I’ve spoken about the importance of user research in a product development cycle in my previous blog. What comes after a solid user research is the phase of prototyping and testing, and these are just as critical as user research. Research, Prototyping and, Usability Testing form the salt and spices of a well balanced dish (yes, we care a lot about food). A dish can be cooked without salt or spice, but you won’t have too many people liking it. The story is very similar in the case of product development too. A product built on assumptions and poor testing will eventually find itself difficult to accommodate the needs of its users and will need more fixing,which translates to a lot of wasted time.
A Wireframe is not a prototype.
While it’s very easy to get confused between a wireframe and a prototype, they’re quite different. A wireframe is a rough visualization of the proof of concept. They are stuff that we sketch on paper or on tools like Balsamiq that gives a rough idea of what an interface looks like. However, the scope of a wireframe ends there. A collection of wireframes that are related to each other and help a user complete a task flow qualifies as a prototype. Mockups are similar to wireframes too, just that their fidelity (level of detail) would be different.
What is prototyping?
Prototyping is a quick an easy way to validate our ideas based on the research information we have. It is a proof of concept and not the actual product itself. And it should not be the actual product. Prototyping is done in the very early stages of ideation to understand what works for users and what doesn’t in order to quickly iterate based on the learnings from these tests. An important thing to note about prototyping is that ‘Prototypes are messy’. And that is how they should be. When thinking of testing, it’s very easy to confuse a prototype for a high-fidelity mockup. They’re two different entities altogether. While usability testing is done on both the interfaces, the purpose and the goals of the testing are very different. With a low fidelity prototype we test if users can figure out their way through the task and reach their goal in the expected way. A high fidelity mockup test would reveal if people can find the right buttons or if they like the colors. Understanding what we’re testing is very important to get the right data.
Usability testing is an amalgamation of art and science. It is more than just watching users complete a task using a prototype or product. The real purpose of conducting a usability testing is not to see if a user can complete the task, but why they’re doing what they’re doing. It’s not a test of the users but of their interaction with the system. If you read the previous sentence clearly, I defined two important facets of design: Interaction Design and Usability Testing. Interaction design is all about creating a dialogue between a user and a system while Usability testing is about understand how the dialogue takes place. While conducting a usability test, it’s all about reading between the lines.
While usability testing is a very detailed subject in itself, I will give a quick list of the most important things to consider before and while conducting a usability test.
- When testing with a low fidelity prototype (A set of wireframes), we should make it clear to the participants that they’re not testing an actual product and it is just a proof of concept built to validate the design decisions. Using the findings of the usability test we would build a high fidelity prototype/mockup.
- Recruit wisely. Do not get people who’d be biased and who are involved in the working of the project. Getting unbiased and candid feedback is the most effective way to build a solid product. Also, do not test with more than 6-8 users per iteration. That’s a good enough number to uncover most usability issues in the prototype.
- Explain the purpose of the test to the participants clearly. When they know why they’re doing something, they’ll have more ownership on their actions and they’ll be more involved in the test, which will give us better findings.
- Remind the users that they’re not being tested, but the prototype and whatever they do are not any of their mistakes. This helps them be more candid with their opinion.
- Always ask participants to ‘think out loud’. Think-out-loud is a beautiful process to understand the psychology of our users. Asking them to speak out what they’re thinking lets us know their thought process which is very important in understanding their mental model.
- Another thing to remember is that people do not remember instructions very well. They tend to get distracted and veer away from the task or may stop thinking out loud. As moderators, we’ve to be aware of their actions and give subtle reminders to think out loud or take another look at the task list.
- If they’re completing a specific task, hand them a clear task list that tells them what they’re supposed to achieve. Also, do not be too specific with the tasks. For example, instead of telling ‘Click on the back’ button’, tell ‘do task x’. Let them figure out where to click. Also, having a clear task list will help participants stay on the task and not get distracted and lost in the middle of a task.
- Before every new task, ask them to read out the task loud. This will keep the task more firmly in their mind.
- Avoid helping users or giving them leading hints. Telling them ‘Ok you can press that button now’ is a direct hint and when people are testing, they WILL more often than not, click it even if they didn’t want to. As tempting as it is to help people when they seems lost, it’s important to know that they do not get such help when they’re using the product on their own. Let participants explore and figure their way out all the while making notes of what’s going on.
- If a person cannot find something and asks you what to do next, do not tell them what to do, rather respond back with a question that helps you learn more. For example, when they cannot find a back button to a webpage, get stuck and ask for help, ask them what they’d usually expect in such a scenario, what they want to do and if they were to design the webpage, where they’d put the back button. It’s amazing how many things we think of as obvious are a total mystery to other people. And sometimes people give the best suggestions from their experience.
- When you find that they’re completely lost and cannot find their way around, give them a couple of minutes to see if they can recover from the error. If they still cannot, give a subtle hint to see if they can find their way from there. If that doesn’t work as well, note that as a major usability issue, ask them what they’d expect and move them to the next task. At any point, Do not make them feel like they’re dumb.
- It’s very important to constantly ask them to explain their actions and decisions. But their explanations must be taken with a pinch of salt. There is a difference between what people do and what they tell. This is why every usability test must have a well triangulated recording process. Usually usability tests will have a moderator, a note taker, screen and audio/video recording tools to correlate data.
- Give close attention to their expressions. People are usually quite expressive and usually give out subtle feedback using non-verbal hints. For example a soft ‘hmmmm!’ to express surprise or a quick ‘oh shit!’ are all very important hints that something they didn’t expect happened. It’s important to note these moments and ask them with a follow-up question ‘please tell me why you said ‘oh shit!’.
- The purpose of a usability test is not always to find faults in a system. If they like something, it is just as important to ask them why they liked it. Even if they do not tell, as moderators, we must make sure to find out the positives of the design. Knowing what worked will give designers a direction to go forward and iterate the design.
- Start and end the usability test with a short interview. Before the test ask users about their experience using similar products, their frustrations, likes and their expectations. End it by asking their experience using the prototype, what they liked or disliked about it, what was most challenging and what they’d change if they could. Understanding their thought process is the goal of a usability test.
Knowing these points above is very important for a successful prototype design and testing. While it appears simple, its quite hard to stay focused while creating prototypes and testing with users. It’s very easy to slip into the user mentality and when we lose focus, we end up losing a lot of data and waste considerable amount of time both for us and the participants.
Prototyping and Usability studies are the equivalent to a physical structure’s foundation. While it may take some extra time and effort, going through these processes gives a product a very solid backbone which can later be used to build upon during successive iterations. This avoids the problem of building a product and later discovering it doesn’t work for its user. This is the reason why companies like Facebook, Google and Twitter put such heavy emphasis on research and design. They understand their users’ needs, goals and frustrations before designing a product and that’s what makes a cult-following billion dollar product.
(image credit: usabilitygeek.com)