When Research Findings Challenge Our Conventions

What do you do when you discover something that might be disruptive in your field?

For some, the notion of sitting on findings that might rock the boat might hold some appeal, especially if those findings challenge the conventions dominating the way we work and approach problems. After all, they wouldn’t be “conventions” if they weren’t widely-adopted and accepted.

First Sacred Cow: Progressive Image Rendering

Last year, we commissioned neurological research with Neurostrata in Europe to examine progressive image rendering and whether or not it was actually beneficial to the user experience. Images compose the bulk of most web pages, and many site owners have rightly sought to optimize their pages for speed by addressing how they handle their images.

In the case of progressive image rendering, we had this to say:

“Much of the debate over image rendering can be reduced to baseline versus progressive images. This has been a contentious issue among designers and developers since the early days of the public internet. The progressive faction believes that this format improves perceived performance by showing the user something while they look at the screen. Proponents of baseline images, however, believe that watching an image load progressively increases user frustration.”

We decided to introduce neuroscience to the debate, since even informed opinions can still lead to an argument that doesn’t go anywhere. There are still strong opinions on both sides of the debate, but what would we discover through the use of Facial Action Coding (tracking micro-expressions) and an Implicit Response Test (measuring relative frustration and emotional engagement)?

The results were interesting, and the 280 test subjects had a definitively less positive reaction to the progressively rendered JPEGs, showing irritation. We found that when, as with the Progressive JPEG method, image rendering is a multi-stage process in which an initially coarse image snaps into sharp focus, cognitive fluency is inhibited and the brain has to work slightly harder to make sense of what is being displayed.

Those were the findings – and some people didn’t like it – which we were reminded of while presenting this year’s research findings at the O’Reilly Velocity conference in New York. After recalling the first study, one attendee thought the assertion that progressively-rendered images weren’t uniformly better than baseline images irresponsible, since those in emerging nations have typically lower-end mobile devices connecting over slower networks.

But we didn’t present a mere opinion – the research could certainly have shown that progressive image rendering was superior for the user experience, had that actually been the case.

We published the paper, even though the findings were disruptive to the conventions.

Second Sacred Cow: Page Speed

This year’s research revolves around the mantra that “Faster is Better” in site loading, tied to the “Three Second Rule” which warns us about the threat of users leaving websites if their pages don’t load within an optimal three second window.
But what if it was more than merely rote speed that actually matters to users? Yes, people want pages to load quickly, which is reinforced by numerous studies showing a correlation in revenue and page speed. But what about people’s internal reactions to site rendering?

Again, we turned to neuroscience to find out; employing eye-tracking equipment and EEGs for neuroimaging to see if merely manipulating a site’s rendering sequence along the lines of an across-the-board speed bump would net positive results in user engagement.

Instead, we didn’t find speed to be the key ingredient by itself in improving the level of engagement for the participants. What we found is that the sequence in which a web page renders has a substantial effect on visitors’ emotional and cognitive response and at which order in which they will look at different items. At the same time, rendering sequence effects also depend on the context and design of websites as well as the speed of the page they are visiting. Some load types are optimal for certain websites, while they fail in other designs.

That last paragraph is certainly less pithy than “faster is better,” but it’s what the neuroscientific research found. The nuanced nature of these findings may be difficult for a marketing department to summarize for Twitter, but they are important in helping us with the direction we take for our products and services. In this case, it’s about the interplay between speed, context and rendering sequence.

The actionable part in this year’s research is this: Rendering sequence – when tailored to the page’s context and paired with appropriate page optimization techniques – will offer the maximum amount of user engagement. This requires a web performance automation solution combined with human intelligence (including a clear understanding of the underlying principles) to determine the optimal rendering sequence for your web pages.

Science Demands Transparency

There are certainly plenty of sacred cows in any industry. There are widely-held attitudes and beliefs, with some being right, and others proven wrong over time.

When there is a level of controversy on how to proceed, introducing actual science into the mix can be illuminating, removing the emotion from the equation.

When we get the results, we all have to ask, “What does this mean?” and just as important: “What will I do with this knowledge?” 

It may mean challenging conventions.

You can read the full report here: Speed vs. Fluency in Website Loading: What Drives User Engagement?