The Cost of Website Obesity
I've been creating websites since the mid 90's, and in that time the web has changed a lot, mostly for the better in terms of style, but certain problems have persisted and I'm going to focus on just one for now: size.
So what about a simple blog? Lets take a look for ourselves...
I decided to compare one of the blog pages on this site to a similar type of page on the popular blog hosting site, Wordpress. I could have made the comparison with almost any site but I went with Wordpress because they're the market leader in terms of customer base and I'm very familiar with Wordpress as I've been using them myself for probably 10 or 12 years. I'd describe myself as something of a Wordpress fan.
In order not shame any specific persons Wordpress site I compared a page on their "demo" site.
This is their Wordpress page based on the "Independent Publisher 2" theme.
It has an eye catching image the full width of the page, uses a clean layout with no ads etc.
The article itself is short with only a single photo they do have some tiny thumbnails to represent the author and a commenter.
The theme is written in the latest HTML 5 using CSS 3 and is a responsive design that automatically adjusts itself to fit different sized device screens, from smartphones through tablets to full desktop monitors.
I really like it.
This is my WenPress page based on the default "blog" theme.
The "featured image" at the top of the post is a similar size to the one in the Wordpress page, again there are no ads and it's a clean single column layout.
The article itself is much longer and has 2 additional photos so we're talking more "content" than the Wordpress "demo" site but that's to be expected.
There are no social media icons or user thumbnail photos but I feel the other 2 images in the page content more than balance that out. If anything there should be a size advantage to the Wordpress page as it has less content...
To demonstrate the time difference in loading speed between a Wordpress blog and a WenPress blog, I used the excellent simulator on www.webpagetest.org. to simulate loading the page via a good old 56k dial-up modem.
We're looking at two numbers. The first is how long before the page appears in the browser so the user can interact with it, and the second is how long it takes for the page to completely finish loading.
|Site||Time to Interactive||Time to Finish|
|Wordpress||34.7 seconds||94.6 seconds|
|WenPress||4.7 seconds||18.8 seconds|
I don't think anyone would want to go back to browsing pages via a dial-up modem, although it's a daily reality for more people than you might think. Even in the USA, AOL has over 2 million dial-up subscribers, even if that were only 0.5% of the market, globally that's a lot of people...
The sub 20 seconds managed by WenPress under those circumstances is probably tolerable but I'll bet a rare thing on the wider web, and the painful 94 seconds taken by Wordpress is probably a lot better than most mainstream sites. I suspect the first 34 seconds it takes to even begin displaying the page is more of a frustration than the total load time, the sub 5 second it takes WenPress must be nirvana by comparison... To understand this massive difference in performance we need to take a closer look at what constitutes a page.
|Images||74.0 KB||245 KB|
|CSS Styling||3.3 KB||59.1 KB|
|HTML||6.0 KB||30.0 KB|
|TOTAL||83.3 KB||433.4 KB|
Is there any reason to be concerned about unnecessary page bloat beyond the sub 1% of users who are accessing the site via modem? Even developing countries have higher bandwidth from their cell service these days right? Maybe, but even if they do, there is the cost of that metered connection to consider.
Cost to Browse
The cost of Internet service varies by country, just as income does. There is an excellent website, www.whatdoesmysitecost.com, that can tell you what it would cost for a user in a given country to access a page on your website, either in cents or as a percentage of their daily income. So let's compare the two sites from the point of view of a user in Madagascar:
Browsing 100 pages per day would cost them 22% of their daily income.
Browsing 100 pages per day would cost them their 4% of their daily income.
It's a sobering thought and another indication that the digital divide is a very real thing.
I firmly believe that access to information should be free, wherever you are in the world, but if as a society we can't afford to make it free, we should strive to make it as cheap as possible. And no, displaying ads and selling data about your users is not giving them something for free. That comes at a very high price...
Granted I can't change the business model of social media networks or search engines, but I can make accessing my own sites as cheap as possible by keeping things as efficient as I can. And as a side benefit, the more efficient something is the less it costs to run, so I don't need to display ads or sell my users data.