spend across the board, and 42% planned to train staff on how to best use their observability tools. More than a third also planned to switch to a more affordable vendor (38%) and optimize their engineering team size (36%).