Usability testing

Effective user research to make the web more usable for more people

Why usability testing?

This isn't a talk about tools, technologies or frameworks

Who am I

Calum Ryan
  • Front-end Developer at Helpful Digital
  • Twitter calum_ryan
  • calumryan.com

My satisfaction

Satisfaction comes from… Knowing what we create satisfies and solves problems for users

Making too many assumptions

  • Our users have the latest devices/software
  • Our users don’t have any disabilities
  • Our users should already know how to do that
  • Our product isn’t that widely used to matter
  • Previous products/approach worked fine

  • We build features people don’t need or hardly ever use Outcomes of not doing usability testing
  • We may fail to consider different use cases, challenges/constraints
  • Potentially break the law, get sued, bad press and reputation
  • We turn people away to alternatives/the competition

Usability testing / research

We’re testing the website/product. Not the user

Quote from Just Enough Research

“The goal is to determine to what extent the product or service as designed is usable”

Erika Hall Just Enough Research | A Book Apart

Measuring usability

5 components of usability - Nielsen Norman Group

  • Learnability
  • Efficiency
  • Memorability
  • Errors
  • Satisfaction

5 components of usability - Nielsen Norman Group

Learnability

Do I need to consult documentation to use your website?

Efficiency

I have a 100-odd other tasks to do today. Will this take long?

Memorability

This website isn’t how I remember it. How do I do that again?

Errors

Users struggle to find X resource. How can we make X easier to locate?

Satisfaction

Is the website accessible and somewhere you’d happily visit again?

Methods of user testing

  • Explorative
  • Comparative
  • Thinking aloud
  • Remote
  • Hallway
  • Assessment
  • Review-based

Consider one or more different approaches to user testing

Explorative

  • Ideally performed by experienced testers
  • Evaluate new or initial design ideas
  • Iterative tests, defines future tests
  • Test sessions are defined by goals

Comparative

  • Two or more design variations
  • Rate positives/negatives in user experience
  • Typically A/B testing
  • Combine best features

Thinking aloud

  • Typically a series of pre-defined tasks
  • Participant describes what they’re doing
  • Gather feedback as they progress (or struggle) to complete the task

Remote

  • Synchronous testing Real-time user testing via video link or remote sharing tools
  • Asynchronous testing Automated/prescribed tests the user may choose to do in their own time and in their own environment

usertesting.com

Hallway

  • Testing done in high footfall public places
  • Voluntary participation random individuals
  • Quick, low-cost way to get feedback

Assessment

  • Testing of a prototype ahead of production
  • Real-time monitoring often from another room
  • Progress and reactions to series of tests

Review-based

  • Typically tested by experienced testers
  • Pre-defined tests, often using automated tools
  • Deep interrogation tests pre-production

Rating the problems

Rate each problem users encountered during the tests with a common scoring system

Severity
  • Fails to complete task (High)
  • Completes with difficulty (Moderate)
  • Minor problem (Low)
Frequency
  • 30% of participants (High)
  • 11-29% (Moderate)
  • 10% (Low)

Participants

Find the right participants and treat them well

Aim for 5-10 participants and usability tests

Avoid using your own team as participants

Testing for accessibility

  • Get representative users
  • Choose participants with a range of different disabilities and combinations of disabilities
  • Allow users to test with their own aids such as screen readers, devices or other assistive tech
  • If required to attend in-person, check the venue and transport to there is accessible

Testing environment

  • Go to where the people are
  • Have separate rooms for the participants and observers with good quality mics and network connections
  • Choose a comfortable space: a relaxed, home-like environment away from distractions and lab-like conditions
  • Encourage honest feedback

What to ask

  • Get to know your participants and build a rapport
  • Profile them and their web usage habits
  • Have scenarios to work through
  • How would you do…?
  • What do you make of this?
  • What would you do here?

Usability Demo

View video demo from Rocket Surgery Made Easy by Steve Krug

Handling feedback

  • Choose a good facilitator
  • Friendly but neutral
  • Encourage honest feedback
  • Acknowledge problems/failures positively
  • Be confidential and protect user data

Guerrilla User Research

“help the participant visualise the future product clearly enough to give useful feedback on whether they can achieve their goals” Jamie Levy Chapter 8 | UX Strategy | O’Reilly

Guerrilla User Research

  • Validated research conducted in short space of time on a small budget
  • Small, structured experiments with a prototype and 5-10 participants
  • Typically done at a coffee shop with good WiFi, power sockets, not too busy and acceptable to stay for the day
  • 1 participant, 1 interviewer, 1 note taker

Final thoughts

What usability testing isn’t going to do

  • Provide a breakthrough design
  • Guarantee a successful product in the marketplace
  • Prepare you for every single scenario or use case
  • Substitute for QA testing (but can help define QA tests)

Some typical feedback from usability testing

  • Difficult to use and overwhelming forms
  • Links and resources aren’t highlighted
  • Challenging/hidden navigation menus
  • Content isn’t presented or tailored for different audiences

Be open about your usability testing. Share your findings

“approach usability testing with the right expectations and conduct it early and often”

Erika Hall Just Enough Research | A Book Apart

Just Enough Research by Erika Hall

UX Strategy by Jaime Levy

thank you