Mobile

22 not-so-short links about web performance

After January’s roundup of performance links was so well-received, I thought I’d make this a regular post. February was a busy month. Here are the best articles, studies, and blog posts I’ve read in the past four weeks.

Web performance overviews, standards and stats

Online performance is business performance
Some excellent data from TRAC Research about the impact of slow websites. Key findings:

  • 4.4 seconds is the average delay in website response times when business performance begins to decline.
  • $21,000 is the average revenue loss for one hour of website downtime.
  • $4,100 is the average revenue loss of an hour of website slowdowns.
  • Website slowdowns occur 10 times more frequently than website outages.
  • Website slowdowns can have twice the revenue impact on an organization as an outage.

I’ll definitely be adding these numbers to my performance cheat sheet.

W3C Launches Web Performance Interest Group
On top of producing use cases and requirements for future deliverables of the Web Performance Working Group, the Interest Group will “provide a forum to discuss web performance initiatives across web publishers, vendors, developers, and users with the goal of identifying areas for standardization.”

Four elements of web performance: Weight, time, processing and perception
This is a really good breakdown — including some useful tips — from Marcus Westin.

Web Performance 101
This is an excellent overview of web performance, its challenges, and its solutions, from Stephen Thair. I’m indebted to it for turning me on to the fascinating “web stress” study I wrote about last week.

The Slow-Motion Internet
A well-crafted article on performance in general, and on Google’s “Make the Web Faster” initiative in specific. It’s from MIT’s Technology Review magazine, so I had to sign up — and use up one of my three feature article credits — to read it, but it was worthwhile. (I was also quite flattered to see data from Strangeloop cited as a source for one of Google’s infographics.)

Case studies and real-world scenarios

The Daily Wait
John Gruber talks about the incredibly slow load times he’s experienced when accessing The Daily and The New York Times on his iPad, and the impact this has had on his use of these apps:

“For comparison’s sake, I timed The New York Times iPad app. That took about 25 seconds to load today’s issue. A lot less time than The Daily, but, still too long. I realized that the delay before being able to read it was the reason I’d slowly stopped using The NYT iPad app over the last few months.”

A Race Between Digital and Print Magazines
In a similar vein (and, ironically, in The New York Times), is this piece from tech columnist Nick Bilton. Bilton conducted a test in which he started to download Wired on his iPad, then timed it against how long it took him to drive twelve blocks to buy the print version. Guess which came first.

How We Got DynDNS.com to Load Faster and How You Can Learn from It
I love pieces like this one from Dyn’s blog. They talk us through the step-by-step performance optimizations they made, and share the awesome end result: pages that load in less than half the time.

Flower Sites Hit Hard by Valentine’s Day
Pingdom, an uptime monitoring service, shares some telling graphs that show how major flower websites — Flower.com, Justflowers.com and Sendflowers.com — all experienced brutal slowdowns on one of the most important days of their year.

Balancing front-end reporting and tracking
A sign that web performance has truly arrived as a mainstream issue: The Guardian writes about its own performance-tuning efforts in its dev blog.

How-to…

Disabling ETags in IIS6
Duncan McDougall explains how to address a common YSlow and WebPagetest suggestion: turning off entity tags.

Creating Fast Buttons for Mobile Web Applications
Google offers some tips and code for making your mobile buttons faster.

Using Human Computation to answer web performance questions
Sergey Chernyshev discusses how to identify the “enough” point, when a user is able to engage with a web page.

Optimize the Performance of Widgets, Buttons and More
Practical tips for fixing performance of common widgets and buttons from Digg, Twitter, Facebook, StumbleUpon, Google, and more.

Mobile

Retailers create better designs on their web sites and beyond
Study: 44% of online retailers have mobile commerce sites, up from 13% a year ago.

Study: Apple’s Mobile Browser Is Fastest
According to a recent Gomez study, Apple’s mobile version of Safari was fastest and BlackBerry’s browser the slowest. Scroll down to see a complete chart of Gomez’s browser data, including its data for desktop browsers.

Maximize Your Mcommerce Strategy
More research from Gomez, this time saying that mobile consumers expect pages to load in 6 seconds, compared to less than 2 seconds for desktop users. (Though Jeroen Tjepkema pointed out to me that, later in this same slideshow, it states that 60% of all mobile users expect sites to load as fast as on their PC.)

Browsers

Focusing on Real World Web Performance with Internet Explorer 9
Good behind-the-scenes post from the folks behind IE9, explaining their five performance objectives.

The Era of Browser Pre-Connect
Mike Belshe makes some good observations in this post, including this one: “IE9 is the first browser I know of which appears to preconnect right out of the gate.”

What you don’t know about browsers can hurt your conversion rate
While many developers have written off IE6, this article about L.L. Bean’s website highlights an important point:

While most consumers now use more advanced browsers than IE6, IE6 is still used by 4-5% of the slightly older than average shoppers at LLBean.com. In China, 60% of online shoppers use IE6. Each retailer must consider its own customers and the browsers they use when deciding how important each browser is.

Connectivity

National Broadband Map shows how connected your community is
A good reality check for those of us in urban centres, developing for people outside of cities.

Home Internet May Get Even Faster in South Korea
The average home in South Korea already has a faster connection than I do, and according to this article, it’s about to get much faster:

By the end of 2012, South Korea intends to connect every home in the country to the Internet at one gigabit per second. That would be a tenfold increase from the already blazing national standard and more than 200 times as fast as the average household setup in the United States.

Did I miss anything? Let me know in the comments.

Related posts:

Review: Blaze Mobile performance measurement tool

Mobile website performance measurement toolOur industry has been in desperate need of a solid tool that measures real-world mobile website performance. Blaze Mobile — which is based on the Webpagetest framework — may be just what we’ve been waiting for.

The tool is still in beta, and the folks at Blaze has a list of known issues they’re working on. It’s up to the rest of us to take it for a test drive, see how it performs, and offer our feedback.

I thought it would be interesting to start by testing the current top 5 sites in Keynote’s mobile commerce index and benchmark Blaze’s results alongside Keynote’s.

Test parameters

Blaze lets you test for iPhone and Android, so I tested on both those platforms.

You can choose between one, two, and three test runs per URL. I opted for three runs. The averages are below.

The tool also lets you capture video results, so I tested with and without that option. As discussed here, capturing video can have a negative impact on Webpagetest’s page load times. Blaze has stated that their tool has the same issue, so I wanted to see how pronounced the impact of video capture is.

Comparison: Keynote and Android

Website Keynote Blaze: Android
(with video)
Blaze: Android
(no video)
Strand.com 2.98s 2.3s 2.14s
Barnesandnoble.com 4.53s 1.06s 1.12s
Walmart.com 4.41s 2.27s 2.82s
Dell.com 4.35s 1.66s 1.5s
Victoriassecret.com 5.94s 2.4s 3.45s

Comparison: Keynote and iPhone

Website Keynote Blaze: iPhone
(with video)
Blaze: iPhone
(no video)
Strand.com 2.98s 3.83s 2.67s
Barnesandnoble.com 4.53s 3.43s 1.73s
Walmart.com 4.41s 6.29s 3.73s
Dell.com 4.35s 3.72s 4.63
Victoriassecret.com 5.94s 5.5s 5.05s

What patterns emerged?

Even allowing for fluctuations caused at the network and delivery end of things, there are some interesting patterns:

  • Across the board, Android load times were significantly faster than Keynote and iPhone load times. In the case of the Barnes & Noble site, the Android load time was about 400% faster.
  • Keynote’s load times surprised me by being, overall, slower than Blaze’s. I had expected that, because Keynote takes their measurements at the highest possible network speeds available at the time of the tests, their results would be faster.
  • Video capture dramatically slowed down iPhone results, in some cases making them almost twice as slow as load times tested without video capture. The effect of video capture on Android results was much slighter.

My review

Usability

I found Blaze Mobile extremely usable. The UI is really clear, and considering the fact that it can only run one test at a time, I was impressed with how fast it was able to deliver results (though that could change if it catches on).

Results page

The test results pages are easy to read and include a waterfall and HAR file. Right now, the results page only shows you the load time and page size.

Suggestions for future development:

  • It would be great to also see time to first byte and start render. Sure, you can find this on the waterfall, but waterfall interpretation isn’t in everyone’s skill set. (If you want to add it to yours, here’s a beginner’s guide to waterfall charts.)
  • I also like Webpagetest’s option to view your test results as a filmstrip. This would be a good feature to see here as well.
  • It would be good to be able to run side-by-side tests.
  • I’d also like to be able to export the video file.

Methodology

I like Blaze’s transparent methodology. Keynote provides a bit of this in a footnote on their mobile index page, but I’ve always hankered for more background from them. I’d love to know what devices and operating systems they’re testing with. With Blaze, I get a better sense, not just of how they run their tests, but also the inevitable caveats that come with interpreting the results. As a result, I feel like I can buy in to their results. They feel truthy.

My only major caveat

Until the video capture issue is resolved for the iPhone, I’d consider these results invalid. So make sure you de-select ‘Enable Video Capture’ when you run iPhone tests.

Other than that issue, I’m adding Blaze Mobile to my toolset, and looking forward to where they take it next.

Related posts:

2011 Web performance predictions for the mobile industry

I recently contributed an article to RCR Wireless that highlights my four big predictions for the mobile industry:

  1. Companies will generate at least 15% of Web sales via their social presence and mobile applications.
  2. Android will become the No. 1 mobile platform, surpassing the iPhone in terms of units and usage.
  3. Retailers will realize that mobile shoppers have a goal-driven “hunter” mentality.
  4. As a result of No. 3, mobile Web performance will become as important as desktop Web performance.

Read the full article here.

Related posts:

Why mobile websites are still disappointing consumers

Last week, I wrote a guest post for Marketing Daily on the fact that, no matter how much press is being given to the success of mobile shopping this holiday season, mobile sites are still not meeting consumers’ expectations. An excerpt:

Fifty-eight percent of mobile users expect sites to load at least as quickly on their mobile devices as on their desktops. You would think this expectation would lead to increasingly faster m-commerce sites, but the opposite seems to be the case.

According to industry benchmarks from Gomez and Keynote, mobile sites seem to be getting slower, not faster. Currently, the average m-commerce site loads in 5.47 seconds. A year ago, that number was 4.73 seconds.

There are a few potential culprits here, such as oversized graphics and poorly optimized widgets, but pointing fingers doesn’t fix the problem. Site visitors want a fast online experience, not excuses.

The rest of the post is here: All I Want For Christmas Is A Faster Mobile Experience

Related posts:

UPDATE: Everything you wanted to know about web performance but were afraid to ask

I’m cleaning out my old bookmarks and rediscovered a Forrester report from earlier this year, The Impact of Poor Web Site Performance in Financial Services. I’ve added the following key bits of data to my performance stats cheat sheet:

  • No activity on the web approaches the frequency of online account access.
  • Web site performance is second only to security in user expectations. Web site performance ranks above even functions like single sign-on or a website that is easy to use.
  • 56% of online bankers and brokers expect web pages to load in 2 seconds or less; this is far above the 47% of consumers who are just shopping online.
  • Poor website performance leads to dissatisfaction more often than any other factor. Sixty-four percent of online US bankers and brokers have had a dissatisfying experience when servicing their accounts. Web performance is far and away the biggest reason for this dissatisfaction.
  • As online tasks get more urgent or complex, online users are less likely to try later and more likely to move to more expensive channels to complete the transactions. Fifty-six percent of online bankers would move to offline channels to ask a general account question; 54% of online brokers would move to offline channels to trade investments if the website was unavailable or slow to respond.
  • The ultimate effect of poor performance is a decrease in willingness to recommend a firm, with 48% of online bankers and brokers saying that poor performance had an impact or significant impact on their likeliness to recommend a firm’s services to a friend or family member.

Also, in the time since I started creating this post, Gomez released a new commissioned report called When Seconds Count. It says that:

  • Nearly one-third (32 percent) of consumers will start abandoning slow sites between one and five seconds.
  • 39% say speed is more important than functionality for most websites, while only one in five rank greater site functionality as more important.
  • More than half of mobile users expect websites to load as quickly, almost as quickly or faster on their mobile phone, compared to the computer they use at home.
  • More than a third of mobile users (37%) said they would not return to a slow site, and 27 percent would likely jump to a competitor’s site.

Related posts: