On the web, page speed matters. If your site takes too long to load, people will go elsewhere. Google proved this by purposefully slowing down its search engine. They found that even just half a second caused fewer searches. Bottom line: users love fast sites!
With that in mind, I decided to look at local media sites – how fast or slow are they? Rather than looking at page load times, I decided to use YSlow to determine the performance of each site. Lots of factors can impact the amount of time a page takes to load (your ISP, the speed of your computer, where you are geographically,etc), but everyone has to download the same amount of data for a web page, so I figured YSlow is a little more fair than a stopwatch.
Here are the fourteen sites I looked at (you can download all the data here):
As you can see, none of the sites received an “A” grade. The only one to receive a B was Only Here for the Food. The grade can be somewhat misleading, however. Here’s what the front page performance for each site looks like:
It turns out that Sharon’s blog is the heaviest of them all with an empty cache (the first time you visit the site). This is due to the large number of images she has on her site. In fact, almost all of the weight of the page is due to the images. If you look at the primed cache (subsequent visits to the site) then the Edmonton Journal is the heaviest. The Edmonton Journal has the worst performance improvement going from an empty to a primed cache:
To be fair, I decided I should compare an “article page” on each site as well. With social media in particular, an article page is more likely to be a visitor’s entry point to the site. For this test, I simply clicked on the top article on each site:
One caveat: I used the second story for iNews 880, because their top story was over 25 MB in size! Evidently they think it is fine to embed full size, uncompressed images.
As you can see, Valerie’s blog is the heaviest, again due to the number of pictures she has. Once again, the Edmonton Journal has the worst performance improvement going from an empty to a primed cache:
I think perceived performance is often more important than actual performance, but that’s obviously harder to measure. In my experience, most of these sites load fairly quickly. When I do notice a speed issue, it’s usually because the page I am trying to load has a lot of stuff on it.
Another thing I learned from this exercise is that all of the sites have room for improvement!
What has your experience been like? Which sites do you find slow?