10 Real Live Examples showing why Performance and Availability Matter

For more than 10 years now, I’m giving talks and doing presentation about IT performance monitoring, Application Performance Management, observability and performance engineering topics. When presenting to a rather broad audience on these topics, it is essential to adequately motivate the importance and relevance of performance and availability. Otherwise, the audience will have lost their thoughts in a few minutes – such as: “What’s for dinner today?”, “Man, this is a nice room here!”, etc. You quickly get the attention of the audience when you talk about money, costs and business success. This is tangible and most people can easily identify with it in some way.

And in fact:

Performance & Availability = Cash Money = Business Success

There are plenty of real live examples that impressively demonstrate this correlation. In the following I will summarize 10 of such prominent examples.

Examples, demonstrating why you should care about Performance & Availability

GitHub Outage lasting more than 6 hours – 2020

“GitHub Outage impacts millions of developers.”

How about some simple math? Let’s prudently assume that around 1 million developers were prevented from doing their work for more than 6 hours due to GitHub’s outage (in fact GitHub has more than 40 million users). With an average developer salary of $50 per hour, in our example the outage would have yielded more than 300 millions of opportunity costs – money lost worldwide due to lower productivity.

Heathrow Disruption – 2020

“More than 100 flights to and from London’s Heathrow airport were disrupted after it was hit by technical issues affecting departure boards and check-in systems”

Taking into account that flight delays cost airlines annually billions of dollars, you may imagine the financial damage caused by such a disruption.

Facebook Outage – 2019

“The outage, which affected millions of users across Facebook, Instagram, Messenger, WhatsApp, and Oculus, lasted for approximately 14 hours […] tens of millions of dollars were lost.”

The root cause for the largest outage since 2012 (according to Tom Sanders, co-founder of DownDetector) has been a server configuration change that triggered a cascading series of issues, affected millions of users and causing tens of millions of dollars worth of damage.

Amazon Prime Day Fail – 2018

“Amazon’s one hour of downtime on Prime Day may have cost it $72 to $100 million in lost sales.”

These huge cost values come from a simple calculation. Taking Digital Commerce 360‘s projection that Amazon would generate 2.5$ billion in sales during 36-hour sale results in ~$70 million loss in revenue with a downtime of approximately one hour.

Google findings on loading times and bounce rates – 2018

“The probability of bounce increases 32% as page load time goes from 1 second to 3 seconds.”

With 70 percent of mobile landing pages taking more than five seconds to load, the corresponding companies lose half of their users and thus half of their revenue already on the landing page.

Akamai Online Retail Performance Report – 2017

“A 100-millisecond delay in website load time can hurt conversion rates by 7 percent.”

7 percent lower conversion rates means 7 percent lower revenue. Let’s take an average e-commerce site with an annual revenue of ~ $4 million. Such an enterprise would loose 280,000$ every year – just because of 100ms delay in website load time.

BBC handling tens of millions of users through compromises – 2016

“At the BBC we’ve noticed that, for every additional second a page takes to load, 10 per cent of users leave.”

BBC wants to impress with the quality of its services. Therefore BBC switches off certain features if it notices that the performance is getting worse in order to be able to reach a high speed again.

Pinterest driving user growth with performance improvements – 2015

“Improving landing page performance by 60 percent leads to 40 percent higher signup conversion rate.”

The equation is trivial: 40 percent more users means up to 40 percent more revenue – purely through performance improvement.

Sales Increase for Internet Retailer AutoAnything.com – 2010

“Faster page loading contributed to a 12-13% increase in sales.”

With an improvement of 50 percent in page load times AutoAnything.com (an Internet car parts retailer) achieved a sales increase up to 13 percent. With AutoAnything.com’s annual revenue of more than 50 millions of dollars, this yields an annual increase of more than $6.5 million.

Kissmetrics – Study on loading time – 2011

“40% of people abandon a website that takes more than 3 seconds to load.”

“If an e-commerce site is making $100,000 per day, a 1 second page delay could potentially cost you $2.5 million in lost sales every year.”

…

This study lists even more such statements – all with the same message: You should care about performance!

Conclusion

I hope you are convinced that performance and availability of IT systems are not just mock factors or annoying software quality characteristics, but crucial success factors and cash money. There are definitely more prominent examples. Can you think of some? I would love to see your thoughts on this in the comment section below!