Time is Money
During a twitter conversation about web performance with some other developers a while back I was introduced to Tammy Everts and her book, Time is Money; when it was offered as a free PDF by Everts’ employer, I grabbed it. I’m so glad that I did, it is jam packed with case studies and data about why this topic is important.
I’ve lately been learning a lot more about performance for my work, helping on some client projects, but also getting familiar with the biggest hurdles and how to make a site more performant. Everts takes all the things that I know about performance and gives me data to help back that up. In essence this is the book you want the bosses to read, so they understand why taking the time to do this work is worth it.
I recommend this book for anyone interested in performance though, as it collates together so much fantastic data, it’s worth having on your shelf as a resource.
I read a PDF and below are my highlights:
42% of men and 35% of women have decided not to use a company again as a result of experiencing a slow website. (p 2)
It’s tempting to label ourselves picky and impatient, but we’re not. There’s a wealth of research into what happens to us at a neurological level when we’re forced to deal with slow or interrupted processes. It isn’t pretty. (p 3)
The Internet may change, and web pages may grow and evolve, but user expectations are constant. The numbers about human perception and response times have been consistent for more than 45 years. These numbers are hard-wired. We have zero control over them. They’re consistent regardless of the type of device, application, or connection we’re using at any given moment. (p 6)
So the goal in getting page load times down to 100 milliseconds is to keep information from falling through the cracks in our iconic memory, while also giving our short-term and working memory ample time to do all the parsing they need to do before they start losing information. (p 8)
It’s also important to remind ourselves that application performance is just one part of the greater world. Our everyday lives are filled with events—from sitting in traffc to standing in line at the grocery store—that challenge our need for flow. Poor web performance is just one problem, but for those of us who spend much of our work and personal time online, it’s extra friction in an already friction-filled world. Its effects are cumulative, as most of us aren’t capable of compartmentalizing our stress. (p 10)
A great deal of attention is placed on retail performance because retail metrics are easy to capture: it’s relatively short work to draw a line between load times and revenue. But performance affects other verticals as well. Media, travel, finance… if there are real people using your service online, then those people’s behavior is susceptible to changes wrought by faster or slower web pages. I have yet to encounter a business that, after gathering enough user data and identifying the right metrics, didn’t find a correlation between performance and their business. (p 17)
Revenue isn’t the only metric that is affected quite differently by outages versus slowdowns. In one of the only studies (if not, the only study) into the impact of outages versus slowdowns on abandonment rates, Akamai found that sites that went down experienced, on average, a permanent abandonment rate of 9%. Sites that suffered from slow performance experienced a 28% permanent abandonment rate—an increase of more than 200%. (p 19)
But while Google’s treatment of performance as a search ranking factor has attracted a fair bit of tech media attention and discussion over the past several years, I’ve never met a site owner who suddenly decided to prioritize performance because of it. This could be due to the fact that the details have always been murky, with most people agreeing that speed is probably just one relatively minor part of Google’s search ranking algorithm. The “Slow” tag, however, may be too in your face to dismiss. It signals that Google is taking performance—and particularly mobile performance—increasingly seriously. Site owners would be wise to do the same. (p 42)
Not only do small delays add up quickly, but users—both internal and external—must concentrate up to 50% harder when pages or applications are slow. And while most of us are fairly good at developing short-term strategies that let us deal with repeated interruptions, this often comes at the expense of our overall satisfaction and our willingness to take on new tasks. (p 51)
Or put it this way: it doesn’t matter if you’re a hard-hearted Scrooge McDuck type who cares only about cutting infrastructure and bandwidth costs. If you tackle performance with those objectives in mind, you’re inevitably going to make your users measurably happier. (Sorry, Uncle Scrooge!) (p 59)
In 2015, the Federal Communications Commission updated its definition of broadband from 4 Mbps to 25 Mbps. According to this new definition, roughly one out of five Internet users—approximately 50 million people—in the United States suddenly did not have broadband access. (p 66)
Frontend optimization addresses performance at the browser level, and has emerged in recent years as an extremely effective way to supplement server build-out and CDN services. One way that FEO alleviates latency is by consolidating page objects into bundles. Fewer bundles means fewer trips to the server, so the total latency hit is greatly reduced. FEO also leverages the browser cache and allows it to do a better job of storing files and serving them again where relevant, so that the browser doesn’t have to make repeat calls to the server. (p 72)
Despite these constraints, user expectations continue to grow: a typical mobile user expects a site to load as fast—or faster!—on their tablet or smart-phone as it does on desktop. (p 74)
Wirth’s law, which states that software is getting slower more rapidly than hardware is getting faster. (p 81)
Not all web pages are created equal. When pages get slower, conversion rates suffer, but some types of pages suffer more than others. (p 86)