Could you beat Steve Souders in a foot race? To find out, the first thing you might do is watch 10-seconds from his talk The Illusion of Speed, at Velocity-Ignite 2013. Which 10-seconds? Start at the 7-second mark and watch until the 17-second mark. Reason? The running performance data that is the most useful is there. And that data is not a number.
Addressing web performance continues to expand and deepen along a spectrum of approaches and models. One might say that at one end of the spectrum “big data” analytics/quantitative have dominated performance discussions, and at the other end of the spectrum Steve’s The Illusion of Speed has highlighted a purely qualitative concept in the form of perception and “perception brokers”.
Starting in the Middle
Here in the humble Midwest (specifically Minnesota), there is a certain bias towards valuing the middle of spectrums in the form of applauding common-sense technical and “mental shortcuts” to solving problems. (There may even be room to call this middle way, a heuristic model.) Whatever the label, this highlighting of the middle of the spectrum will hopefully add value and spur discussion.
The case for using Perception and Analytics…Together
Specifically, this heuristic model involves adding a video playback of a web application (perceptual data) to a waterfall chart of element load times (analytics) pulled from web application monitoring. The captured video is synched with the captured data, so that the user can cognitively mix the video with data. So, an experience of the web application and an analysis of the data from the web application is used together to make “mental shortcuts” for performance decisions.
To whit: the most useful aspect of running performance data in Steve Souders video is a picture of Steve running. Tasked with making performance decisions based on limited time and limited analytical resources mental short-cuts are not uncommon. The video synched with data method (we call it UserView Monitoring™ with Video Capture) merely puts that mental short-cut squarely on the table and, where applicable, for use in achieving web performance goals.
Back to Steve running….If you have any experience with running then the picture of Steve running alone is, in fact, “experience enough” of performance data to make a few assumptions about Steve’s running abilities. Moreover, you can take a mental shortcut and extrapolate a few things about Steve’s age, height, and conditioning – enough so, to make a determination whether you could beat Steve in a foot race.
Also, if you had quantitative data alone about Steve’s running performance such as – miles per hour, distance run, top speed etc…- you’d a good chance at determining what size bet you want to place on racing Steve.
However, if you have both a video (or picture) and quantitative data you have, in many cases, the best chance of knowing that you could place a bet with Steve and, if you win the foot race, and say…win a free slot at Velocity 2014 as a speaker.
Data is not enough and Perception is not enough
In many circumstances quantitative data isn’t enough. For example, if all you had was the quantitative data on Steve’s running performance, but not the picture, Steve could theoretically hire the fastest man alive, Jamaican sprinter Usain Bolt, to stand in for him at a race and win the bet. The quantitative data wouldn’t be enough for you to determine that the “Steve” you just lost to just had a slow day when you looked at the original mph etc…data.
Moreover, perceptual data, like a picture, is also not enough. For example, at the 24-second mark Steve’s presentation launches down the perceptual rabbit hole of Matrix-esque references to the “Red pill, Blue pill”, social experiments in perception, etc… However, part of what is challenging about the topic of perception is that perception itself is a tricky topic – as Steve’s talk highlight, perception is filled with illusions.
A perceptual illusion of the cups shadow makes grey in squares A and B appear to be different shades (they are the same) in the mind’s eye.
But, how is subjective illusory information quantified as “data”? How do you use subjective illusionary data to make decisions? Too improve performance of a website? …well you get the idea.
Going down the rabbit hole of perception, ala The Matrix, involves many intractable problems with perception; psychological, neurological, and philosophical. Moreover, breaking “perception” down into, say user experience, runs smack dab into the problem of Subjectivity. Souder’s inadvertently notes the subjectivity issue with experience starting at the 14-second mark of the video, “if you look at the picture” (ah, visual data) “it looks like (sensory data) I’m going pretty fast” (compared to everyone else in the room he was) “and I’ll admit I felt (subjective perception) like I was going pretty fast” “but I wasn’t going fast…I mean I’m too old for that” (relative to who, age data may in fact be helpful).
Taking the pill that wasn’t offered…purple
There’s big data and then there’s “perceptual” data, such as that of a picture. The ability to combine the two, in some cases, helps speed up and inform decision-making given limited time and resources. Dotcom-Monitor has developed a tool that syncs video playback of web application performance with a waterfall chart of web page load time data the helps in that process.
As was noted by the red pill/blue pill note in Steve’s video at the 24-second mark, there is a temptation to make a choice based on what is presented. However, the ends of a spectrum also provide the opportunity to address questions from an entirely new perspective – that of a consolidated middle. Therefore, given an option of taking the Red pill A. Steve’s running performance data (foot speed, distance of the race, etc…), or B. Blue pill, seeing a more subjective picture/video of Steve’s running form to determine if you could beat Steve Souders in a foot race the answer can often be the pill not offered – a mixing of red and blue – the “Purple pill”.
Brad Canham, is the VP of Sales and Marketing at Dotcom-Monitor, an endurance athlete and Ironman finisher. Based on a quick review of personal quantitative performance data and the picture of Steve Souder’s running, Brad believes he would beat Steve in a footrace (unless Steve hired Usain Bolt… then Brad believes he would finish in the middle spot).
Editors Note: You can try the full version of Dotcom-Monitor for 30-days free of charge: Learn More Here.
Brad Canham (@BradCanham) is the VP of Sales and Marketing at Dotcom-Monitor. He's passionate about leading an organization that believes deeply in the principle of “constant improvement” and delivers that to its users via web performance monitoring. He blogs at http://www.dotcom-monitor.com/blog/ and organizes the Web Performance-Minneapolis/St. Paul Meetup group. When he's not talking and writing about constantly improving performance, he enjoys spending time with his family, racing in Ironman triathlons, snowshoes races, sipping craft beers, and reading everything, everywhere, all-the-time.