One Magic Formula to Calculate UX?

tl;dr → user experience = (purpose*interaction) / velocity

My quest for a formula to calculate User Experience started with a little experiment to see if I could calculate cognitive loading time of a design. When talking about loading speed of digital products the focus always seem to be the digital loading time, but we sometimes forget that trying to understand an interface also requires time – cognitive loading time. The more I thought and researched this problem, the more I discovered that loading time – digital and cognitive – might just be a small part of a bigger and more important question: Is it possible to quantify User Experience?

The problem with User Experience is that it is — just like Psychology — quite vague. Many people, including myself, talk about improving User Experience and what factors might affect it, but it seems impossible to quantify or calculate UX (I will call it UX from now since I will be using this term a lot in this post).

Of course, there are ways to discover how your product is experienced by asking your users with a questionnaire or interview, but the results are qualitative. There’s nothing wrong with qualitative data or users telling you that your product experience sucks or is brilliant, but you can’t trust people’s opinions. What people tell you is just a version of the truth and always influenced by external factors. Cognitive biases prevent users from accurately reporting their own experiences with your product.

Wouldn’t it be awesome if we could figure out a formula to calculate what the UX of your product is?

Wouldn’t it be awesome if we could figure out a formula to calculate the success of the UX of your product? Not saying you should stop talking to your users, but having a formula with hard data would make the UX field less fuzzy. The thought alone of creating this magic formula to calculate the success of a design and product made me glow in the dark, hence I gave it try!

Break it down

The first step we have to take is breaking down the UX into smaller measurable chunks. What facets influence the UX of a product? This goes beyond designing an understandable interface for your product, there are many factors at play that influence your users.

Digital loading speed

v digital = v internet + v server + v device + v code

Digital loading time is easy to measure. We can all agree that loading time has a huge effect on the UX of a digital product. If software is slow users become cranky, and if the speed is too slow they may abandon your product altogether. Google found that by serving a 400ms slower version of their website, people were less likely to use it.

The digital speed of your product can be broken down even further into smaller pieces: internet speed, device speed, server speed and product code speed. For our formula I will use the letter v for velocity.

v internet 

The speed of the internet connection is which is out of our hands, but has a huge impact on how your product is experienced. Something to consider here is whether the user thinks any slow speed issue is your products fault or a weak internet signal. You can help frame the user’s thoughts in a number of ways, including choosing the right loading animation.

v device 

Your users device speed is out of your hands, although you could choose to make your application only available for newer devices; for instance, to avoid users with older slower devices that would experience a slow product. Of course, this raises the question of whether it’s better to offer no product than a slow product.

v server 

The speed of your server is something that I (as a designer) don’t know much about, but my co-founder & buddy Petar Radošević assures me that your server choice will affect the application and database speed, which in turn affects the overall speed of your product. Therefore, we can’t ignore the server speed in calculating the overall speed of a product.

v code 

The last part of digital loading speed is the loading speed of the product code. Having worked with Joeri and Petar for many years now, I know how much effort is required to make queries as fast as possible. So don’t just lazily loop through your string quick and dirty (this sounds worse than I intended), put an effort into writing beautiful efficient code.

Cognitive loading speed

v cognitive = ic / (n * dx) 

The cognitive loading speed of the product is where this all started. I find it a particularly interesting subject because it is such an important determinant in the realisation of a great user experience. It is an area where design meets psychology, where flat digital pixels become more than the sum of its part because the human brain is able to translate it to a meaningful interface.

For the ones asking themselves “what the f*ck is cognitive load Mr. de Bres?”, don’t worry, it’s a fairly simple concept, especially for you, since you probably work a lot with computers.

The cognitive load theory was first mentioned by G.A. Miller in the 1950s; he mentions in a paper that the working memory of humans seems to have an inherent limit. He states the working memory seems to be capable of only holding 7 (±2) chunks of knowledge at a time, being numbers, words or visual elements. The working memory of our brain works exactly the same as the working memory of computers. It processes all information inputted. Dr. Graham Cooper created this explanatory image in his 1998 research paper that makes it understandable:

ic = interface complexity 

Interface complexity plays a huge factor in cognitive load. If your product has a complex interface with many elements then it has a higher cognitive load, meaning it takes longer for the user to comprehend, resulting in a less happy user. When a user has the feeling they don’t understand what they see a feeling of discomfort rises and self-esteem plummets. I think we can all agree that’s not a great user experience. That’s why creating simple interfaces is not just a hipster designer thing, it is essential to make your users feel they are in control. This results in a happy confident user able to make quick decisions.

n = number of visits
dx = digital experience

Although complexity of an interface is the key determinant in the cognitive loading speed of an interface, two other factors play a big role in counterbalancing this factor; n, the times a user has successfully used your product in the past and dx, the general digital experience level a user has with digital products like yours. When a product has a complex interface, but the user has previous experience with this or similar products, the cognitive load can still be low. If there is a lot of previous experience with the product, the complexity might not even matter. This can result in  the fact that a product with a complex interface and experienced users can be more effective than redesigning a product to use a simple interface that sets all legacy user understanding back to zero.

Having read the above we can calculate the total speed of the product by adding the digital speed and the cognitive speed.

v digital = v internet + v server + v device + v code
v cognitive = ic / (n * dx)
v = v digital + v cognitive

Great we now determined what the total speed of the product is. But, having a fast product is only a small part of a great user experience. This is where calculating UX gets tricky and I may need your collective brain to help me out.

Purpose

p = b + s

As I mentioned in a previous blog post, the fundamental element of a solid user experience is purpose (p). A digital product must have a purpose to provide value. Without a purpose there will never be a user to experience the user experience at all. A products purpose can differ on a per user basis, but it always comes down to two factors:

b = benefit

The most important thing for a user to use a particular piece of software is that they believe there is a benefit in using it. This can be acquiring information, social interaction, (digital) goods or just good old fun.

s = status 

A factor that can’t be ignored since we are dealing with human behavior here is status. Status can be a strong force that drives people to take action. If a product has the power to increase someone’s social status then it boosts the intrinsic motivation of your users.

Interaction

i = e + c 

The part where most designers focus when trying to improve the UX of a product is the interaction (i). This is the easiest factor to improve as a designer. The problem is that because of the Dribbblisation of design young designers think the esthetics are the sole factor in creating a good UX.

e = esthetics 

I have a hard time admitting this, but the esthetics (e) of a product are only a small part of a great UX. Designers — including myself — like to overestimate the influence of their designs on the complete UX and overall success of a product. As you probably realise by reading this article, there are so many factors determining a successful UX, that creating a great UX is a team sport, not something owned by just the designer.

c = communication

Last but not least is communication. By communication I mean everything from the copy used in your interface to the emails from customer support. The way you communicate with users should be human, kind and understanding, with a dash of humor. Every human being likes to be treated nicely, your product is your voice, so use it wisely and kindly.

Designers like to overestimate the influence of esthetics on the complete UX and overall success of a product.

The magic formula

If you are still with me, congratulations for getting this far and thank you! As you have read above we now have all the building blocks that play a part in a great UX. If I missed anything — which I inevitably did — please let me know on Twitter @WdeB. So, is this the moment we have all been waiting for? Can we create one magic formula out of the above to calculate the User Experience of a digital product? Let’s give it a try. Mathematicians and physicists, please look the other way.

v digital = v internet + v server + v device + v code
v cognitive = ic / (n * dx)
v = v digital + v cognitive
p = b + s
i = e + c

ux = (p*i) / v

user experience = (purpose*interaction) / velocity

Conclusion

Have we now cracked the UX code? Can we now stop talking to our users and just calculate this magic number to see if our UX is brilliant? No, absolutely not. First, the purpose of the exercise was to uncover the various factors that play a role in User Experience. Second, too many of the factors mentioned above are almost impossible to quantify, which makes the formula difficult to use.

The bottom line is that a great UX is not just a beautiful interface, it is part of a bigger equation in which not just the designer is responsible but everyone in the company, from the backend engineer to customer support. Maybe this formula can help us determine where the strong points are in a product and in which areas it has to improve. But as I mentioned in an earlier blog post, purpose is the key to a great UX. Loading speed, communication, and aesthetics can all be improved fairly easy, but purpose, that’s the core of your product.

I would love to hear what you think of this little thought experiment. If you have any additions for the formula or if you have any other feedback, please let me know on Twitter @WdeB.

Thank you for making it to the bottom of this post. If you want to hear more of my ramblings, or read my previous post The 10 Commandments of Digital Product Design.

Wouter de Bres

Wouter de Bres@WdeB
Psychologist turned Designer & Co-founder gibbon.co