A recent blog post at Radware took on two sacred cows about website load times and said what many of us think about image loading and page speed is wrong. They based their claim on neurological testing they commissioned.
First, Kent Alstad says in his post, progressive images are irritating. The rule of thumb has been that when images load to a website progressively, users see that the site is working and they appreciate it. The alternative is that the image only appears once it is fully accessed by the browser.
But test subjects had a “definitively less positive reaction” to progressively rendered .jpg images, Alstad says. “We found that … cognitive fluency is inhibited and the brain has to work slightly harder to make sense of what is being displayed.” (I recently wrote about the role cognitive fluency plays in conversion optimization.)
Second, the study concluded that load speed alone doesn’t improve the level of engagement for website users. Instead, “the sequence in which a web page renders has a substantial effect on visitors’ emotional and cognitive response and (the) order in which they will look at different items.”
But the effect of a page’s rendering sequence depends on the context and design of the website, as well as the speed of the page. Rendering sequence has to be tailored to the page’s context and paired with appropriate page optimization techniques to ensure the maximum amount of user engagement.
The results of Radware’s neurological testing points up the never-ending need for testing as a part of conversion optimization. The results of their tests are interesting but, to me, they mainly say this needs to be tested further.
Load Testing Under-Rated as a Conversion Booster
The load time of your web page, including images and other elements, should be among many factors tested as part of your ongoing conversion optimization efforts.
Most of the information about “testing” that you’ll find when you research conversion optimization is about A/B testing. That’s great; A/B or split testing is a crucial component of optimization. But it is far from the only type of testing required to optimize the user experience on your e-commerce website.
I’m guilty of an over-emphasis on split testing, too. I use “A/B Testing” as the head on my testing services page, and I’ve blogged about A/B testing more than a dozen times on my own site as well as for other sites.
But load testing, seeing how quickly a site fully appears and is functional in a browser, is often the initial work I perform for clients.
I’ve written about load times before, too, stating that page load time is one of the most overlooked areas of conversion optimization, yet also one of the quickest ways to improve conversion rates.
You can look at that post for several examples of how businesses decreased their load times and quickly increased conversion rates. That post is about a year old, so here’s a more recent example boasting a 38 percent lift, which you can also check out.
Or you can just trust me – and your common sense – when we both tell you that faster loads lead to more conversions.
Test and Improve Your Load Speeds
There’s no wizardry to testing your page / site speed. You just have to know to do it and then follow up by fixing any problem areas the test turns up.
But, first, to review, a good rule of thumb says a load speed of 1 or 2 seconds is your target. Three to 7 seconds is not bad; Google said last year the average page load time is about 7 seconds. But you would still want to strive for faster. Once you get beyond 7 and approach 10 seconds or more, you have a problem.
You can find many page-load speed testing tools online. Two I’ve recommended in the past are Google’s PageSpeed Insights and Pingdom. They’ll not only give you a speed/grade for your site, they’ll suggest things to do to speed it up. There are many, many more test tools you can find by searching. Test your site with a few of them to confirm findings and to find the one whose instructions for improving best speak your language.
Here are several typical suggestions for changes to make that will decrease your load times:
- Optimize images, that is, properly size them and compress them
- Clean up CSS code to optimize delivery
- Compress content, i.e., eliminate elements like spacers, image segments, Flash elements, background audio loading, autoplay rich media, etc.
- Reduce the number of plugins or make sure you have newer, more efficient versions
- Deliver cached copies of requested content when available
- Set time limits for how long databases cache information.
Make Speed Inducing Changes – Or Find Someone Who Will
If my suggestions above have you scratching your head, don’t feel bad. A marketing professional won’t necessarily know and probably doesn’t need to know how to configure a website. This is partly why I get funny looks and hesitant responses when I start a conversion optimization engagement by checking load times and suggesting the kinds of steps listed above.
But your technical folks know about this stuff. If they’re not handling it correctly (which your use of the test sites may indicate), you should have some questions to ask them. Or just send them a link to this post.
If you are on your own and don’t understand coding and configuration and such, one solution is to obtain a dedicated or managed hosting service that provides these kinds of services. Or, you know, hire a consultant.