Published 2010
http://www.sensible.com/rocketsurgery/index.html
Steve Krug wrote “Rocket Surgery Made Easy” so it can be easily read on a plane. I did just that on a short flight from Toronto to Chicago. It was such a quick, easy and enjoyable read that I had time to write down a few notes for this post.
Published in 2010, “Rocket Surgery Made Easy” is a good follow up to “Don’t Make Me Think” written by Krug about 10 years earlier. This book is a great read for anyone involved in interactive design and creating user experiences. Rocket Surgery Made Easy targets how to do effective usability testing, debunking the myths of it being complicated and expensive (Nothing new – Krug admits Jakob Nielson has been doing this from the 80s).
Below are a few notes of what I took away from this book and some of the highlights.
Usability Testing: Watching people trying to use what you’re creating/designing/building (or something already created/designed/built), with the intention of (a) making it easier for people to use or (b) proving that it is easy to use.
Krug's maxims are his core values of what you should remember from this book.
- A morning a month, that’s all we ask.
- Start earlier than you think makes sense
- Recruit loosely and grade on a curve
- Make it a spectator sport
- Focus ruthlessly on small number of the most important problems
- When fixing problems, always to the least you can do.
IF you can afford to hire a usability professional to do your testing for you – DO IT.
Test frequently and test often
- Whether it’s a sketch on a napkin, wireframe or completed design, usability testing can be done.
- Minimum Test once a month, 3 users and 3 Tests.
- You can test more but that requires more deliberation, scheduling and planning. Try to get the most value from the least impact/investment. Focus on tweaks.
- Test other peoples sites/apps – auditing your competitors is a very under-utilized resources.
- Find out what already works well
- A good way to wet your feet if getting into new usability
Who do you test with
- Subject Matter Experts do not make optimal users. Users types are changing constantly. Users with domain knowledge will look for keywords and signs they are familiar with. You can get as good of results testing users without domain knowledge. If someone who knows nothing about a subject can complete a task, odds are, someone with domain knowledge will do it as fast or faster.
- Recruit Loosely and don’t test the same person twice.
How to conduct testing
- Steve Krug outlines a simple testing example/method for you to follow and gives some suggestions on how to conduct the tests
- Some of the core concepts are to keep the tests short, and test areas that you think need testing.
- The question I hear a lot from other departments is “What do we test”
What to test
- Often customers who have a monolith of an application ask “what to test” when you have a huge application and have never done any usability testing in the past this can be quite a task.
- Come up with a list of tasks by what is the most important thing for a user to do on your site/application (5-10). The secret is to make the tests reflect your users actual goals opposed to what your idea of what you think they ought to do.
- Focus your tests on those 5-10 main tasks. Depending on the scale of the task you may want to spread it across multiple sessions
- Allow all developers, project managers and stakeholders to sit in and listen to the test (not in the same room)
Fixing the Problems
- Once the tests are complete, fix the issues that require the least amount of effort. The idea… make it better for users right now.
- Tweak… don’t redesign
- Take something away
