Web Development News, Tips, Insights | Website Magazine

Optimizing for Speed

Written by Linc Wonham | May 9, 2011 5:00:00 AM

Web marketers have spent the last couple of years getting a grip on web analytics and A/B testing so they can measure user behavior, learn what works best and methodically improve their sites.

 

Today, when we talk about site optimization, we're usually referring to the content that a user sees - the graphics, copy, layout and offers - and then testing is focused on tweaking these elements to make the site more persuasive, usable, relevant and engaging. While these efforts are absolutely necessary, now that measurement and testing have become staples of the web marketer's toolbox, another trend is emerging: optimizing for speed.

 

Recent studies published by Amazon, Google, Bing, Yahoo and others reveal a direct connection between site speed and user behavior. The faster a web site's page-load speed, the better the user experience, as measured by higher conversion rate and greater engagement levels. It's fair to say that given two sites which are identical in every way except for speed, the faster site will have the competitive advantage.

 

So how do you measure and optimize for speed? Free tools like webpagetest.org and the Yahoo! "YSlow" add-on for Firefox can provide you with a speed grade (A-F), detailing why your site is slow and then making recommendations for improvement. There is also an emerging industry called Web Performance Optimization (WPO) which specifically aims to deliver products and services which make it easier for site owners to shave seconds off their page-load times.

 

Fortunately, not only are there tools and well-documented techniques for increasing speed, but as a Web marketer you already have two of the most important tools already in your arsenal: Web analytics, for measuring changes in KPIs like conversion rate, and A/B testing, for learning how users respond to site changes.

 

So, when trying out speed-savings techniques - for example, combining multiple JavaScript or CSS files into one - you should always test them first and then measure impact on user behavior. Testing and measurement remain two of the core capabilities that enable you to continually improve your site. Only now, you should be thinking not only the visual elements, but also the underlying page elements that influence page speed.

 

This post was contributed by Eric J. Hansen, founder and CEO of SiteSpect.