惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Google DeepMind News
Google DeepMind News
人人都是产品经理
人人都是产品经理
H
Hacker News: Front Page
Stack Overflow Blog
Stack Overflow Blog
B
Blog
I
InfoQ
GbyAI
GbyAI
T
The Blog of Author Tim Ferriss
F
Fortinet All Blogs
Y
Y Combinator Blog
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
月光博客
月光博客
Hugging Face - Blog
Hugging Face - Blog
爱范儿
爱范儿
F
Full Disclosure
Hacker News - Newest:
Hacker News - Newest: "LLM"
Recent Announcements
Recent Announcements
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
Jina AI
Jina AI
T
Tailwind CSS Blog
S
Secure Thoughts
P
Privacy International News Feed
美团技术团队
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
L
LINUX DO - 最新话题
H
Hackread – Cybersecurity News, Data Breaches, AI and More
C
Cybersecurity and Infrastructure Security Agency CISA
Last Week in AI
Last Week in AI
W
WeLiveSecurity
Google Online Security Blog
Google Online Security Blog
P
Privacy & Cybersecurity Law Blog
D
DataBreaches.Net
Engineering at Meta
Engineering at Meta
Know Your Adversary
Know Your Adversary
P
Palo Alto Networks Blog
I
Intezer
Application and Cybersecurity Blog
Application and Cybersecurity Blog
Project Zero
Project Zero
V2EX - 技术
V2EX - 技术
H
Heimdal Security Blog
博客园 - Franky
阮一峰的网络日志
阮一峰的网络日志
D
Darknet – Hacking Tools, Hacker News & Cyber Security
T
Troy Hunt's Blog
V
Vulnerabilities – Threatpost
H
Help Net Security
Martin Fowler
Martin Fowler
cs.AI updates on arXiv.org
cs.AI updates on arXiv.org
G
GRAHAM CLULEY
博客园 - 【当耐特】

Catchpoint Blog

SRE Report: AI optimism and the economics of effort SRE Report: Why fast is what users trust SRE Report 2026: What surprised us, what didn't, and why the gaps matter most The SRE Report 2026: Defensible Ns Why Synthetic Tracing Delivers Better Data, Not Just More Data A New Chapter: LogicMonitor + Catchpoint – A Personal Note from Mehdi Mezmo + Catchpoint deliver observability SREs can rely on The four pillars holding up your digital business, and what happens when they crumble When payments pause: lessons from a global payments outage Observability 2025 Decoded: What the DZone Report Means for SLO-Driven Ops The next evolution of WebPageTest has arrived, and it’s a game-changer The Monitoring Blind Spot That Could Cost You Black Friday Powering Mexico’s Digital Future: Expanded Internet Observability with Catchpoint The Next Chapter of WebPageTest: Your New Experience Starts Soon SRE Report Retrospectives — Have AIOps Predictions Held Up? When BGP becomes UX: The inside story of a SaaS routing decision gone wrong (or right) Session Replay explained: A guide to seeing digital experience through your user’s eyes Making the invisible visible: Are your cloud firewalls and DDoS protection really working? Why it’s time to move beyond APM: Monitoring from the user’s perspective When metrics mislead: Inside the 2025 Retail Web Performance Benchmark The vendor trap: why your next outage won’t be your fault—but will be your problem LLMs don’t stand still: How to monitor and trust the models powering your AI Semantic Caching: What We Measured, Why It Matters The Annual SRE Survey Is Open—We Want to Hear from You Observability isn’t about the tool. It’s about the truth Invisible dependencies, visible impact: Lessons from the Google Cloud outage Real-time detection of BGP blackholing and prefix hijacks Leading analyst firm reveals the real cost of internet disruptions The Power of Over 3000 Intelligent Observability Agents Monitoring in the Age of Complexity: 5 Assumptions CIOs Need to Rethink Why Intelligent Traffic Steering is Critical for Performance and Cost Optimization Retail digital performance event recap: Key insights from IBM & Catchpoint Zendesk outage: A case for proactive monitoring and faster incident response Silence during chaos: Why the X outage is a call to arms for proactive monitoring The $1 Million Lesson: Building a Culture of Quality Through SLAs When AI tools fail: How to map your AI dependencies for proactive visibility Why Super Bowl 2025 was a triumph for Internet Resilience Why Internet Performance Monitoring is the new health check for IT organizations Why use Playwright in Catchpoint for synthetic monitoring Introducing WebPageTest Expert Plan: Real-Time Insights, Synthetic + RUM together in One Platform The shift to digital: How businesses are reshaping their priorities for 2025 The SRE Report 2025's Call to Action Monitoring in the Age of the Internet: DEM, IPM, and APM—What You Need to Know SSL Monitoring, Trust, and McLOVIN Performing for the holidays: Look beyond uptime for season sales success Lessons from Microsoft’s office 365 Outage: The Importance of third-party monitoring Web Performance Experts Look into the Future of Web Performance The hidden challenges of Internet Resilience: Key insights from 2024 report When SSL Issues aren’t just about SSL: A deep dive into the TIBCO Mashery outage The curious case of Marriott and the untold impact of web performance on revenue Preparing for the unexpected: Lessons from the AJIO and Jio Outage It’s time to stop neglecting the elephant in the room: Performance Matters! The Need for Speed: Highlights from IBM and Catchpoint’s Global DNS Performance Study Learnings from ServiceNow’s Proactive Response to a Network Breakdown Webinar Recap: Taking Web Performance to the Next Level Use the Catchpoint Terraform Provider in your CI/CD workflows Is the Internet ready for L4S? Takeaways from the CrowdStrike outage: third-parties can pose risk July 19th global IT outage reminds us of digital complexity 5 Actions you can take to improve digital performance 2024: A banner year for Internet Resilience APM vs Observability: Both-and, not either-or AppAssure: Ensuring the resilience of your Tier-1 applications just became easier APM vs observability: why your definitions are broken APM vs Observability: What comes next? APM vs Observability: Observing beyond APM Achieving stability with agility in your CI/CD pipeline AWS Outage: How do you prepare for the failure of your own safety net? Agentic AI: Powerful But Fragile—What You Need to Know Catch frustration before it costs you: New tools for a better user experience Catchpoint Expands Observability Network to Barcelona: A Growing Internet Hub Catchpoint Peak Performance Summit 2025: Redefining Observability for the Outcome Economy Catchpoint named a leader in the 2024 Gartner® Magic Quadrant™ for Digital Experience Monitoring Consolidation and Modernization in Enterprise Observability Connected Devices: Unlocking the next frontier of Internet Performance Monitoring Cloud Monitoring's Blind Spot: The User Perspective Cloudflare’s Resolver Outage: More Than Just DNS Cloudflare outage: another wake-up call for resilience planning Demystifying API Monitoring and Testing with IPM Creating the IPM Category: Catchpoint’s Journey to Leadership and the LogicMonitor Era Critical Requirements for Modern API Monitoring Customer Survey 2024: Unveiling insights and impact Diagnosing Wi-Fi failures that traditional tools miss: a case study DNS misconfiguration can happen to anyone - the question is how fast can you detect it? ECN explained: Navigate congestion for faster, smoother data delivery Don’t get caught in the dark: Lessons from a Lumen & AWS micro-outage Escalating risk, shrinking margins: The 2025 Internet Resilience Report From refresh to results: the metrics that shaped Election Day 2024 coverage Fast and furious: The importance of performance in the digital age Getting Started with Traceroute From the source to the edge: the six agent types you can’t ignore From SEO to AEO: Why Web Performance Is the Key to AI Search Success Going for gold: Testing the resilience of Olympic websites Here’s the proof: What the fastest sites on the web have in common Google’s Agent-to-Agent (A2A) Protocol is here—Now Let’s Make it Observable How IPM helped a top tech brand catch an OpenAI outage before it became a crisis How AI Turns Monitoring From “What Now?” Into “What’s Next?” How SAP achieved world-class uptime through modern observability How to Monitor AI Agents in Commerce Systems
Did Delta's slow web performance signal trouble before CrowdStrike?
2026-05-31 · via Catchpoint Blog

in this blog post

The CrowdStrike outage was a reminder of how quickly the dominoes can fall—especially when the foundation is shaky. Delta Airlines was hit harder than its competitors. While United and American Airlines were able to recover within days, Delta faced ongoing struggles, leading to the cancellation of 7,000 flights over five days.

CEO Ed Bastian estimated this cost the company $500 million. And unlike United, which canceled a smaller number of flights, Delta’s difficulties lingered. It comes as no surprise that the fallout was a 26% drop in third-quarter earnings. Delta earned $971 million, down from $1.31 billion a year earlier, and is now seeking compensation for the July outage. These numbers paint a stark picture of the financial toll taken by Delta's slow recovery, and it’s hard not to ask: Why was Delta so vulnerable?

According to Microsoft, the answer may lie in their infrastructure. In an ongoing war of words, Microsoft suggested Delta’s IT simply wasn't up to par. "Unlike its competitors, Delta apparently has not modernized its IT infrastructure," Mark Cheffo, an attorney for Microsoft, pointed out. It raises the question: Did Delta's struggles really begin with CrowdStrike, or were the seeds sown well before?  

Maybe the fundamental problem is that Delta has under-invested in IT. I have yet to meet a CIO who has too many resources and no backlog of systems to modernize, legacy tools to retire (SiteScope, anyone?), or security enhancements to implement.  

This post is not trying to blame IT or even point the finger at Delta. We are looking at them as an example of a large, complex organization that may need to increase its investment in IT overall to be better prepared for the future. This connection became evident when we revisited our Airline Website Performance Benchmark Report. Published before CrowdStrike’s devasting impact on Delta, the report highlighted significant deficiencies relative to industry benchmarks – and customer expectations.  

Lagging website performance: A window into deeper issues

Catchpoint has been conducting web performance benchmarking reports for many years, often as exclusive reports for our customers or other internal purposes. Commenting on the purpose, the report states, “Our aim isn’t to pick winners and losers or point the finger at brands with slower sites. Instead, it’s to learn about the range of customer experiences being delivered, identify areas for improvement, and spark a conversation about best practices.”

Our benchmarking report, published July 2024, revealed that Delta's website lagged behind its competitors in categories such as page load time, where Delta was the slowest among the airlines analyzed. For context, a recommended page load time is less than 5 seconds. Delta's was over 7 seconds.

A graph of a number of passengersDescription automatically generated with medium confidence

You may be asking: How is that relevant? How damaging is slow page load time?  

Page load time matters more than you think

In its simplest terms, page load time is the average amount of time it takes for a page to show up on your screen. It measures the time from initial navigation to receiving the last byte of the final element on the page. More than just a single metric, it encompasses factors like the number of requests, the amount of data downloaded, DNS quality, and third-party requests. Essentially, it's a crucial metric and an effective indicator of how well your website is performing.

Slow load times aren't just technical problems—they’re business problems. It’s well established that the longer people stare at a blank screen, the less likely they are to spend money on a website. Research shows that 40% of users abandon a site that takes over three seconds to load—and that goes up to 53% if they're browsing on a mobile device. As page load time goes from one to ten seconds, the probability of a mobile user bouncing increases by 123%. Perhaps, the most frightening statistic of all is that a one-second delay in page load time can lead to a 7% reduction in conversions. Now consider that in our tests, Delta's page load time was over 7 seconds.  

Document complete time: The hidden cost of delay

A graph of a document completeDescription automatically generated

Delta also struggled with document complete time—the point when a browser has fully loaded all the necessary elements for a page to be functional. Delta's score of 5133ms was well above the 3-second benchmark. This means customers were left waiting, often staring at a partially loaded page while trying to book their tickets or make changes to their reservations.

In an industry where every minute matters, these delays are more than just inconvenient—they translate directly into lost bookings and lost revenue.

Building a culture of excellence: Why every detail matters

There's a school of thought that building a culture of excellence means striving for greatness in every corner of the business—not just the obvious parts like getting planes from A to B, but also in areas that might not seem directly connected to flying. It's about establishing high standards and continuously improving—before outages like CrowdStrike's happen. Things like web performance might feel peripheral, but in today’s digital world, they absolutely matter.

So, at first glance, Delta’s poor digtial customer experience and page load times may seem disconnected from their primary business of flying passengers. But having poor web performance metrics isn’t just about website speed; it’s a red flag for the entire operation. It signals a gap in the commitment to excellence, which always becomes apparent under stress. When CrowdStrike happened, it stress-tested Delta’s resilience as an organization, and Delta struggled.

Was Delta's vulnerability inevitable?

The CrowdStrike outage may have been the spark, but the dry tinder—lack of investment in IT systems from cybersecurity to web performance—was already in place. In times of crisis, organizations that have built Internet Resilience into every part of their business are the ones that come out on top. Our Inaugural Internet Resilience Report, based on insights from 300+ digital leaders in North America and EMEA, found that overwhelmingly, resilience was deemed to be a moderate or high driver for every part of business operations.

A graph of different colored rectanglesDescription automatically generated

Source

Internet Resilience matters across the entire business, from improving operational efficiency and maximizing workforce productivity to mitigating business risk.

For those in the airline industry—and beyond—every industry now relies on the Internet as the primary enterprise network. Delta’s recent struggles highlight why a proactive approach to web performance is essential. Internet Performance Monitoring is crucial. In a world where slow is the new down, companies must invest in Internet Resilience now or risk losing out when the next outage inevitably occurs.

For a detailed look at the web performance of major airlines and what they can do to improve, read the full benchmark report.

The CrowdStrike outage was a reminder of how quickly the dominoes can fall—especially when the foundation is shaky. Delta Airlines was hit harder than its competitors. While United and American Airlines were able to recover within days, Delta faced ongoing struggles, leading to the cancellation of 7,000 flights over five days.

CEO Ed Bastian estimated this cost the company $500 million. And unlike United, which canceled a smaller number of flights, Delta’s difficulties lingered. It comes as no surprise that the fallout was a 26% drop in third-quarter earnings. Delta earned $971 million, down from $1.31 billion a year earlier, and is now seeking compensation for the July outage. These numbers paint a stark picture of the financial toll taken by Delta's slow recovery, and it’s hard not to ask: Why was Delta so vulnerable?

According to Microsoft, the answer may lie in their infrastructure. In an ongoing war of words, Microsoft suggested Delta’s IT simply wasn't up to par. "Unlike its competitors, Delta apparently has not modernized its IT infrastructure," Mark Cheffo, an attorney for Microsoft, pointed out. It raises the question: Did Delta's struggles really begin with CrowdStrike, or were the seeds sown well before?  

Maybe the fundamental problem is that Delta has under-invested in IT. I have yet to meet a CIO who has too many resources and no backlog of systems to modernize, legacy tools to retire (SiteScope, anyone?), or security enhancements to implement.  

This post is not trying to blame IT or even point the finger at Delta. We are looking at them as an example of a large, complex organization that may need to increase its investment in IT overall to be better prepared for the future. This connection became evident when we revisited our Airline Website Performance Benchmark Report. Published before CrowdStrike’s devasting impact on Delta, the report highlighted significant deficiencies relative to industry benchmarks – and customer expectations.  

Lagging website performance: A window into deeper issues

Catchpoint has been conducting web performance benchmarking reports for many years, often as exclusive reports for our customers or other internal purposes. Commenting on the purpose, the report states, “Our aim isn’t to pick winners and losers or point the finger at brands with slower sites. Instead, it’s to learn about the range of customer experiences being delivered, identify areas for improvement, and spark a conversation about best practices.”

Our benchmarking report, published July 2024, revealed that Delta's website lagged behind its competitors in categories such as page load time, where Delta was the slowest among the airlines analyzed. For context, a recommended page load time is less than 5 seconds. Delta's was over 7 seconds.

A graph of a number of passengersDescription automatically generated with medium confidence

You may be asking: How is that relevant? How damaging is slow page load time?  

Page load time matters more than you think

In its simplest terms, page load time is the average amount of time it takes for a page to show up on your screen. It measures the time from initial navigation to receiving the last byte of the final element on the page. More than just a single metric, it encompasses factors like the number of requests, the amount of data downloaded, DNS quality, and third-party requests. Essentially, it's a crucial metric and an effective indicator of how well your website is performing.

Slow load times aren't just technical problems—they’re business problems. It’s well established that the longer people stare at a blank screen, the less likely they are to spend money on a website. Research shows that 40% of users abandon a site that takes over three seconds to load—and that goes up to 53% if they're browsing on a mobile device. As page load time goes from one to ten seconds, the probability of a mobile user bouncing increases by 123%. Perhaps, the most frightening statistic of all is that a one-second delay in page load time can lead to a 7% reduction in conversions. Now consider that in our tests, Delta's page load time was over 7 seconds.  

Document complete time: The hidden cost of delay

A graph of a document completeDescription automatically generated

Delta also struggled with document complete time—the point when a browser has fully loaded all the necessary elements for a page to be functional. Delta's score of 5133ms was well above the 3-second benchmark. This means customers were left waiting, often staring at a partially loaded page while trying to book their tickets or make changes to their reservations.

In an industry where every minute matters, these delays are more than just inconvenient—they translate directly into lost bookings and lost revenue.

Building a culture of excellence: Why every detail matters

There's a school of thought that building a culture of excellence means striving for greatness in every corner of the business—not just the obvious parts like getting planes from A to B, but also in areas that might not seem directly connected to flying. It's about establishing high standards and continuously improving—before outages like CrowdStrike's happen. Things like web performance might feel peripheral, but in today’s digital world, they absolutely matter.

So, at first glance, Delta’s poor digtial customer experience and page load times may seem disconnected from their primary business of flying passengers. But having poor web performance metrics isn’t just about website speed; it’s a red flag for the entire operation. It signals a gap in the commitment to excellence, which always becomes apparent under stress. When CrowdStrike happened, it stress-tested Delta’s resilience as an organization, and Delta struggled.

Was Delta's vulnerability inevitable?

The CrowdStrike outage may have been the spark, but the dry tinder—lack of investment in IT systems from cybersecurity to web performance—was already in place. In times of crisis, organizations that have built Internet Resilience into every part of their business are the ones that come out on top. Our Inaugural Internet Resilience Report, based on insights from 300+ digital leaders in North America and EMEA, found that overwhelmingly, resilience was deemed to be a moderate or high driver for every part of business operations.

A graph of different colored rectanglesDescription automatically generated

Source

Internet Resilience matters across the entire business, from improving operational efficiency and maximizing workforce productivity to mitigating business risk.

For those in the airline industry—and beyond—every industry now relies on the Internet as the primary enterprise network. Delta’s recent struggles highlight why a proactive approach to web performance is essential. Internet Performance Monitoring is crucial. In a world where slow is the new down, companies must invest in Internet Resilience now or risk losing out when the next outage inevitably occurs.

For a detailed look at the web performance of major airlines and what they can do to improve, read the full benchmark report.