How to Bring UX into the Modern World
Starting the Conversation of Adaptive UX
Are you that child in the family?
You know the one. The person who gets the three sequential 10am Thursday morning frantic phone calls in the middle of your meeting, but you have to call her back because it’s your mom and three calls in a row means it must be important. So you hand off the meeting and step out of the room while profusely apologizing to your annoyed coworkers only to find out that the reason she’s calling is because she thinks she deleted that email with “the link thing” to the photos from her vacation last summer. See, she wanted to show her friends at brunch and she just saw the email but she went to click the download button and “something happened” and now she can’t find it anywhere…
Yeah… me neither. Just kidding, I’m am so that child. (Love you, Mom)
If you are as well, I will be holding support groups under the guise of MeetUps in a city near you!
So I proceed to explain to her that she should first not click anything, and second look at the bottom left of the screen.
Me: “Is there a dark rectangle with a blue “Undo” in it?”
Mom: “Oh, yes, there is. Do you want me to click that?”
Me: “Yes, that will bring the email back to your inbox. Go ahead and click it.”
Mom: “Awesome, it worked! Thank you so much dear, I hope you weren’t busy.”
Me: “Never too busy for you, Mom. If that happens again, you can just search for words you know are in that email in the search bar and it should also come up.”
Am I clairvoyant? Nope (I’ve been tested). I just know she is using the Gmail account I set her up with. I know she most likely archived the email on accident and mistook Gmail’s archive icon for a download icon. And I know she doesn’t intuitively know to hover and confirm the tooltip descriptor before she clicks the icon. I’m sure you also know things about your loved ones that helps you solve their problems like how your color-blind friend struggles with reds and greens. Or how your uncle has bad eyesight and buys large print books because they’re easier to read. Or how your sister is a self-proclaimed “visual learner” and refuses to read any direction longer than three words. In this article I am going to discuss what I personally think should be the new industry standard for User Experience (UX), a method that would utilize all of these individual nuances in order to help your mom, friend, uncle, and sister navigate their technology easier in accordance to them.
UX Currently
The fact is that we, as UX professionals, claim that we are designing for the singular user experience but we aren’t. We are designing for someone who doesn’t exist and this realization feels as wrong as the plural use of data (I get the argument but “the data are” will never feel right). We are meeting individuals, or observing them, and collecting very personal information (just go down the rabbit hole of finding your frequent locations in your iPhone). Yet we take those personal details and throw them into a huge pool of insights, slap an over-generalized classification on them as a whole, and only revisit once a massive trend emerges from their similar demographic. Tell me, if I say “plug-in,” what concept comes to mind? Do you first think of outlet, cord, or Chrome? Understanding your end user is an elemental part of communication, and while you may be a white 29-year-old female of Irish decent,that doesn’t mean you first thought Chrome, like I did.
In 1983, a psychologist named Dedre Getner (now a professor in the Department of Psychology at Northwestern) published a research paper titled Structure Mapping: A Theoretical Framework for Analogy, in which she established a method of structure mapping from the base of an analogy to the target in order to communicate meaning beyond the literal words used. In that cornerstone paper, cited over 6,200 times according to Google search results, Getner breaches the topic of domain knowledge. In order to successfully map the intended meaning from an analogy (in the case of user interfaces, think visual analogies like paper filling a recycling bin when a file is deleted) the user needs to understand the relationship you are attempting to portray. In other words, the user needs to have sufficient domain knowledge of what you’re referencing in order for the analogy to work. If I personally have never known what a floppy disk was or how to identify it, *cough* Gen-Z *cough*, then how am I intuitively supposed to know that icon means save?
In a similar mindset, how is my mom supposed to know that, in modern UI systems, the download button is usually just an arrow pointing down and maybe a line below that, or even what “archive” means in the Gmail setting?
With respect to Getner’s research, the designers of Google Mail are faced with a structure mapping dilemma. I would put money on them knowing that some people don’t have that domain knowledge, but. alas, they must build one product to serve their entire target population. So what do they do? They hope to train, try to include happy-mediums, but not too much information, and they iterate. Here is where I take issue. That used to be the reasonable best practice. Not anymore.
We need to evolve the product development process to include adaptive user interfaces (AUIs) and UX systems based off of a user’s cognitive model, behaviors, domain knowledge, and context.
For further comprehension, AUIs “are defined as systems that adapt their displays and available actions to the user’s current goals and abilities by monitoring user status, the system state and current situation,” (Rothrock et al., 2002). That’s right, people; that definition was published in 2002!
So why the lag? Simply put, building an adaptive system that helps users when they need it and leaves them alone when they don’t takes a lot of information that we didn’t have readily accessible or obtainable in 2002. Wasn’t that the era of T-9 texting after all and arely three years after the first Blackberry came out? Think of where we are today compared to then. The extent of data we collect is so elaborate that the concept of user data has penetrated non-industry related news cycles. People may not realize that we can track their movement by tracking the cell towers their phone tries to connect to, but they sure as heck know that something's weird when they stub their toe and immediately see an ad for bandaids (exaggeration of course…. Or is it? Looking at you, Alexa). They caught on, and why shouldn’t they? Facebook can collect data on you even if you don’t have an account, but you need a treasure map and a lexicon to figure out how to turn off notifications about people you haven’t seen in years (Wagner, 2018)?
It’s one-sided and unfair.
As a professional who is supposed to fight for the user, and as a citizen of the afflicted population, I move that we put in more effort to truly make each user’s life easier and the experiences we design better. I’m talking about using all of Toni’s data points to help serve Toni just as much as our own business goals instead of throwing all users into the current system just to be watered down with the rest, resulting in five personas that are supposed to represent the 95 million user base…and Toni.
Now, take a second and hear me when I say there are a lot of things to consider when we shift the industry towards this smarter way of designing. There is such thing as over personalization (Flaherty and Moran, 2019). There are times where adaptive interfaces are more helpful and times where it is better to leave one standard interface for the masses (Lavie and Meyer, 2010). There are ethical issues. There are different implementations (my thought is to have a data bank and handle it like a payment processor but more to come on that). There are still obstacles when collecting data stemming from things like hardware. There are the help forums to consider. What happens when I can’t blindly direct my mom (Akiki, 2017)? What does that look like? I could go on, and I plan to… in my next article. However this one is to get the conversation started.
So how about it? Is your interest piqued?
Our developers are going to loveee the technical logic write-ups that this will bring.
Citations
Akiki, Pierre. (2017). CHAIN: Developing Model-Driven Contextual Help for Adaptive User Interfaces. Journal of Systems and Software. 135. 10.1016/j.jss.2017.10.017.
Flaherty, Kim & Moran, Kate. (2019). “The Dangers of Overpersonalization.” Nielsen Norman Group. https://www.nngroup.com/articles/overpersonalization/
Gentner, D. (1983). Structure-mapping: A theoretical framework for analogy. Cognitive Science, 7(2), 155-170. http://dx.doi.org/10.1207/s15516709cog0702_3
Lavie, Talia & Meyer, Joachim. (2010). Benefits and costs of adaptive user interfaces. International Journal of Human-Computer Studies. 68. 508-524. 10.1016/j.ijhcs.2010.01.004.
Recycle bin image from ComputerHope.com
Rothrock, Ling & Koubek, Richard & Fuchs, Frederic & Haas, Michael & Salvendy, Gavriel. (2002). Review and reappraisal of adaptive interfaces: Toward biologically inspired paradigms. Theoretical Issues in Ergonomics Science. 3. 47-84. 10.1080/14639220110110342.
Wagner, Kurt. (2018). “This is how Facebook collects data on you even if you don’t have an account: There’s little you can do about it.” Recode, Vox Media. https://www.recode.net/2018/4/20/17254312/facebook-shadow-profiles-data-collection-non-users-mark-zuckerberg