Slow is officially the new down. That’s a major finding of The SRE Report 2025, with 53% of study respondents agreeing with this expression, and 44% stating “performance should be tracked against a service-level objective.”
In the context of the SRE Report, the research was ostensibly talking about web performance and the need to deliver fast, personalized and seamless digital experiences. The report goes on to say, “poor [app] performance is as bad as complete downtime or unavailability,” and that IT should measure performance against experience-based objectives (e.g., ensure 95% of our users can check out on our site in under 5 seconds).
This got me thinking. How can you litmus test the importance of performance for yourself without all this jargon? (Think of this advice: “Before you marry a person, you should first make them use a computer with slow Internet to see who they really are.”) Getting to the more important, serious question, what can organizations do to account for this new paradigm?
First, for funsies, here are some personal litmus tests to gauge your own view on the importance of performance.
The next time you go through a drive-through window and the entity on the other side asks, “May I take your order?” do not respond. Instead, start counting slowly: “One Mississippi … two Mississippi …” I’d venture to guess they might again ask, “May I take your order?” before you get to three or four Mississippis. Note, if you order ahead on an app, instead of ordering at the drive-through, then slow is already the new down for you!
The next time you go to place an online order and you get any form of an hourglass (e.g., the hourglass icon, a “please do not refresh this page or go back” message, etc.), be mindful of how long you wait before you start thrashing your cursor or checking your phone for the latest notifications or social media updates. It probably will only be a few seconds.
The next time you start watching a ”how-to” video, take note of how long you watch before you start thinking, “Just show me how to do it already.” Chances are, it won’t be too long before you go back and look for another video with a shorter duration.
Now that we’ve warmed up a little bit, let’s get serious. Faster websites and apps mean faster money for your organization. There is no shortage of data on the positive impact of “fast” sites versus the negative impact of “slow” sites (which really means fast vs. slow user experience).
With that in mind, let’s look at five actions to help you address the “slow is officially the new down” problem.
It’s crucial to align the perceptions of reliability practices across different managerial levels. Our research highlights a stark contrast in how various levels of managerial responsibility perceive reliability practices. If IT and business teams can’t agree on what’s broken, it becomes nearly impossible to prioritize web performance initiatives.
To address this, acknowledge the gap between IT and business perspectives and work towards bridging it. Ensure there is a clear, shared understanding of the current state by comparing “work as imagined” versus “work as actually done.” Additionally, understand how IT efforts to improve web performance can positively influence business revenue and other critical KPIs.
Tracking performance against experience-level objectives is essential. Most IT organizations focus solely on good or “green” internal IT metrics, but this doesn’t necessarily mean the users’ digital experience is also good.
Monitoring and measuring internal metrics alone won’t guarantee a good user experience. Therefore, it’s important to add outward-in, experience-based indicators, such as the number of users who complete a checkout on your site, and track them against experience-based objectives. Visualize whether you are on track to meet or miss these objectives using a burndown chart to provide a clear picture of your progress.
Leveraging AI and analytics can help you get to the answers faster. In addition to tracking experience-level objectives, use AI and analytical capabilities to look beyond simple averages. Central-tendency averages can be misleading and don’t reveal the percentage of users having different types of experiences (e.g., fast or slow).
It’s important to identify the long tails (the less frequent values), as they may represent significant revenue sources. Use advanced analytics to present full distributions of performance data and break down data by important business dimensions, such as which markets generate the most revenue. Employ AI to find trend shifts and correlations within your datasets.
Measuring the performance of your application solely at its source, such as from the same cloud where it’s hosted, provides no insight into what impacts the user experience. This is why you must apply a performance mindset to all stack components between a user and your sites or applications. Since experience-level objectives are crucial, having visibility into what affects that experience will help you address performance issues as they arise (e.g., did performance regress after this last update?).
Therefore, in addition to measuring performance at the source, measure the performance of the user’s experience — where it actually occurs — to gain a comprehensive understanding.
Lastly, lead with a culture of continual performance improvement and optimization. Performance is not a one-time effort; it requires ongoing work to improve and optimize both the frontend and backend.
As customers’ expectations continue to rise, their demand for faster and faster experiences will also continue to rise.
To meet these expectations, consider these 4C’s. Establish a culture of communication and collaboration, catalyzed by an evergreen performance mindset. This will help ensure that your organization remains focused on delivering optimal performance.
To learn more about this research, download The SRE Report 2025 or read it online (no registration required).
Slow is officially the new down. That’s a major finding of The SRE Report 2025, with 53% of study respondents agreeing with this expression, and 44% stating “performance should be tracked against a service-level objective.”
In the context of the SRE Report, the research was ostensibly talking about web performance and the need to deliver fast, personalized and seamless digital experiences. The report goes on to say, “poor [app] performance is as bad as complete downtime or unavailability,” and that IT should measure performance against experience-based objectives (e.g., ensure 95% of our users can check out on our site in under 5 seconds).
This got me thinking. How can you litmus test the importance of performance for yourself without all this jargon? (Think of this advice: “Before you marry a person, you should first make them use a computer with slow Internet to see who they really are.”) Getting to the more important, serious question, what can organizations do to account for this new paradigm?
First, for funsies, here are some personal litmus tests to gauge your own view on the importance of performance.
The next time you go through a drive-through window and the entity on the other side asks, “May I take your order?” do not respond. Instead, start counting slowly: “One Mississippi … two Mississippi …” I’d venture to guess they might again ask, “May I take your order?” before you get to three or four Mississippis. Note, if you order ahead on an app, instead of ordering at the drive-through, then slow is already the new down for you!
The next time you go to place an online order and you get any form of an hourglass (e.g., the hourglass icon, a “please do not refresh this page or go back” message, etc.), be mindful of how long you wait before you start thrashing your cursor or checking your phone for the latest notifications or social media updates. It probably will only be a few seconds.
The next time you start watching a ”how-to” video, take note of how long you watch before you start thinking, “Just show me how to do it already.” Chances are, it won’t be too long before you go back and look for another video with a shorter duration.
Now that we’ve warmed up a little bit, let’s get serious. Faster websites and apps mean faster money for your organization. There is no shortage of data on the positive impact of “fast” sites versus the negative impact of “slow” sites (which really means fast vs. slow user experience).
With that in mind, let’s look at five actions to help you address the “slow is officially the new down” problem.
It’s crucial to align the perceptions of reliability practices across different managerial levels. Our research highlights a stark contrast in how various levels of managerial responsibility perceive reliability practices. If IT and business teams can’t agree on what’s broken, it becomes nearly impossible to prioritize web performance initiatives.
To address this, acknowledge the gap between IT and business perspectives and work towards bridging it. Ensure there is a clear, shared understanding of the current state by comparing “work as imagined” versus “work as actually done.” Additionally, understand how IT efforts to improve web performance can positively influence business revenue and other critical KPIs.
Tracking performance against experience-level objectives is essential. Most IT organizations focus solely on good or “green” internal IT metrics, but this doesn’t necessarily mean the users’ digital experience is also good.
Monitoring and measuring internal metrics alone won’t guarantee a good user experience. Therefore, it’s important to add outward-in, experience-based indicators, such as the number of users who complete a checkout on your site, and track them against experience-based objectives. Visualize whether you are on track to meet or miss these objectives using a burndown chart to provide a clear picture of your progress.
Leveraging AI and analytics can help you get to the answers faster. In addition to tracking experience-level objectives, use AI and analytical capabilities to look beyond simple averages. Central-tendency averages can be misleading and don’t reveal the percentage of users having different types of experiences (e.g., fast or slow).
It’s important to identify the long tails (the less frequent values), as they may represent significant revenue sources. Use advanced analytics to present full distributions of performance data and break down data by important business dimensions, such as which markets generate the most revenue. Employ AI to find trend shifts and correlations within your datasets.
Measuring the performance of your application solely at its source, such as from the same cloud where it’s hosted, provides no insight into what impacts the user experience. This is why you must apply a performance mindset to all stack components between a user and your sites or applications. Since experience-level objectives are crucial, having visibility into what affects that experience will help you address performance issues as they arise (e.g., did performance regress after this last update?).
Therefore, in addition to measuring performance at the source, measure the performance of the user’s experience — where it actually occurs — to gain a comprehensive understanding.
Lastly, lead with a culture of continual performance improvement and optimization. Performance is not a one-time effort; it requires ongoing work to improve and optimize both the frontend and backend.
As customers’ expectations continue to rise, their demand for faster and faster experiences will also continue to rise.
To meet these expectations, consider these 4C’s. Establish a culture of communication and collaboration, catalyzed by an evergreen performance mindset. This will help ensure that your organization remains focused on delivering optimal performance.
To learn more about this research, download The SRE Report 2025 or read it online (no registration required).
This is some text inside of a div block.
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。