🔻Why do we need to test our designs?
Testing is like the holy grail of the UX process. You work so hard on a project, it goes untested and then developed and delivered the end user. But what happens if you didn’t help solving their ‘real’ problems? Moreover, if it’s a new product, what if your users don’t get it? Usability testing will help you find these answers. Also it will help to increase the conversion rates, customer retention, decrease customer support costs and re-development costs. Testing early and often will cut a lot of unwanted costs and a lot of companies could benefit from this. 💰
It is like, you try out the trial version before subscription.
🔻 What does testing help us with?
Testing helps us to get to the WHY. Why did it go wrong? Why didn’t the user see it? Why did the user find it difficult to understand? You can answer a lot of these questions by testing.
🔻Who are involved in testing?
👩 Facilitator — the specialist who conducts the UT
👨Observer — the specialist who observes the UT
👲Participants — the users in a selected user group
🔻 What testing are we talking about here?
The kind of user testing we do at our studio is called formative testing. That is, we start testing a product while its development is going on or before it is sent to any development.
We do this to see if our designs do what we intended them to do. Basically, to see if a design objective is being met.
The approach we take when doing a UT is small scale. Here we use small sample sizes. This will get us many opportunities for observation. Small scale UTs are based on expertise and judgement since the facilitator will be conducting the UT face to face, one on one with the participant. However small scale UTs are not quantitative. Meaning, we are not doing these to find statistics or measurements. This is only to find issues in the designs and to gather insights. How we find these insights are by interpreting the participants’ language, body language and emotions.
🔻How do I prepare for a Usability Testing (UT)?
1️⃣ Understand which part of your product you are going to test
Maybe you have designed a new feature of your product, or you have redesigned your product, etc. First, you need to finalize which parts of your product are going to be tested.
Example: I am going to test a login screen.
2️⃣ Pick your participants
You need to decide what sort of users are needed to test this prototype and how much. If you are testing a feature or a small part of the design, 5 users would be enough. They will help you to find 80% of the problems. This also depends on the criticality of the product. From my experience, the sample size varies between 5–15 depending on the product and the areas that we are testing.
It would be better if you can do iterative testing as well. That is, you test with 5 users, fix the issues and test with 5 users again. It goes on like that. But the product time lines should be kept in mind if you want to do that, otherwise this will not be a realistic goal.
The user profiles will help you select participants for the testing. Data that is gathered through market research, or other data that is collected within the company will play a big role at this point.
To recruit participants, you need to create a screener. Screener will include the user groups, number of the participants and their expected characteristics. The screener will then be sent to a recruiting agency, where they will recruit the participants, schedule the appointments and let you know. Or you can do the recruitments in-house. That will cost you less but will be time consuming.
Example: I want users who are between the ages of 50–60,who use an android phone.
3️⃣ Prepare the testing protocol
Simply said, the testing protocol is a document that will guide you through the testing process.
Here you have to create scenarios or tasks for the user to perform. These are for the features/parts that you decided to test.
You could be a little descriptive while telling about the tasks/scenarios to the user. Be unambiguous and provide the details in need for the related tasks.
Example: For the same login screen, you will need to provide the username and the password. Username is [email protected] and password is ‘hannah123’
4️⃣ Get the prototype ready
It’s the prototype that you are testing since the real product is not being built yet. You will have to create your prototypes to suit the scenarios that you created in the testing protocol. If it doesn’t match the expected outcome, the participants will get confused and deviate from the task at hand.
Example: If you want your participant to enter ‘[email protected]’ as the username in the login page, and what’s displayed is ‘[email protected]’ in the prototype, it will confuse the participant and cause a disorientation.
5️⃣ Have a look at the venue
Usually the normal setup will look like an FBI interrogation room. But not so hostile. There will be furniture. And some snacks and beverages on the table. You have to have a consent form for the participant to sign.
The room will have a one-way mirror for the observer to observe. A camcorder focusing on the prototype to record and for it to be displayed on the TV for the observer.
🔻During the UT
Be polite and welcoming to the participants. If they seem nervous, be friendly and try to talk with them a bit before starting the session. Before you start any session, explain to the participants why you are doing this and what for. It’s important to be transparent.
Ask the participants to ‘think aloud’. It will help both the facilitator and the observer to understand the participant more. Don’t use the words that are on screen while describing a task/scenario. Try to get to user’s terminology and use them instead. Also, before a user completes the task, don’t give away the answer. You can guide them if needed but hand holding is not recommended. Be kind and polite. Avoid stating things like ‘yes! That’s how you do it’ or ‘nope. You didn’t get it right’. Remember that, users are never wrong. They don’t get it because there is a flaw in the design. And we are here to find out why.
Ask more open-ended questions. And they will try to explain you. If you didn’t get to the exact reason, probe till you figure out why. We are doing this not to get opinions, but to understand the big fat WHY.
🔻After the UT
After every session, have a quick chat with the observer to recap what happened during the session. This will help to clarify things that are missed or misunderstood.
After you complete all the sessions, you need to gather all the insights to iterate your designs. The data logging sheet which the observer was working on and other screen recordings will be of help.
🔻Checklist before testing
✅ Set up – camcorder with an inbuilt microphone, TV of the observer’s room, snacks and beverages for the participant, venue
✅ Testing protocol – scenarios and supportive items. eg :phone numbers, passwords
✅ Prototype – all set for the scenarios
✅ Incentives – for the participants
✅ Consent forms – for the participants
✅ Pen – to sign the consent forms
✅ Screen recorder – to record the flow of the prototype
✅ Data logging sheet – for the observer to take notes
If it’s your first UT, it would give you heebie jeebies for sure. So, try to relax as much as possible, talk to your observer about how things would go. You can avoid a lot of this anxiety by doing a pilot test before the real one. Ask one or two of your peers to help you with it. It would be better if they had previous experience so that they can give you feedback then and there.
Happy testing! 💁You’ll love it, as much as I do.
I hope you gathered some ideas about usability testing with this article. Don’t forget to show your support by clapping! 👏