I have voiced the possibility that doing testing of designs or prototypes on real people to get feedback at early stages in development and continually through production could improve the developed products.

I have been given the go ahead to look into this and our options. So I'm after some pointers from some people already involved in working on or with usability tests.

I need some guidance on the first steps to take (from a designer's perspective) to initiate this change.

  • First types of tests I could carry out?
  • Who needs to be on board initially i.e. can it be done independently without support from other staff members or managers in the very early stages?

Where to start?

  • Get who ever is in charge of the project on board - and preferably the CEO...
    – PhillipW
    Jun 13, 2016 at 19:15

2 Answers 2


There are several ways of doing user testing. My preferred way is with the help of a ux tool. I am personally using loop11 for my user testing but you can do it with a pen and a paper. The reason I like to use tools is because this way i focus on creating meaningful scenarios that can really help me test my prototypes.It helps me quickly make specific design decisions based on customer recommendations and am able to iterate quickly. Have a panel of testers which you can use at different stages of your project. Create persona. Then choose between moderated or unmoderated tests. Usually, it depends on your specific goals and what you need to accomplish. A moderated test is the most controlled type of usability testing, and is recommended for testing specific use cases, like whether or not the user can find/access a certain feature (or how long it takes to do so). Scripted tests tend to produce more quantitative data, beneficial for usability metrics, but can also generate qualitative data as well, depending on the how tight or controlling the script is. Unmoderated testing benefits include: - time savings - mote natural product usage - cost savings - simpler coordination


You can start with a moderated usability test of a prototype. There are steps which will help you to organize your first simple usability testing session.

1. Research

First of all, you need to know what do you want to test. You don't need a test just for test. You need to test design assumptions, e.g. would users understand what that label means or that this section is scrollable. So, write down the list of design assumptions.

2. Plan

The next step is a testing script. It could be a simple word document which contains: the lists of assumptions and participants, intro (brief) and a list of tasks.

List of assumptions - done

List of participants - You need to know who is your user. This step might include creating personas and recruiting process. The question is about conducting a usability test in general, so I won't go deep. Here you need to list the names of actual participants of the upcoming testing session. You need at least from 3 to 5 participants, they will detect the 80% of the most crucial issues.

Intro - Your introduction for participants. It should be helper text for yourself to not forget anything. The key points which you should mention are that you are not testing the user and that you want to know honestly what they think: e.g. we are testing an application, not you, we want to hear exactly what you think, you can't do anything wrong here etc.

Tasks - You should prepare a list of ~7-10 real scenario tasks which will test your design assumptions. Each task might have the following structure in the document: task (what you have to do), actions (what this user need to do to accomplish this task, what you think he will do), response (how the system/prototype will respond to user's actions).

Here is an example:

6. Approve the current image
  a) Click on approve_image button
    - Current image is marked as approved
    - Next image is unlocked

3. Preparation

Schedule a meeting for each participant. Usually, it should take ~30 minutes per user. It can be done solo, but it's always good to have a second person who will be taking notes. It might be one of the stakeholders, e.g. project manager. Prepare an excel file for the results. There should be presented tasks, participants, success rate, satisfaction rate, identified issues.

An example for identified issues: task (# of task), description (description of an issue) and occurred (how many times this issue occurred in this session).

4. Test

Find a quite place to run a prototype test. This could be paper prototype, static HTML web page, InvisionApp's prototype etc. Start with your introduction to explain what are you going to, that you want the participant to think out loud, that you will answer their questions in the end because you want to know how people do when they don’t have someone sitting next to them. Ask if they have any questions. You might want to sign some kind of NDA agreement with the participants

After the introduction, show to the user initial state of the system you're going to test. You may ask first what the user think it is, on what he would click first. Then start with the tasks. Always make notes, you may also make an audio or video record of a testing session.

5. Analyze

After testing session create a report document. List there again the participants and the tasks. Then write down the results from your excel file. You should highlight the errors which appeared most of the times or which stopped the user from accomplishing the task. Write down a list of key takeaways as well as notable quotes from users, e.g. “So it shows me ‘swipe right to done’ okay let’s swipe it (swiping in the middle of the screen)” Then, after a brainstorm session, the output should be the recommendations for improvements to designs.

Hope this will help and good luck!

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.