Studies show that a great user experience can boost conversion rates up to 400%, which business owners can't afford to ignore. One of the best ways to determine what pleases and what bothers the users is to get them involved by running usability tests.
This usability testing guide will tell you exactly how to plan and execute effective user testing. The goal is not just to get data and user opinions – we want data you can immediately put to use and improve the most critical aspects of your business.
That will directly turn to better user experience and satisfaction, indirectly leading to more sales and better customer lifetime value.
First thing first: Determine the test scope
Before you do anything, you need to know precisely what you want to accomplish with user testing. While every design and usability project is different, here are some ideas to start:
What do you want to test?
You need to have something specific you want to test with real users. This can be your new app, online store or a complete website redesign.
Whenever you aren't sure about a change you want to make with your product, usability testing might give you an answer. You can also test only a section of your product or website, which will give you more precise insights.
What’s the goal?
While uncovering usability issues and friction points is something that every user test will help with, it's a better idea to have a more precise goal.
For example, even though you are testing your whole sales process, you can primarily look into improving one aspect of it, such as upsells. You can do this by testing various recommended products that go along with the one a customer just bought.
To eliminate guesswork, It would be a good idea to ask users about their choices after they complete the purchase, not just to count how many of them added another product to the cart. Asking will help you identify the reasons behind purchase decisions, which will help offer better products in the “related” section.
Which type of user should perform the test?
This will heavily depend on the product or feature you want to test. If you’re checking something more general, anyone can give you information.
But, if you’re offering a more "niche" service or product, it makes no sense to ask random people for their opinion. For instance, if you just created a new SEO plugin for WordPress, people who already use Yoast are an excellent choice for testing, as they already have an idea of how a good SEO plugin should work. But someone who strictly does user interface design coding may not even know the first thing about SEO plugins.
Which type of test should you run?
Answering the questions above will help you decide what to test, how to identify a specific goal and who should be in the test group. With those items out of the way, you only need to determine which type of test you want to run
For commodity-type physical products, random users and "guerilla testing" can work. You can have a stand and a promoter asking people passing by to try your product or test your app on a tablet computer and answer some quick questions.
If your product has a narrower or strictly-digital audience, performing online tests can work best. You can contact your existing user base or find potential customers on LinkedIn and ask them if they’re willing to participate in the test. This is an excellent option as it's easy to organize and get a lot of feedback.
Lastly, you can carefully pick test subjects and organize what’s known as moderated usability testing. Here, you will interact with people while they interact with your products. This will give you the best answers as you will exactly see how much effort each user puts into the testing.
This can be done in person or online. It will also allow you to uncover sticking points quickly.
How to write a usability test script
A usability test script will help users relax and understand the goal of the test, which will also allow them to focus on the purpose of the test, helping them give you better quality feedback.
It will also ensure all team members involved in testing do a good job and correctly guide the users through the process. Here's how to create a good usability test.
Add a brief introduction
This section should be designed to break the ice. Briefly introduce your team, and tell users more about the company. If you're testing person-to-person, this is a great opportunity to have some small talk to build rapport.
Also, tell them all the practical information, such as how long the test will take if it's going to be recorded and how you will use the collected information.
You should also check your local privacy laws and ensure you’re not infringing on their rights. In general, getting written consent and using the information exactly the way they authorized you won't get you in trouble.
Get familiar with the user
Even if you’ve carefully selected users and pre-screened them, asking them some basic questions will help break the ice even further, especially if you’re doing the test live.
Ask them about their background – where they live, their current position, education, age, previous experience with testing or using other products of your company, or competitors'.
Make sure you only ask questions that are relevant for the test and let them know that all you want is to collect valuable information about your product and not the users themselves.
Pick an easy task first
Give users something simple to get started. This should be something quick like completing a "show me around" procedure. Doing something simple will help them relax and boost their confidence, helping them go through the more challenging parts later.
Set a scenario
A scenario will give users context, which will make the task more interesting and give you more insights later, as the user will understand what they need to accomplish and why.
The scenario doesn't have to be complicated – it can be something as simple as telling users they need to use your website to find a school backpack for their children, a girl and a boy.
This will make them go through your store categories, find backpacks, select children's models, pick a color and finalize the checkout process. Through that, you’ll see how the whole buyer's journey goes and if there's a phase you can improve.
Take it step by step
In most cases, users should face only a single task at a time. This means they will never get overwhelmed, but, more importantly, you’ll know exactly which task causes friction, especially if you’re doing remote/online testing where you can't interact and see their behavior.
If you have the option, separate groups can test different parts of your product. Or you can ask users new questions when they face the same task – this might be a better approach than facing a single user with multiple tasks and questions simultaneously.
Also, ensure users go through a flow they typically would while using a live product. If it's a store app, it should start with logging in, then continuing through the menus, categories, products, and checkout at the end. Don't jump from one point to the next, and don't break the flow if it’s not fully necessary.
Don't be too helpful
While you want to ensure your users don't feel lost, giving too detailed instructions is something you should never do. Every question you ask should be unbiased, and if possible, open-ended.
That way, users will still need to explore and figure out things independently (like they would in the real world)
If you ask questions narrowing down users' answers, you’ll practically make user testing useless. On the other hand, if you don't ask questions that are precise enough, you will get answers that are too vague, and the participants might lose interest.
Here are some unbiased questions you can use after the test is done:
What part of the product did you find the most useful?
If you could change one thing, what would it be?
What do you think about the XYZ feature?
What suggestions do you have to improve XYZ's part of the process?
On the other hand, a biased question would be something similar to "How entertaining did you find the first level?" This question suggests that the user should feel entertained and would skew responses in that direction.
Keep it short
Lastly, never overwhelm users. Depending on the test difficulty, each user should face 5-8 tasks, at the most, especially if you are doing the testing online where the bounce rate is higher.
Remember, users are basically doing you a favor here, as most of the time, they don't get anything in return for their feedback. Therefore, the test should never take too long and be too difficult.
What's too long and too difficult? That depends on the test subjects and goals. It's very important that the users know in advance what they will face and that the test doesn't take longer than you informed them. That way, they are less likely to find the test too long, too dull, or too complicated, making it less likely for them to quit.
Usability testing will give you valuable insights and rapidly uncover bugs and friction points, helping you improve the product you are working on or test ideas you haven't yet launched.
They are one of the best ways to cut the total web app development cost as they can often help identify minor issues in the early stages before they compound into massive problems later.
Because user testing is flexible and adaptable to any industry and every type of product, and at the same time ultra-effective and affordable, you should utilize it to learn more about your brand and improve it further.