1. Set your prototype test goals

When running a prototype test, an important first step is to set your end goal. What are you looking to test and discover about your product? Answering this question will help ensure you get the results you need to take action at the end of the testing period.

Be specific with your goals, for example:

• Instead of “I want to test my prototype,” ask “I want to find out if users can find information about a home loan and fill in and submit an application form.”

• Instead of “Is the product engaging?” ask, “Does the prototype influence users to engage further and purchase a streaming subscription?”

• Instead of “Do users like the design?” ask, “Can users interact with the design and navigate to book a walking tour?”

Remember, prototype testing is iterative. You want to be continually testing and gathering feedback throughout the design process. So it’s important to be clear about what you want to discover at each stage.

2. Choose the level of fidelity

The prototype you create and test will depend on your goals and the stage you’re at. 

The word prototype might suggest a model of the finished product, but in reality, they don’t need to be perfect. In fact, there are different levels of fidelity, which we cover below:

Low- and medium-fidelity prototypes

The earlier you can test your prototypes, the better. Low-fidelity prototype testing happens in the earlier stages of the design process with a paper or digital prototype. Use them to generate and share feedback and ideas with stakeholders so you can quickly iterate your designs.

What are you testing at this stage?

• Whether the layout makes sense to your users.

• Experiments in your initial designs.

• The information architecture.

• Basic interactions with the design.

As you take user feedback onboard and improve your design, retest new prototypes using a medium-fidelity prototype. At this stage, you can also add copy.

High-fidelity prototypes

Once you’ve tested and taken in user feedback on your low- and medium-fidelity prototypes, you can move onto creating a high-fidelity prototype. 

This is interactive and closely resembles the final product.

At this stage, you shouldn’t expect huge problems to emerge. Use a high-fidelity prototype as a last step in your design testing process to validate your final design and uncover any hidden usability issues before handing over to the development team.

What are you testing at this stage?

• The overall design direction.

• The copy, menu links, and information architecture.

• User flows: Can users find their way when completing a task?

• UI components, e.g. drop-down menus, pop-ups.

• Graphic and visual elements, e.g. image quality, text readability.

Live data prototypes

A live data prototype uses data or APIs to create dynamic experiences. It’s much like the final product, so you might use it if you wanted to test software solutions in real-life environments and conditions.

Building a live data prototype is more involved than a low- or high-fidelity prototype, as you’ll need a developer. But it can be a useful way to learn how your product performs with users in the wild and generate data about how the final product will work. And if the test goes well, you’re already on your way to production-quality software.

Feasibility prototypes

A feasibility prototype focuses on a key feature, technology, or component of the product. You might opt for this type of prototype if you wanted to test something considered high-risk, like a specific feature innovation.

The aim with a feasibility prototype is to establish evidence that the feature, technology, or component will work. You might use it in the early stages of the development process in order to assess and reduce future risks.

Before creating and testing your prototype

You’ll need to decide which tools you’ll use. Again, this will depend on the type of prototype you’re going for. For example, if you create a digital prototype in Figma, you can then embed the Flows directly in your Lyssna test and run a remote unmoderated test with real users.

In the Lyssna test builder, you can paste or type the prototype Flow link for the prototype you’d like to test.

3. Choose a user testing method

The usability testing technique you choose is how you’ll gather user feedback.

Will you choose a remote or in-person test? Moderated or unmoderated? Again, this all depends on your goals, what stage you’re at, and the tools you’re using, along with other considerations, like resourcing, time, and budget.

Remember that tests can be a combination of these methods, such as unmoderated remote tests.

4. Create a prototype test

The purpose here is to give users a specific task to perform related to the problem your product or website is trying to solve.

Here are some top tips when creating your test:

• Set a clear task for your users: It might help to create a story that places them in a realistic scenario. This puts the focus on the user’s goals, not on the product’s features and functionality. For example, if you’re designing a clothing resale marketplace, you might test whether users can set up a profile, choose a clothing category for their item, upload photos of their item, add a description, and set a price.

• Ask specific questions: This is your chance to ask usability questions, so be sure to write questions that will give you useful feedback. 

• Assure participants they can’t be wrong: Be clear to your participants that the test isn’t about them, it’s about testing the functional design. So they don’t need to worry about being wrong.

• If testing in-person, be open, engaged, and neutral: If a participant knows you created the prototype, they might avoid offering critical remarks. So try to avoid emotional language to encourage honest feedback.

5. Recruit participants 

In the early design stages, you might test with colleagues in your organization, or with a select group of valued customers. You can also run a pilot test with friends or non-design colleagues to feel confident before testing externally.

Ideally, you want your test participants to represent your target group of users. This might include current customers and those who haven’t used your product before so you can get well-rounded feedback. For example, if you’re designing a recipe organizing app, it makes sense to do app prototype testing with people who like to cook rather than those who use a food delivery service.

If your product will have a global audience, aim to test it with diverse and global audiences. Different cultures and customs might affect how your users use your product.

6. Evaluate and share test results

By now, you’ll likely have a mix of qualitative and quantitative data. Next, you’ll need to evaluate and share the results with your team and your stakeholders.

Prepare a clear report so everyone agrees and understands the next steps.

Whether or not the feedback is positive, it should give you a sign of what needs improving. For example, using our recipe organizing app, if users found the grocery list feature difficult, consider ways to help unblock them.

After making your changes, it’s time to repeat the process with a new prototype. This process continues until you’re satisfied with the prototype and no further changes are needed.

Viewing your prototype test results in Lyssna.

Top tips for prototype testing

You might be tired of us saying this by now, but UX design is iterative. So test your prototypes early, and test often.

To wrap up, here are our top tips for prototype testing:

• Begin prototype testing early, after the preliminary design stage. This will help you identify and solve problems before you get too far.

• Prototypes don’t have to be perfect. They can be low-fidelity or high-fidelity.

• Set clearly defined goals.

• Choose a testing method that best suits your goals.

• Define your target audience and recruit participants representative of your target audience.

• Set clear tasks for your participants.

• Evaluate the results and implement changes.

• Keep testing. It’s a great way to gather feedback at different stages of the design process.