A quick guide to usability testing

Dan Tase
4 min readOct 24, 2016


Even if you’re a small and scrappy startup or a billion dollar company, usability testing is helping you evaluate your product based on feedback from real users, with little or no resources. It’s the easiest way to spot problems early on or to validate ideas before going to production.

In this small article I’m going to cover the basic aspects of usability testing, from recruitment to analysing results.

How it’s done

A session usually consists in observing a number of people use your product under controlled circumstances. This can be done either by giving them a detailed task or by guiding them through the process in order to achieve the desired outcome.

Depending on the type of test you’re doing, the number of users can vary, but Nielsen Norman Group identified that 5 is the ideal for the maximum benefit-cost ratio. That being said, it’s important to test the feature in each stage, from early concept to final prototype, as every small change can affect the user’s behaviour.

Step 1: Prototype

So you’ve designed an awesome feature and you want to make sure it’s working according to your plan. One of the key aspects is to create a prototype in order to simulate how it’s gonna work on the live platform.

Don’t worry, it requires no coding. Actually, coding the feature is not ideal, as the purpose of the session is to understand how/if users are engaging with the feature. There are awesome tools like Marvel that can help you quickly create interactive prototypes, just by importing your Sketch or Photoshop file.

Step 2: Recruitment

Recruiting the right people is another essential aspect. In order to get useful feedback, users have to fit within your desired persona, so creating a recruitment criteria is gonna help you decide on the right ones. A good example might include:

  • Personal details: Name, language, profession, age, gender, hobbies.
  • Tech experience: iPhone user / Likes to try new products, etc
  • Behaviour: Buys something online every week.
  • Exclusion criteria: Does not work in Sales
  • Approval to be filmed. This is important, as the last step is to analyse the results.

See who ticks all the boxes, schedule an hour with each one of them and you’re ready for the next step.

Step 3: Session

There are no clear rules on how to run a usability testing session, but there are some things you might want to follow in order to achieve better results.

  1. Preparation: Set up the room to create a safe environment and make sure there are no distractions. Don’t forget to record the whole session (prototype and user) so that you can analyse the video at a later stage. You can also invite stakeholders to watch the session from a different room, helping you by taking notes.
  2. Introduction script: It’s really important for users to feel comfortable, same as in their natural environment. State that the sole purpose of the session is to see how real people would use the feature, encouraging them to think out loud.
  3. Questionnaire: Before the session kicks off you might want to get a better understanding of the user’s context or opinions regarding a certain aspect. This can easily be done with a simple, non formal questionnaire, disguised as a conversation.
  4. Task: Most of the times a session starts by giving users a detailed task and observing their behaviour. At times you might have to guide the user in the right direction if the conversation slips. Don’t forget to observe verbal cues as well as non-verbal ones (facial expressions, posture, etc). It is estimated that 80% of what we communicate comes from these signals.

And now… time for the next step.

Step 4: Analysing results

Stay strong, you’re almost there. The following step is to gather the results by watching each recording, studying all the cues and taking notes of all the potential issues, while grouping them by Critical, Serious and Minor.

Now create a stakeholders deck, add the summary, testing, results and recommendations, and you’re all set to go. Of course, analysing the results is no piece of cake, but every researcher has their own way of dealing with it, so the results depend on your level of expertise and (why not?) patience.

I hope you find this helpful.


I’m Dan, Product Designer living and working in sunny London. If you enjoy what you’re reading don’t hesitate to contact me. I’m also on Twitter.



Dan Tase

Founder & Lead Product Designer at Rubber Studio. Previously led design at Farfetch, Just Eat & Fresha. Slightly dyslexic.