🧠 How To Reduce Cognitive Load In UX

People are complex. We are often impulsive, and typically juggling between dozens of things at the same time. We follow hunches. We trust their own beliefs and experiences. And often what they say, do, think and believe are all very different things. But ultimately we need to make 1000s of decisions — often very difficult ones, in very challenging environments, with a lot of complexity and a myriad of dependencies that we can't control.

How can we, as designers, then help other people make better, accurate decisions? How can we prevent mistakes and errors efficiently? Well, typically by creating clarity, order and structure around their work, and reducing what we call "cognitive load" — the amount of mental resources needed to understand and interact with a product.

In this post, let’s explore how people make choices and what we can do to make our products slightly less demanding — dismantling common UX myths along the way.

Meet Design Patterns For AI 🔮 and How To Measure UX 🍎 — friendly UX courses with practical techniques for product designers and UX leads. Use code 🎟 SNOWFLAKE to save 20% off today.

Islands Of Clarity In UX

People don’t always use our products the way we imagined they would. In fact, it’s common to see people using only small portions of a complex product frequently — almost identifying small islands of clarity that help them in their daily work, while avoiding obscure or daunting parts of the product because they haven’t managed to learn how to use them yet — and because they don't necessarily need them.

And sometimes people don't use of a product simply because they are just to difficult to use, or the "cognitive load" required to complete a task is too much. In other words, when the amount of information coming in exceeds the space we have to process it — that's when tasks become more difficult, details are missed, and we begin to feel overwhelmed.

Conservation of Complexity

According to Tesler's Law, complex systems always preserve their complexity. We can’t really remove complexity as it must match the circumstances of daily life; it has to be dealt with. The question is, how much technical burden we put on the user, and how much on supporting the user in their complex daily tasks.

While the dental software above might look a bit daunting at first, we shouldn't underestimate how quickly people adapt their habits around the tools they use daily. Often people can't choose the tools they use for work, and so over time they stop noticing the UI entirely. It almost becomes a second nature.

The question isn't how daunting it looks to us — and we are not the users — but rather how error-prone or reliable the system is in daily use. Judging such systems by the cover is often a wrong path; instead, figuring out where people get stuck and what their issues are is more helpful.

In complex environments, successful organizations put the burden of complexity on technologies: i.e. standardized, optimized and automated processes. If a product can reduce wasted time, errors, workarounds and improve efficiency, it can have a significant impact at scale. Unsurprisingly, the path there typically lies through better UX.

Every Product Is Like A City

When we look at complex applications, it might feel like the best way to improve efficiency is by "cleaning up" the UI — with all its weird inconsistencies, confusing layout, poor typesetting and ambiguous iconography. And it's indeed a good direction to explore, but personally I wouldn't actually start with it.

Instead, I want to understand how people who rely on this tool actually use it. What might appear like inconsistency to me might be merely the result of complexity leaking through the UI — e.g. when lengthy titles break the layout, or when a product has to support multiple user groups with very different user needs, or when there are legal or security considerations that have shaped the product to become what it is.

I would often start by asking the same 6 questions:

1. What's your working environment like (e.g. how many screens, surroundings, mobile, desktop, stationery or on the go)?
2. What other products or services do you use together with our product?
3. Could you guide me on how you use the product frequently at your work?
4. Could you explain the product and its most useful features to me?
5. What are most frequent tasks that you are performing in this product?
6. What do you find most confusing and broken in the product?

Every product is like a city. It's a complex, intertwined set of requirements, infrastructure, unpredictable behaviors, shortcuts, snowflake scenarios and everything in-between. Consistency might help in some way, but it might also break a neatly orchestrated way of working to map and manage the complexity of the work.

The best way to optimize a product is by looking at the most frequent tasks across user groups and optimize them. Similarly, to improve the quality of a city, we can how people spend their time, and where the bottlenecks are — to explore opportunities for improvement.

Five Pillars Of Reducing Cognitive Load

When we want to reduce cognitive load, there are a few key aspects we need to consider: most importantly choice, organization, and working memory.

1. Beware of Oversimplification

People don't necessarily struggle with complex environments. Life is incredibly complex, and we must navigate all its confusing and contradictory parts every single day. People struggle with products and features that they don’t understand or can't figure out. And they also learn products and features over time, making tremendous progress in just a few days. Complex doesn't mean unmanageble.

In fact, expert software must be complex as it must match the complexity of the real world. It requires a vast number of attributes, settings, views, panels, data points. However, complex products don’t have to be complicated in use if they make sense to end users and they can be proficient with them. There, the worst thing we can do as designers is to oversimplify.

2. Make Choices More Manageable

People need to make a huge number of choices every day, and they have become very efficient at scanning, skimming, and skipping to reach their goals more quickly. However, oftentimes the differences between choices aren't obvious — especially in light of consequences, dependencies and their advantages and disadvantages. People need support to make choices more manageable for them, and we can help them there.

To help people manage choices, it’s always critical to avoid ambiguity and explain differences between directions. This means chunking: we break options into distinct, scannable groups. Maybe we could highlight or pre-select options that fits a vast majority of use cases in similar situations.

We could also use color coding and indicate what requires attention, or rank options based on data, or label options as "Frequently chosen" or "Recommended" or "Featured items" (in that category). It’s worth noting that experts often prefer to see many options or features at once, while non-expert users often rely on small “islands” that they use regularly.

3. Create Order Around A Task

People can’t really multi-task, especially in very stressful situations or emergencies — and especially with a big chunk of work in front of them, people need some order to make progress, reliably. That’s why more simpler pages often work better than one big complex page.

People are making better progress when they only have a small set of options to choose from. This holds true even if the amount of options hasn't changed, but the number of displayed options at a time has. Similar options can confuse users, making them feel frustrated or paralyzed.

Contrary to the often propagated “3-clicks-rule,” people typically don’t mind an extra click if it’s clear and predictable — and as long as it’s not repetitive or slows them down. The number of clicks/taps is a poor indicator of good or bad UX.

Order means giving users a clear plan of action to complete a task. No distractions, no unnecessary navigation. We ask simple questions and prompt simple actions, one after another, one thing at a time. An example of the plan is the Task List Pattern, invented by fine folks at Gov.uk. We break a task into a sequence of sub-tasks, describe them with actionable labels, assign statuses, and track progress.

To support accuracy, we revise default settings, values, presets, and actions. Also, the order of actions and buttons matters, so we put high-priority things first to make them easier to find. Then we add built-in safeguards (e.g., Undo feature) to prevent irreversible errors.

4. Support Working Memory

Surely some people are forgetful, especially when they move between tabs and tasks, but also when they are disrupted by a phone call or an urgent email. Yet most of the time it's not forgetfulness that is problematic, but the amount of detail that the product expects them to keep in their working memory — and it's usually very limited.

As Raluca Budiu noted, if a task incurs a high cognitive load, it usually means that it puts a high burden on the working memory. At this point, usually the famous “7±2 rule” (also known as Miller's rule) appears. This research by George Miller dates back to 1958 and relates to short-term memory’s approximate capacity.

Research shows that broad and shallow menu structures may even work better than deeper menus. Also, link-rich e-commerce homepages, like that of Amazon with 100+ product category links, are found to be more usable than homepages with only a few links. What matters more is how many of those items a user has to keep in their working memory.

It's better to not force people to remember things. Yet we do it unconsciously when modals block access to content needed for comparison (see above), or when similar records are scattered across different tabs or pages.

Instead, we can allow people to favorite, store and access critical details that are needed during a task — perhaps in a sidebar, or in a separate workspace (external memory). Sliding panels and overlays might not work as well compared to showing content in persistent side panels instead.

Over the years, we've learned how to better map users’ mental model. We can run card sorting on features, filters, attributes, and menu items. We can run tree testing on navigation. We break down complex decisions into a set of smaller decisions. The same goes for flows: flows with more pages might work better than one single page.

5. Just Enough Friction

Not all cognitive load is harmful — it's when there is too much of it that it becomes hard to manage. As Krystal Higgins points out, if there is not enough friction when onboarding new users and the experience is too passive or users are hand-held even through the most basic tasks, you risk that they won’t realize the personal value they gain from the experience and, ultimately, lose interest.

Good onboarding requires just the right amount of work from users to bring them to their first success moments in the product. At best, it should feel like a series of little achievements that lead to a bigger achievement, rather than just a series of meaningless clicks to just get through onboarding tutorials.

Friction has some value. It’s when people start paying attention. When they put in some effort to achieve a meaningful success. To do that, learning-by-doing with guided interaction — instead of onboarding visuals as tutorials — will work best.

Wrapping Up

There are no universal templates or established shortcuts we can follow to reduce cognitive load. Some applications are complex and needs a healthy dose of cognitive load. So beware of oversimplified UX rules that promise quick solutions. They might limit your design decisions and prevent you from making the right design choices for your users.

As always, the first step towards reducing cognitive load is to understand users' tasks and their understanding of your product — along with their expectations, goals and needs. Once you have these insights, you can start exploring ways to make their experience less difficult to manage and a little bit more straightforward.

Useful Resources

• UX Myths (+ Free PDF Poster), by Zoltan Gocza , Zoltan Kollin
• Laws of UX, by Jon Yablonski
• Minimize Cognitive Load to Maximize Usability, by Kathryn Whitenton
• Designing For Stress and Emergencies, by yours truly
• How To Reduce Cognitive Demand In UX, by Alexander Benz
• De-Stressing Design, by Lucy Scott
• Eight Design Tips to Reduce Cognitive Load, by UXcel

🥁 Just Released: Design Patterns For AI

Meet Design Patterns For AI Interfaces 🔮 — a brand new video course (10h) with me on design patterns for AI products, with 100s of real-life examples, live sessions, practical techniques and everything designers need to know to design AI experiences that people understand, use, value and trust. Ah, use the coupon code 🎟 NOVEMBER to save 20% off today. Jump to details.

Thank you so much for your support, everyone — and (as always!) happy designing! 🎉🥳