Don't Make Me Think
- Jacob Rodriguez
- Mar 15, 2024
- 5 min read
Steve Krug’s Don’t Make Me Think Revisited gives a “common sense approach” to UX for web and mobile. The big idea is that good design should not require users to think. I read the third edition, the latest as of now, which was published all the way back in 2014. Its age shows with the old UI screen captures and photos of chunky iPhones with home buttons. Despite that, the principles it discusses hold up. The book is about how people use tech; luckily, people have not changed much in the past 10 years.
Krug’s definition of usability:
A person of average (or even below average) ability and experience can figure out how to use the thing [i.e., it’s learnable] to accomplish something [effective] without it being more trouble than it’s worth [efficient].
In addition to the three attributes italicized above, he cites delight and memorability as another two things that good design should have. Delight is if the application is fun to interact with and memorable is if controls for the application can be recalled later.
To prevent users from thinking about the design of an application, it should use conventions. While SpaceX might think it’s cute to replace the submit button with a button that says, “Blast Off,” this could confuse users and lessen their experience. Users should not be questioning an application’s design for even a fraction of a second. If something doesn’t have a clear convention and cannot be self-evident then it should at least be self-explanatory. Things like size, color, text, and layout can be signalers for a function.
The reason that sites need to be so obvious and simple to use is because of the way people look at digital content. Krug compares it to driving past a billboard. Screens are scanned, not read. I remember Nicholas Carr citing a study in The Shallows that found that people look at webpages in an F pattern. They are looking for the features that are relevant to them, tuning out the fluff. Users will typically choose the first reasonable option they see in a practice called satisficing. If a platform is not designed with this in mind, a user will end up making a lot of interactions they don’t need to, continuously backtracking. Krug claims the back button is the most used button on the internet.
How do you design something that is not only common sense but needs to be processed in almost no time at all? By using conventions and visual hierarchies, breaking up content, making clickable objects apparent, eliminating distractions, and good formatting. These conventions are all over the internet now. When the book was originally published in 2000, not so much. It’s obvious when a convention isn’t used because it makes it harder to use an application. Conventions should only be broken when a designer knows they can do it better. Better, in this case, means more clearly.
Design helps a user navigate through an application. Just the way streets have signs, design should let users know where they are. Good navigation should: tell users what’s on the site, tell them how to use the site, help them find what they need, and tell them where they are. The last one is especially useful for users who didn’t enter the site through the homepage. Navigation should pass the trunk test. The trunk test takes any screen and asks the designer to identify the site, page name, site sections, options, etc. can be identified. The name comes from the idea of waking up in a trunk and using your surroundings to find out what’s going on.
Like all things that are good in theory, good design is hard to implement in the real world. Shocker. One way Krug touches on this is with the homepage and the tragedy of the commons. Because items on the homepage get the most interactions, groups will want their work featured there. This can lead to teams shoving all their projects to the homepage until it becomes hard to use for users. Krug uses university sites as an example of this where they feature a lot of material that visitors don’t care about, but faculty and staff do. This prioritization of things not desired by features is a result of opinions driving decisions.
Usability testing also costs time, money, and resources that all departments think are better spent on their objectives. Usability testing may be hard to justify to the decision-makers. Different roles also have different goals. “Correct” design’s benefits may not seem apparent or worth the cost. Teaching staff UX is expensive. According to his website, Krug’s all-day remote workshops for companies are $15K!
Krug tries to solve the problems companies have with usability testing by offering a method called do-it-yourself usability testing. This method costs hundreds of dollars instead of thousands. It works by, throughout the development process, recruiting about three testers once a month to perform tasks on an application. After they perform tasks, stakeholders deliberate which issues found are the most severe which will determine what gets priority over the next month. Scrum. What differentiates this method from standard testing is the limited testing pool, shortened testing period, increased frequency, and accelerated analysis. This method also doesn’t require a testing site and instead is performed in a convenient neutral environment.
It’s important not to confuse usability testing with focus groups. Focus groups are more for determining the validity of something with a target market whereas usability testing is to see what issues a person has with operating a product.
Krug also states the importance of accessible and ethical design. Accessibility is making sure that those with disabilities can effectively use a platform. This is harder to do for newer companies with limited resources and know-how, however, it’s important nonetheless. Ethical considerations must be made when it comes to using design to influence users into taking actions they do not want.
Krug wrote this book for everyone. Product managers, engineers, marketers, and of course those working in UX. This makes it very easy to grasp the material and apply it on a basic level. As I mentioned earlier, it still holds up despite its age. Some things are dated. Mobile browsing has come a loooooong way and so has desktop browsing. Design has evolved. Krug could not have foreseen the sleek and sometimes convoluted modern UI/UX design. I’m curious as to what he thinks of modern design and how it challenges some conventions.
This book gave me some ideas for some of my designs. I know this blog breaks a few of Krug’s rules and I intend to fix some of those infractions when I get the chance. To be fair, I think I’m the only person who cares about this blog (or even reads it) so I don’t think it’s of the utmost importance yet. This text will be a good guide to help me revise Nectar if I find the time to as well.
I really enjoyed the cartoons used to illustrate points. I felt like I was reading a children’s book for business workers. It wasn’t until after finishing the book that I found out his name is pronounced “kroog” and not “krug.” He also mentioned The Design of Everyday Things which I took as a sign that I’m on the right track for picking reading material. He uses the “wrong” definition of affordances, which he acknowledges, which I thought was funny because of what a big deal Norman made of it in his book. He also gave a shoutout to UserTesting.com where I unintentionally conducted a somewhat successful do-it-yourself usability test while developing GnoME.
Two videos he mentioned that I liked. I might steal these for a presentation at some point.
Comments