Thought Leadership


Subscribe to our blog

Your email:

Connect with Ness

Software Engineering Services Blog

Current Articles | RSS Feed RSS Feed

Speed Matters - But Performance Improvement Takes Time


About a year ago, researchers from both Google and Bing published findings showing, lo and behold, that user satisfaction declined with slower load times for search results.

While that caught no one by surprise, the fact that delays as short as 200 milliseconds had a measurable impact on user satisfaction did. Not only that, there was also evidence that negative experiences with site performance had a lingering effect on site usage even after speeds increased.

Furthermore, when presenting these results at the Velocity 2009 conference, the folks from Bing indicated that a 2 second slowdown could reduce revenue per user by more than 4%.

That the impact of site performance reaches far beyond the world of search was reinforced more recently by a study commissioned by Gomez which showed, among other things, that 88% of people who've had a poor experience on a website are less likely to return, that 47% leave with a negative perception of the company, and that 43% discuss the experience with their friends. 

The Gomez study focused specifically on user experience during "peak times" (holiday shopping season, Mother's Day, Valentine's Day, etc.) when one would expect high traffic volumes to affect site performance. Unfortunately, web users don't really care about that. They expect your site to perform well when they are using it (and frankly, they are more likely to use many sites precisely during these peak times); a negative experience is a negative experience to them whenever it happens.

Add into the mix that Google now uses site speed as a signal to help determine search ranking and suddenly you've got something to think about if the web plays any kind of role in your business!

The good news is that there are things that you can do to improve site performance, including elements that are of particular importance to ecommerce such as onsite search. The bad news is that the process of improvement takes time (in fact it never stops) and it involves making informed trade-offs. 

First of all, you need to be monitoring site performance and diagnosing problems. There are a variety of tools out there that serve exactly this purpose. This sort of monitoring can't be episodic; it needs to be ongoing and should inlcude things like regular load testing, as Gomez recommends.

Second of all, as you consider and add features to your site you need to be testing the effect they will have on site performance. As Google puts it, "Because the cost of slower performance increases over time and persists, we encourage site designers to think twice about adding a feature that hurts performance if the benefit of the feature is unproven."

Third of all, as that last statement implies, you need to be ready to make some trade-offs. For example, the Gomez study showed that 53% would abandon a travel Website at peak traffic times and book somewhere else after only one or two bad experiences. At the same time, however, onsite content (pictures, videos, detailed descriptions, comparisons, etc.) can have a huge impact on travel site conversions. Certainly you can optimize this content, but even if it is having an impact on performance you need to consider the business impact of not including it.

Long story short, site performance is always something that can be improved and these improvements can have positive effects on your business. To take but one example, at Velocity 2009 Phil Dixon of Shopzilla described a year-long performance redesign they undertook which reduced load times from 7 seconds to 2 resulting in revenue increases up to 12%. 

Still, while 12% more revenue definitely sounds great, don't forget the "year-long" process that led to it. These things take time. That's why my fourth and final recommendation would be: Start now.

blog comments powered by Disqus