The holy grail of Web performance seems to be load times of two seconds or less. At the same time I hear people saying that the specific purpose of their Web site requires more and heavier content and consequently leads to higher load times. “I am running a media site, I need more images and videos” or “all our content is highly dynamic that’s why our pages load slower” are phrases typically used to justify higher load times.
If you compare a media site to Google’s search page it becomes obvious that there is a significant difference in the content that is shown.
At the same we know that people get bored or lose their attention when activities take too long. The easiest way to figure out how performance relates to the specifics of an industry site is to test the top sites and look at their performance metrics.
I used speedoftheweb.org to do this and decided to look at three industries which typically have a strong Web presence; retail, travel and media.
First I looked at First Impression Time. Interestingly the median across all industries was about 1.5 seconds.
However when I then looked at the Total Load Time the picture was completely different. Media sites were noticeably slower and the slowest pages took up to 23 seconds to load.
So where is the time actually lost? To figure this out I created a chart that visualizes the lifetime of page loading. The first time is lost between First Impression and the onLoad event but the by far biggest chunk is lost after the onLoad event.
What makes pages slow?
Having done quite some page optimization analysis I know that content size and number of resources are typical sources of performance problems. While there is not a big difference among the top pages across industries it is noticeable that the biggest pages are larger by a factor of 5 to 10 times compared to the best in the industry. Big media sites for example require up to 3.8 MB of content. This might create the impression that these pages look very different.
However looking at MSN (fast) and CBS (slow) shows that there are not really big difference from the content that is shown.
Performance differences beyond content
So what can we learn from this exercise? First of all there are differences in performance between industries. At the same time the top pages across industries show quite similar and pretty excellent performance characteristics. So the argument that a site cannot be fast just because of the specifics of it’s industry does not really hold. If you want to learn how your site performs compared to others in your industry try speedoftheweb.org
Alois Reitbauer (@aloisreitbauer) works as Technology Strategist for dynaTrace software and heads the dynaTrace Center of Excellence. As a major contributor to dynaTrace Labs technology he influences the companies future technological direction. Besides his engineering work he supports Fortune 500 companies in implementing successful performance management.