Monday, April 18, 2011

Website Performance – Are you doing enough?

If you’ve read my earlier postings on the beneficial effects that webpage performance improvements can have on SEO (and this recent video where I also give a basic explanation of the eCommerce implications), then you will know I’m pretty keen on the subject of website performance optimisation. This perspective is all part of my overall view that a company’s website needs to work as hard as possible and at all times.

In my opinion, minimising the time that content takes to appear on the average user’s browser (or range of browsers) should have the same sort of priority as reducing call waiting times in a customer services centre or ensuring diners are quickly seated at a restaurant. In all three of these cases, a faster time to deliver the service is what matters.

Speed counts… and in the case of a site that transacts, speed can have a real impact on the bottom line.

So what’s a fast page response time and what’s a slow one? The complex answer to this question is “its complex”, but luckily the simple answer is “it’s simple, there are several online sources to help you”. Including: http://pagespeed.googlelabs.com/ This is the first place I would visit to measure the speed of your website pages. The performance from Google is scored and suggestions given to improve your page speed in: high, medium & low priority (as well as other best practice recommendations). For example the homepage of http://www.idealinterface.co.uk/ gets a score of 85/100 http://pagespeed.googlelabs.com/#url=http_3A_2F_2Fwww.idealinterface.co.uk&mobile=false Note: You can currently only use this site to measure one page at a time. It is not possible (as far as I am aware) to record a series of actions.... such as a user’s typical transaction process through your eCommerce site.

So don't dismiss the subject of website performance. You may find it could mean the difference between a good website and great one.

1 comment:

Charlie The Dog said...

It's interesting to note that the Google's explanation of the result is not based on how long it takes for a page to load but how much room for improvement there is.

My website www.klyndesign.co.uk gets a score of 89/100 so no alarm bells are ringing.

You mention 'the average user's browser'. This is quite prevelant to my current project whereby the website I am working on will soon have a worldwide target audience and the average users browser can be very different between countries.

In Germany Firefox is streets ahead in popularity, while in other countries Firefox is less so and Internet Explorer still has a stronghold.

Therefore, for websites with a worldwide target audience or specific country audiences, the browser type ideally needs to be factored into performance testing rather than a simple page load stat as the different browser technologies will have an impact on the last mile page rendering, and a performance issue may only manifest itself in a specific browser.