One of the first things people do when they start using a product is spend time learning the interface. This process can be good or bad, fun or annoying. If a product is straightforward on the first try, first-time users will likely return. If they have to guess how certain features work or where to find an individual option, they would be hard-pressed to describe the experience as a good one.
To improve the chances that users get what they want out of your product, it’s essential to invest time in measuring its learnability. In this article, we will discuss the concept of learnability and explore how learnability testing can maximize the usability—and, by proxy, the appeal—of your app design.
What is learnability?
Learnability is the quality that allows users to quickly become familiar with and able to use the features and capabilities of a product. “Learnability” is often erroneously used interchangeably with the term “usability”; while they are similar concepts, learnability (as well as efficiency, memorability, and satisfaction) is actually a category of user testing.
Why does learnability matter?
Good learnability is directly relevant to usability. The less time and effort users have to invest in learning how to use a product, the more money you save on training costs. This can be especially important for products that require a certain amount of training, such as CRM products or industry-specific software. Additionally, good learnability makes users feel confident in their abilities and increases the likelihood that they will be satisfied with a product.
How is learnability measured?
When UX practitioners measure learnability, they track how easy it is for users to accomplish a task the first time they encounter the interface and how many repetitions it takes for them to become efficient at that task. The power law of practice proves that the improvement isn’t linear, but logarithmic. That’s why UX practitioners use a learning curve, a 2D chart that has the number of repetitions (trials) on the X-axis and efficiency on the Y-axis, as a tool to measure human behavior.
A learning curve reveals how long it takes for users to learn the interface and achieve maximum efficiency on specific tasks. Image credit NNGroup.Similar to any other usability testing, to measure learnability, you define metrics first. Usually, it’s enough to rely on the “time on task” metric, which, as the name implies, tells you how much time users spend on the task. By tracking this metric, you will see how fast users are learning: The easier it is to learn how to use a product, the less time it takes for a user to do a specific task without prior training.
Three facets of learnability
Alita Joyce, a user experience specialist for NNGroup, identifies three aspects of learnability that are important to different kinds of users. These are important to keep in mind when testing learnability. Ideally, a product should demonstrate excellent results in all three aspects:
- First-use learnability. How easy is it to use the product the first-time users try it? This is important for users who plan to use a product only once.
- Steepness of the learning curve. How quickly do users get better with repeated use of the design? This is important for users who want to feel a sense of progression with the interaction (getting better at using a product).
- Efficiency of the interactions. How high is the productivity that users can reach once they have fully learned how to use the product? This aspect is vital for users who will use the product frequently.
What makes an interface learnable?
Good learnability is the result of three things: good information architecture, good navigation systems, and good individual pages/screens. Here are three elements that influence learnability.
Mental models
Highly learnable products are intuitive—their users can navigate and complete certain operations with ease. Such products are created in accordance with user mental models, which represent an individual’s understanding of how something works. Designers who leverage mental models of their users make it easier for them to discover and learn features and functionality. For example, when users interact with a website, they expect to see high-level navigation at the top of the page. Navigation placed in another location on the page makes it less discoverable, and, thus, less learnable.
Content and feature priority
The size of an individual object influences the way users perceive it. Larger objects are likely to be located first and more easily found again. That’s why consideration should be given to the relative importance of tasks and features; put more visual weight on objects that are connected with those tasks. Things that most people do, most often, should be prioritized first.
Key navigation controls are clearly visible in the Google Map interface. This design decision aids learnability. Image credit Google Maps.Simplicity
How many features does your product have? The more features you have within your interface, the more complicated it will look, and the more time users will need to invest in learning its features. Also, an interface with dozens of different features is often perceived as difficult to learn. This perception can influence a user’s motivation to start the process of learning in the first place. Generally, users are less motivated to learn something that looks complex.
How is UI tested for learnability?
Quantitative usability testing is well-suited for studying product learnability. Typical learnability testing can look like this:
1 – Hire test participants who represent your target audience. These participants should have little to no experience using the product that’s being tested for its learnability. Participants who have prior experience using a product might cause biased user testing results.
2 – Ask test participants to complete a set of tasks using the product. A typical task usually represents a real interaction scenario. For example, test participants are asked to purchase a product on an ecommerce website and the amount of time it takes them to complete the task for the first time is measured.
3 – Ask test participants to do the same set of tasks a second time. Each participant should repeat the X tasks Y times. Similar to the first trial, measure the task completion time.
4 – Repeat the process several more times. The results will show a learning curve that plots the task time over a set number of trials. Repeat the trials until you see the same time completion for two trials. A flat curve indicates participants have learned the product (specific to this task) as much as possible.
What are learnability testing best practices?
Learning is a sum of understanding and remembering. A few factors have a tremendous impact on the learnability of a system:
- Understanding (ease of function learning). How long does the user take to learn how to use a function?
- Efficiency. How long does the user take to learn how to perform the specified task efficiently?
- Accessibility. How well-suited is the system design for the needs of various groups of users?
These factors inform the following best practices for making your UI more learnable for users.
Do learnability testing of the product’s core features first
Every product has a core set of features that deliver the maximum value for your users and your business. It’s vital to ensure that those features can be learned with ease, so they should be tested first.
Compare the learnability of different versions of your design
It’s possible to analyze the learnability of a few different versions of your interface for the same task. For example, you have two versions of a UI that look good, but the team is unsure which version will work better for users. By creating and analyzing learning curves for both versions of the UI, you will identify the winner.
Do not measure learnability for one-time tasks
Measuring learnability for tasks that users complete one time (such as signing up for a service) doesn’t provide value. In real life, even when users complete the task again, they will most likely behave like new users.
Recruit a large number of test participants
Since learnability testing is quantitative testing, you need a large number of test participants to get statistically significant results. NNGroup recommends hiring at least 30 to 40 test participants, but the exact number will depend on the complexity of your tasks (complex tasks will require more test participants to account for the variability of the results).
Define the number of trials
Trials are repeated uses. In most cases, task performance improves after repeated trials (more practice results in less time needed to complete tasks). But how many repetitions does it take for test participants to become efficient at that task? In many cases, you want to aim for 5 to 10 trials; after users become familiar with the interface, their trials will give you data on the fastest possible task completion time.
Identify how much time you need between the trials
Anticipate the timeframe your typical user will spend interacting with the product and re-create that interval as closely as possible in your testing. For example, if your target audience interacts with your product once a month, you should select 30 days as an interval for your testing and repeat testing with participants in 30 days.
Pair learnability testing with other types of usability testing
Usability testing methods will help you to determine how easy (or difficult) it is for users to interact with a product. Robust usability testing will help you identify further issues. You need to understand how users navigate in your product, track the success rate for common tasks, and identify elements that cause friction.
Always balance learnability and efficiency
Good learnability doesn’t guarantee good user experience. The product may be highly learnable but completely inefficient. The easiest solution is not necessarily the best solution for users long-term.
Imagine a step-by-step wizard that you use to complete a certain operation in a finance app (e.g., sending money). The system puts you in a funnel and guides you along the way, helping you perform the task as fast as possible right from the first trial. At the same time, the wizard might be designed in a way that doesn’t allow for shortcuts; this might cause a lot of unnecessary interaction cost (extra clicks or taps). After a few trials, regular users will be reluctant to interact with the app because they have to go through so many unnecessary steps.
This example demonstrates why designers should always carefully balance learnability and efficiency. It’s always worth implementing shortcuts for expert users to give them the power to complete regular operations with less effort.
Conclusion
When you design a new product or redesign an existing one, you should invest time in learning how fast users can interact with your interface. The learning curve of your product will impact how well it performs on the market. Your goal is to minimize the number of trials required to become an experienced user, without sacrificing product efficiency.