Last week I wrote how site speed was starting to play a greater role in Google Rankings. I wrestled with myself a little over whether or not to publish that post because I knew that the bpodr site wasn’t as well optimised for speed as it could have been (WARNING: Low flying excuses incoming…) The site had only recently been re-launched under tight timescales, we have a lot of client work on at the moment, and the page speed of bpodr.co.uk wasn’t as important as getting the site up-and-running, and publishing useful content. However, the speed of a website is important, and this website is no different.
I decided to post it anyway, and use it as an opportunity to ease my embarrassment about bpodr.co.uk’s speed issues – I’d set myself a 7-day challenge: Improve the speed and performance of bpodr.co.uk.
I knew what I needed to do, I knew how to do it, but I was struggling to find the time. A public announcement that I was going to improve things seemed the perfect piece of pressure, and it proved to be exactly what was needed (of course, there’s also the danger that I could have failed miserably, and shattered any trace of credibility I may have). Anyway, here’s the results of my 7-day page speed challenge:
bpodr.co.uk: Now 60% Slimmer and 33% Faster!
How we did it…
The last post outlined a few techniques for improving the speed of a website, so I started my speed improvements there. I decided to implement these small changes to the bpodr.co.uk home page, and ensured none of the changes affected the look or the content on the page. I documented the size of the page, and the number of objects (images, files, scripts) that are requested when the page is loaded, at the start of the challenge and after each round of changes. This showed me what gains (or losses) I made for each speed enhancement.
N.B. I didn’t measure the speed after each change – it’s a fair conclusion that speed, although relative to each user (based on their computer speed, browser used, connection speed, etc), would improve as the page got lighter, and the number of images downloaded decreased.
Before we started…
Firstly, I wanted to know how things were before I started making changes, so I benchmarked the speed and size of bpodr.co.uk using Browsermob.com.
The results were: Size: 431kb, Average Speed: 2.30 seconds, Number of Objects:32.

Compared to Google’s published web averages, this made us about 25% larger than the average website waddling around in cyberspace, with a few extra objects being loaded. A little bloat around MY waist I can handle, but I’m not prepared to accept it on a website of mine, and so I needed to draw up a plan to shed those extra kilobytes, and take bpodr.co.uk from overweight to fighting weight (cue rocky music…)
The website weight loss programme begins…
1. Combine JS and CSS files
I started with a quick win: Reduce the number of files that the browser has to load for the page to work correctly. A lot of sites call multiple Javascript files, some of which are hosted externally to the site that’s loading, as well as multiple CSS files (CSS files control the way your web site looks). By reducing the number of files that need to be loaded, there’s less trips between the server and the user’s browser, saving precious nano-seconds! I managed to reduce the number of files on bpodr.co.uk from 5 to 2, and also moved a file from loading on an external server to our own server.
|
Before |
After |
Gained |
| Size: |
431kb |
417kb |
+ 3.25%
|
| Files: |
5 |
2 |
+ 60%
|
| images: |
19 |
19 |
+/- 0
|
2. Compress Image Sizes
Compressing image sizes takes a little longer than reducing the number of scripts loaded on a page. The benefit here is in reducing the size of each image your page is loading (as long as you’re not compromising the quality of the images). There were 10 images that were in need of a good compression on the bpodr.co.uk site…
|
Before |
After |
Gained |
| Size: |
417kb |
222kb |
+ 46.7%
|
| Files: |
2 |
2 |
+/- 0
|
| images: |
19 |
19 |
+/- 0
|
3. Turned images to Sprites
Creating sprite images can be a time-consuming task – it’s basically converting multiple background images from a page into just one image. That one image then loads once, but displays a different portion of itself to show the various images contained within it (uncross your eyes now!). The page benefits from having to load less separate images from the server – less trips to the server = faster loading, yippee!
|
Before |
After |
Gained |
| Size: |
222kb |
172kb |
+ 22.5%
|
| Files: |
2 |
2 |
+/- 0
|
| images: |
19 |
12 |
+ 36.8%
|
There are other compression techniques that were already implemented on the bpodr.co.uk site, such as file compression, and some caching of files and images. If I hadn’t implemented these on the site during the build, I’m confident we would have seen some extra improvement to the size and
speed of the site.
The Total Site Improvement…
We already know that I reduced the waistline of the page by around 60%, but let’s see it in a fancy table:
|
Before |
After |
Gained |
| Size: |
431kb |
172kb |
+ 60.1%
|
| Files: |
5 |
2 |
+ 60%
|
| images: |
19 |
12 |
+ 36.8%
|
So what was the final result on the speed test?

From 2.30 seconds to 1.53 seconds average loading time: a 33% increase in speed! It also means we’ve got a significantly smaller size, and less images loading than the web’s page average (according to Google). So the page is now slimmer and faster, and better than your average page – a true athlete amongst web pages. bpodr.co.uk snr would be proud.
These improvements show us that big gains can be made to the size of a site, and therefore it’s size, using some fairly basic techniques. We can draw some conclusions from this 7-day challenge:
- The biggest size reduction came from compressing images,
- All of the above activity improved the size of the site,
- Less visits to the server and smaller file sizes equals faster loading times for sites.
How optimised is your site for speed? Remember, if it’s not well optimised, and below the ‘Google average’ you could be doing your site rankings some harm. It’s not going to result in jail time, but it may mean your site isn’t working as hard as it could be.
Why not set your web developer the 7-day page speed challenge, and trim some weight off of your web pages…