Skip to main content

Usability testing and focus groups

Quinten Steenhuis, updated August 2022

Why test?

The word "testing" can encompass many overlapping goals, including:

  • Discovery - learning what product to create, and what features it should have
  • Identifying usability problems in your tool
  • Verifying the correctness of your tool

Forms of testing

Focus groups

A focus group typically includes:

  • A moderator
  • A group of 3-5 representative potential users of the product you intend to build
  • A series of questions designed to learn what the users value and desire, and their particular needs
  • A session that can be between 1 and 2 hours long

Focus groups are most effective early in your product design process, before you have actual wireframes or code that they can interact with. Once you have a real product, one on one usability tests may be more effective.

Before holding a focus group, come up with a few key goals that you hope it will achieve. Then, think of 10-20 questions, with follow-ups, that help explore your big themes.

You should expect to pay focus group participants. Typical honorariums range from $25 to $200.

It's a good idea to record your focus group sessions, but you should also take notes. If you have multiple observers, have each observer write down 3 key takeaways. After the focus group session is complete, compare notes and discuss the main points you can use to drive your product's direction.

Without focus groups, it can be hard to know if the product that you are planning to build will actually meet the needs of your users. But it's important to remember: you are asking your users to guess what they want, but they may not be able to accurately state what that is.

Usability tests

Usability tests also involve talking about your product with real users. They are not the same as focus groups. Usability tests:

  • Involve a single user at a time
  • Require a series of tasks that you will observe the user attempt, not questions for them to answer
  • Involve a moderator who will be in the "room" (camera on) with the user
  • Should have multiple observers who can be camera off, and will be looking for the key usability problems that the test surfaces

Usability tests are critical to producing a product that your user can actually use without difficulty. If you don't have an actual product to test yet, you can actually run usability tests on your competitor's products! You can learn a lot about how to design your tool "right" by seeing what your competitor or benchmark site does "wrong."

When designing a user test, your first goal is to come up with a series of tasks. Tasks should be somewhat open-ended. Assigning an open-ended task to guided interviews can be tricky as guided interviews tend to be linear, but you can always start the task on the landing page. Even very simple guided interviews do tend to involve some branching logic, however. And you can also assign tasks such as changing an answer, downloading a form, or viewing instructions that require the user to select among many possible menu items visible on the screen.

For example: get a child support plan as part of your restraining order.

Start off the usability test by reassuring the user that you are testing the product, not the user. Warm the user up with a few general questions. Ask for their general impressions of the site, and then move into observing the user attempting your tasks.

This script from the US government agency 18F is a great place to start.

Some other key things to remember during the usability test:

  • Have your user narrate what they are thinking out loud.
  • While observing your user, be careful not to offer help. You can answer questions at the end.

Below is a video demonstrating a model usability test from start to finish:

You should expect to pay usability test participants. Typical honorariums range from $25 to $200.

It's a good idea to record your usability tests, but you should also take notes. If you have multiple observers, have each observer write down 3 key takeaways. After the usability session is complete, compare notes and discuss the highest priority usability issues that you can fix.

Keep usability tests simple so that you can conduct them regularly. A good cadence is once a month.

Recruiting and scheduling participants for a usability test

Don't waste time coordinating individual appointments with participants. Instead, make the times that work for your team and invite participants to choose from one of the open slots.

  1. Coordinate the availability of your team. Choose 2 time blocks, each about 2-3 hours long. Give yourself a little breathing room by scheduling them at least a day apart.
  2. Create 1 hour "slots" for participants to select from. It may be helpful to use a tool like Signup Genius for this task.
  3. Send an email to a large pool of potential volunteers, and invite them to choose one of the slots.

Sample email invitation:

Hello! We are looking for volunteers to help us test [a new tool that helps tenants
report problems in their home].

If you choose to participate, you will be paid [$100] for 1 hour of your time.
You can participate remotely on your Android, iPhone or desktop computer. You will use
the free Zoom app to share your screen and participate remotely.

During the test, we will show you a website and ask you questions about it. Then
you will be given some short tasks to complete. We will ask you to narrate your
experience aloud as you work on the task.

What you tell us will be used to help us find parts of the website that are
hard to understand and can be improved. We are testing our website--not you!

This is your chance to improve an important tool that will help [tenants around the
state]. We're looking forward to your signup!

Please click this link to pick an open time: [https://signup.example.com/my-signup]

Yours,

[Quinten Steenhuis and the app team]

Sample welcome email:

Dear [Jane],

We are looking forward to seeing you at [11:00 AM on Wednesday September 28]!

How to join the meeting
The usability test will be on Zoom. Please use this link: [https://us02web.zoom.us/j/1234]

We will ask you to share the screen on your phone or computer while you meet with us. There is a screen sharing feature built-in to Zoom.

What to expect
The goal of this meeting is to help us improve a new website that [helps tenants get repairs made in their home].

We will ask you to talk out loud as you use the site. What you tell us will be used to help us find parts of the website that are hard to understand and can be improved. We are testing our website--not you!

We will not be able to give you specific legal advice during the meeting.

Our meeting will take about 1 hour.

What to plan before you join
Please make sure you have a steady Wi-Fi or data connection while you meet with us. Try to find a place where we can hear you clearly while you talk out loud, such as a quiet room.

Compensation
We will give you a $25 gift card to Market Basket as our thank you for taking the test.

Please let us know the best way to get you the gift card. We are happy to send it to you by mail if you share your address.

Thank you so much for agreeing to help us!

Yours,

[Quinten Steenhuis, Lemma Legal Consulting and the website team]

Sample prompts for a usability test

While there is a lot of information about testing websites for usability, a test for a guided interview is a little different. Primarily because the process is much more linear.

Here are some questions that Jarret and Gaffney suggest you use when testing a form:

  1. Could you tell me what that question is asking you?

  2. What kind of information do you think that question is asking for?

  3. Did you expect to be asked that question?

  4. Does it explain why it asked that question?

  5. Did it leave out a question you expected?

  6. How would you figure out how to answer that question?

  7. Where would you look to answer that question?

  8. Is it OK for the website to ask that question?

  9. Is that the kind of information you'd be willing to provide?

Testing for correctness

Testing for correctness usually means:

  • Ensuring that your user can download a completed form, or otherwise reach an end screen
  • Ensuring that invalid input (however you define it) is not submitted
  • Ensuring that the user does not see an error screen
  • Ensuring that your form is "legally" correct--that is, that the business rules of your form are correctly implemented

Testing for correctness is important throughout the form development process. Errors can creep in at any stage. Errors may also be "regressions". A regression is a bug that comes back after being fixed. A feature might be complete and work one day, and a change that you make to seemingly unrelated code can reintroduce an error. That means you will want a plan to run and re-run your tests multiple times. Automated tests can help make the re-running step realistic.

Keep in mind that reaching the state of "zero" bugs is almost impossible. Instead, think about testing the most common situations and provide a way for your user to let you know about bugs you didn't discover in advance so that you can quickly resolve them.

Software tests usually come in two varieties:

  1. Unit tests, in which a specific function or method is tested in isolation
  2. End to end tests, in which multiple full "paths" through your tool are tested to ensure that the different components work in combination.

Unit tests can be very helpful for testing Python code, but less useful in the most common kinds of guided interviews which involve little code. Instead, you will mostly rely on end-to-end tests of the entire form. End-to-end testing can be tedious, but it's important to release software into the world that is relatively free of bugs. Bugs will reduce user confidence in your tool.

The ALKiln framework, developed by the Document Assembly Line Project, provides a method for automating end-to-end tests in a relatively flexible way. However, setting up an ALKiln test is still an advanced task. Most users will get by with a series of manual end-to-end tests that concentrate on getting good "coverage" of the most common paths through your guided interview.

The next section discusses a simple manual method of conducting end-to-end tests by developing realistic user scenarios.

When is testing complete?

No software has finished testing before it is in the hand of real-world users. Your users will find bugs that the most carefully planned testing will miss. You need a plan to gather feedback from your app once it is released in order to catch these late bugs.

Further reading