The Uncertainty Principle

Building better websites by learning from our exploration of the known universe


“One certainly cannot predict future events exactly if one cannot even measure the present state of the universe precisely!”

Stephen Hawking wrote this in his book A Brief History of Time to describe the uncertainty principle, which has had profound implications on the way we view the world and ultimately led to the creation of quantum mechanics. Sure, it’s weighty material, but it has some very real applications in the world of the Web.

Most organizations out there already have a website. So, as an agency, we’re often building someone’s second or third website, rather than their first. As we look at where to take a website project, we see a particular correlation between the emergence of quantum mechanics and web design: that there is not one clear outcome from our observations, but several possible outcomes with varying levels of probability.

Obviously, building a website is not the same as understanding the known universe; however, we can learn a lot from the people who have dedicated their lives to the pursuit of the final frontier.

One of the major challenges that has always faced science has been the ability to accurately measure things; see, for instance, Clair Patterson’s efforts to eliminate lead from his instruments so that he could accurately predict the earth’s age.

When we look at technology — especially website design — there is so much that we can measure. The value of measurement is so profound that it’s no longer an optional perk; it’s a necessary part of a site that truly works for you. While the code builds the function of the site, research (and measurement) determines the function you need. We can’t know what to build next if we don’t know how what we’ve built is performing.

With web design measurement available at your fingertips, and for little to no cost, we have progressed light years from the visit counter at the bottom of our websites — the metaphorical equivalent of looking at the stars through the crudely made telescopes of Galileo’s time. Now we can map the online galaxy with finely tuned instruments like Google Analytics that tell us not only what site visitors are doing, but where they came from to do it.

While the technologies of science have evolved over hundreds of years, web design (and development… and measurement) has evolved in less than a decade, and the rate of growth happens even faster. Over the past few years, data science has given rise to heat- and and eye-mapping technologies, and we are even beginning to see real-world analogs in tracking techniques.

“In general, quantum mechanics does not predict a single definite result of observation. Instead it predicts a number of possible outcomes and tells us how likely each of these is.”
When reading this defence, the likelihood of a catch seems incredibly low, however it is one of several possible outcomes

An amazing advantage of being able to capture more (and different) data is that we can understand the customer journey so much better, and as a result, can make better decisions about website modifications and improvements. Are users reading the content we have on the site? Could users get more information from the site if we simplify it? These are just a few of the ways that we can apply the scientific method to web design:

Purpose

When we build a new site, we always begin with the question, “Why do we have a website? Because we have to,” no longer cuts it as an answer. We need to know what users expect when they come to the site so we can give that to them (and make it easy). Are we answering commonly asked questions for customers, creating high-quality leads for the sales team, increasing online sales for the organization?

Research

Once we’ve established the “why” of building a site, it’s important to understand the variables to its success. Who is accessing the site? What are they using it for? Where did they come from? When do we see spikes in traffic (e.g., seasonally, certain times each month)? What other reason might someone visit that we aren’t addressing yet? How are people accessing the site (mostly by phone or primarily from a desktop)?

We need to understand the organization that’s requested the website, and the users that will be navigating it. Some of that comes from the questions above, but it also helps to look at past website measurement, and industry trends.

Hypothesis

Now that we know who we are building the site for and why they are there, we can build a site that meets their goals.

Let’s say that our client tells us that they want to improve customer satisfaction rates. Let’s also imagine that we look at call centre data and find out that 90% of customers are asking the same 10 questions, and that the call centre is so overloaded with enquiries that customers are frustrated when they finally reach someone.

“If we make it easy for customers to find answers online, where they’re already looking for the phone number, then they won’t have to wait on hold to get an answer, which should raise the customer satisfaction rates.”

Experiment

So we build a site with that hypothesis as the driving factor. We make it easy at all steps to get the questions answered without having to take the extra effort of calling. This starting point drives everything, from the site structure, to the font selected, to what the site looks like on different platforms. Every decision we make is driven by our hypothesis and the goal of increasing customer satisfaction.

Analysis

And then we start to measure and analyze. To do that well, we ask questions during the design stage to make sure we’re working toward concrete goals. Will we be successful if we see a decrease in call volumes, or do we measure success by the number of people accessing the site vs. the number of calls that are coming in?

Conclusion

In this step, we review all of the information that we gathered in our analysis, and discover if our hypothesis was correct, and use everything we learn to tweak the site as necessary.

As with so much of life, there are no definite answers, no “if I do x, then y will happen.” Like in quantum mechanics, x can have a number of unexpected outcomes, and there can be multiple causes for y. There is never one single right or wrong step. This is why it is important to do as much research as we can, and instead of dwelling on our missteps, learn from them and iterate accordingly.

And finally, remember that in web design, as in science, it is important to celebrate your successes and learn from your failures.