Based on the tenets of product-market fit, you’re more likely to achieve success with a minimal product with errors that has product-market fit, than with a great product experience that doesn’t land with your customers. That said, while product-market fit is not the same thing as great product usability (and it’s easy to confuse the two), poor usability can impact how your users use and perceive your product, and therefore, the likelihood of product success. Enter: usability testing, also known as user testing. In this guide, you'll learn: When you should conduct usability, or user, testing What to look for when conducting usability testing The steps of successful user testing When to conduct usability testingUsability testing is done prior to the launch of your product as well as anytime you’re developing a new feature. During and post-launch, testing is usually referred to as QA (quality assurance) to test live bugs and errors with a different yet overlapping purpose: to create the best product experience.Usability testing occurs on live prototypes and MVPs of a product, yes, but many companies conduct usability testing prior to any product development through design prototypes using tools like Figma and with the expertise of a UX designer. Even scrappier versions include paper prototypes and someone asking questions about how the user/tester is feeling and what they’re thinking as they walk through scenarios and tasks as if they’re on an actual website or product.Testing on a design prototype can help work out some kinks and issues before development begins, saving time and money by addressing them earlier on: IBM conducted research that showed that user testing only after a product has already been released costs 30 times more than if the errors had been discovered during the design phase.What to look for during user testingAside from signs of product-market fit, when conducting usability testing you should be honing in on the actual behavior of users and their experience of your product.Here’s what you should be on the lookout for: User satisfaction - how satisfied are they overall? Ease of use - how easy is it to complete a task/action? Errors - what’s preventing them from completing a task/action? Time on task - how long does it take to complete a task/action? However, on a design prototype, you’re less likely to be testing for things like errors and more likely to be testing for the user’s understanding of how to take desired actions and how they move through your product. This is helpful when you’re considering different designs, for example, and would like some insight before moving forward.With that, let’s go through how to conduct usability testing.Steps to user testingDefine goalsBefore you start, define your goals. What are you testing and why? What’s the desired outcome? If you have many assumptions to test, you may want to set up different tests. For example, if you’re testing customer segments, you’ll want to run through the same test with different groups of users. Or, if you’re testing for purchase conversion, you may want to test a different user flow from if you’re testing for signups. Take a look at the assumptions you have for your product to determine what needs to be tested. Hire testers You’ll need to find people to test your product. Make sure you recruit participants that are representative of the people who fit into your ideal customer profile. This could include people you’ve already been in contact with who've expressed interest in your product—in other words, your potential customers. In this case, you should provide an incentive for them to provide early feedback such as a gift card, swag, cash, early access, or a combination of these. Your testers could also include people hired from user testing platforms such as: UserTesting Userbrain Userfeel User Interviews  TestingTime UserZoom Some of these platforms even provide a full suite of services to help you conduct your user testing, for a fee.How many testers is a large enough sample size? According to the Nielsen Norman Group, you should have a minimum of five participants. Plan your test Decide where the test will take place. You may be able to observe behavior more accurately and closely by conducting a test live and in-person, but remote user testing is increasingly common. Write a script based on your test goals. You can veer off script but having one makes sure you cover everything you need to. Define success criteria and metrics. If you’ve conducted previous tests or have existing data, you can use these as a benchmark for measuring success based on any changes you’ve made. Prepare your participants. Introduce yourself, build a bit of rapport, and let them know if you’ll be recording them and/or taking notes. ExecuteTime to conduct your test! Here are are few tips: Make sure your participants are comfortable and at ease; you’ll receive more accurate feedback. Encourage participants to think out loud and talk through their actions and feelings. Avoid influencing the test by asking leading questions that hint at desired actions or behaviors. If your participant gets stuck or silent, you can jump in with a non-biased approach like “How might you..” or “How are you feeling?”  You’ll want to make sure you can capture behavior both on-screen and by observing your participant. If you’re in-person, that’s easy enough. But if you’re remote, you can use tools such as Rayfeed. AnalyzeWhat does your usability test tell you about your product? Let’s analyze. Here’s where recordings and notes come in handy. Write down any general observations as well as observations and insights based on your specific goals. Organize the information using a spreadsheet or table, then sort based on urgency and impact. You can rank the insights in order to understand priority. Once you’re done, apply your findings and if necessary, repeat the process all over again.  The data you get from usability testing is just one factor of many pieces of data you’ll need to look at. But as the bridge between a concept and a real, live product, it’s one of the most important in getting you one step closer to product-market fit.