Akamai Diversity

The Akamai Blog

Mobile Web Performance: 23 stats you should know about mobile web performance

I love a good roundup of stats, and the webcast I did yesterday -- a "state of the union" for mobile web performance optimization, which you can watch here -- offered up some good roundup fodder on everything from user expectations to common performance issues.

There are, of course, many more than twenty-three interesting stats about mobile usage and performance (it was only a one-hour talk, after all), but if you're interested in mobile, this is a good start.

We have a love/hate relationship with mobile

1. 7% of US adults rely on their smartphone. This group of mobile users -- which includes a significant number of people with lower incomes (more on that later in this post) -- has little to no online access other than their phone. This group also has no broadband service at home other than their phone's data plan. [source]

2. Almost half of US adults "can't live without" their smartphone. When surveyed, 46% of adults said that their phone is vital to their daily lives. [source]

mobile web performance stats

3. 39% of mobile users are unhappy with their online experiences. The top two reasons: slow pages and site freezes/crashes. [source]

4. 57% have experienced problems when using mobile. See above. [source]

5. 3 out of 4 consider mobile slower than desktop. This number surprised me. I expected it to be much higher. [source]

6. 46% will not return to a site that performs poorly. [source]

7. Only 9% of online shoppers prefer to shop via mobile. This is another number that surprised me. I would've thought even fewer. Perhaps the convenience of shopping at your fingertips mitigates the performance issues? Just a guess. [source]

Mobile matters

8. By 2020, 4 out of 5 people worldwide will own a smartphone. That's a total of roughly 6.1 billion mobile users. Many of these users will never have used a desktop computer or landline. [source]

mobile web performance

9. Mobile shoppers are more engaged than desktop shoppers. 57% of tablet users conduct product searches at least once a week, compared to 37% of desktop users. [source]

10. Mobile users shop more. 35% of tablet and smartphone users make at least one online purchase per week via mobile, compared to just 15% of desktop users. [source]

11. Mobile users spend more. Looking at annual spending per device, consumers spend more than 50% more via their tablets than they do via their desktops. These findings could be interpreted to mean that tablet owners tend to be more affluent, and therefore bigger spenders. If this is, in fact, the case, then it's a compelling reason for site owners to invest in optimizing for that audience. [source]

mobile web performance stats

12. When it comes to mcommerce, the US is not a global benchmark. In the US, currently only about 15% of online transactions take place via mobile devices. In Japan, South Korea, and the UK, that number is approaching -- or in the case of Japan, exceeding -- 50%. Not to worry, though. The US will get there eventually. [source]

13. Mcommerce conversion rates are on an upward trajectory. Again, we can look to Japan for an idea of where this trajectory is headed for North American shoppers. Currently, mobile shopping in the US has an average conversion rate of almost 2.5%. In Japan, the average mobile conversion rate is more than 9%. And again, the US will catch up here. [source]

How fast do mobile users expect pages to be?

14. I wish I could say there's a single answer to this question, but there isn't. Every site is different, which is why every site owner should ideally be looking to their own data to understand how their users interact with their site.

But in the interest of addressing this question, I looked at one of SOASTA's customers, a leading online retailer who uses mPulse to do real user monitoring for both their desktop traffic and mobile traffic. The graph below depicts one month of data for all mobile sessions. You can see that the site's conversion rate peaks at 1.9% for visitors who experienced average page loads of 2.4 seconds, and the conversion rate drops to 1.5% for visitors who experience average page load times of 3.3 seconds. In other words, for this site, conversions were 27% higher for visitors who enjoyed a load time that was about one second faster. (I'll be writing about this in deeper detail in a future post.)

mobile web performance stats

What's wrong with mobile website performance?

15. In July, the average page served to mobile was 1180 KB in size. Compare this to the average page served to mobile in 2011, which weighed in at 390 KB. In other words, the average page is three times bigger today than it was four years ago. And bear in mind that more bandwidth isn't a performance cure-all. [source]

website monitoring for mobile devices

16. Images account for 63% of total page weight. The vast majority of images on the web are either in the wrong format or they're the wrong size or they're uncompressed -- or some combination of all three of these factors.  [source]

17. 1 out of 5 pages contains 100 or more resource requests. In 2011, almost half of all pages served to mobile contained fewer than 25 total requests. Today, more than half of all pages contain 50 or more requests, with 20% of pages containing 100 or more. Latency for major mobile carriers in the US ranges from 340 to 362 milliseconds per request. Those milliseconds add up fast. As you may or may not know, latency is a serious performance-killer. [source]

18. JavaScript usage continues to rise. 60% of pages served to mobile contain 10 or more JS script requests, and each of these requests increases the complexity of page rendering. Increased complexity = slower mobile site load times. [source]

19. Almost half of pages served to mobile use custom fonts. Back in 2011, only 2% of all pages served to mobile devices used custom fonts. Today, just four years later, that number has leapt to 48%. [source]

20. 1 out of 5 pages contains 5 or more redirects. These numbers surprised me. I expected pages to contain a few redirects, but I didn't expect 20% of pages to contain five or more. And 3% of pages contained 10 or more. That's a lot. [source]

21. 16% of pages contain HTTPS requests. In 2011, only 3% of pages served to mobile contained HTTPS requests. Today, that number has climbed to 16%. While this is a fantastic step forward in terms of offering a more secure user experience, it also comes with performance considerations that site owners need to be aware of. HTTPS requests require a few extra handshakes in order to verify their authenticity, which incurs an extra bit of latency. This latency can affect load times, though it doesn't have to. [source]

Page bloat has a real cost for mobile users

22. 27% of smartphone users exceed their data cap every month. [source]

mobile web performance

23. Low- and middle-earners are more likely to max out their data. It may not be a huge issue for you if you blow past your data cap, because you can easily absorb an extra $10-20 on your bill. For people with lower incomes -- and bear in mind that these are people who are more likely to rely on their smartphones and 3G/4G networks to access the internet for necessary services -- that extra $20-30 is a big deal. This is why it's concerning to know that 28% of low-income earners (less than $30K/year) exceed their data cap every month. And this is yet another reason why managing page bloat should be a priority for site owners.

To recap...

More and more, mobile has become a fundamental tool that we use in almost every aspect of our personal and professional lives. Yet most mobile users are dissatisfied with the quality of their online experiences, citing slow performance and outages as their main complaints. The culprits: web pages that are bigger and more complex than ever. These slowdowns don't just hurt your visitors, they hurt your business -- but the good news is that this problem is fixable.

Related posts: