I’ll never forget a conversation I had with a customer many years ago about how success would be measured at the end of a proof of concept. At the time I laughed at the answer, but looking back on that conversation I realize they were right. Their goal: “We want our users to be happy.” They knew something then that I have since learned - it doesn’t necessarily matter how fast or slow a page loads, if the user thinks it is fast and they aren’t frustrated by the experience then you have a good user experience AKA happy users. The user’s perception is more important than the actual experience.
There is a lot of talk about wanting to create a good user experience, but how much of the user experience is within our control? The problem with web performance metrics is they focus on objective time or clock time as this is measurable and can be used for comparisons. We focus on time to first byte (TTFB), start render, document complete and more recently speed index.
These are all valuable objective metrics for comparison but what is missing is the subjective experience on the passage of time, known as psychological time. The perception of the end user can be more important than the objective measure of time. In the web performance world we are focused on objective time and look for ways to shave milliseconds off load times, even if the user won’t perceive the page as being faster. Steven Seow in Designing and Engineering Time: The Psychology of Time and Perception in Software presents the 20% rule: a response time needs to be 20% faster for an end user to perceive it as being faster. That means if a page takes 5 seconds to load and it is reduced to 4.5 seconds an end user will not perceive this as being faster, the page has to load in under 4 seconds to be perceived as faster. This applies to whether you want your site to perceived as faster than a competitors or faster than a previous version of your own site. To learn more about objective and psychological time check out Denys Mishunov's 3-part series in Smashing Magazine.
Not only do people experience time differently their perception is also influenced by biases. Everybody is biased and these biases influence how a user can perceive an experience. If you think you aren’t biased that’s a bias-blind spot. Wikipedia has a great list of cognitive biases that include decision making, belief & behavioral biases, social biases, and memory errors & biases.
A third piece to the user perception puzzle is people are not always rational, and sadly this irrationalism impacts our perception of the world around us. Decisions are often driven more on emotion than on logical thinking. Think about the recent billion dollar Powerball jackpot, I don’t play the lottery as I know my chances of winning are close to zero but that didn’t stop me from buying a ticket for that mega jackpot. The purchase was driven completely by emotion, logic had nothing to do with it.
How do these biases and irrational beliefs influence the user experience?
First experiences are imprinted on our brain. As the saying goes you never get a second chance to make a first impression. If the first visit to a web site is bad this will be remembered forever. We tend to rely too heavily on the first piece of information we acquire about a subject.
People are bad at comparing things. People may not be able to judge which site is faster than another especially when viewed at different times. We tend to find more differences when comparing items simultaneously.
Negativity Bias & Selective Perception
Negative events have a greater impact on an individual than a positive experience does. If a user visits your site 99 times but 1 time they have a negative experience they will remember that negative experience more than the 99 times they had a positive experience. This may then lead them to expect a negative experience on future visits.
The inability to compare small differences. When shaving milliseconds off a response time, users may not notice any difference as the just-noticeable difference (JND) has not been reached. An improvement of 500 ms will be more noticeable on a page loading in 2 seconds than a page loading in 6 seconds.
Where to go from here?
We have a plethora of metrics to measure the user experience but very few ways to measure the user’s perception. There have been a number of “tricks” that have been utilized over the years to make the page appear to load faster, the challenge is investing time and resources into making a change that “feels” faster without any way to back up the feeling with data. What we need is a way to measure the user’s perceived experience which is subjective and driven by emotion.