UX so simple, test early, test often; fail fast, fail cheap

My maxim as a button.

The title says it all, you basically learned the most important thing of this post. Continue reading in case you are interested in the reasoning and further resources how to get this done.


My Maxim

This is one of my maxims I teach to my students. If there is one thing they should remember from my course after ten years it is this phrase:

Test early, test often;  fail fast, fail cheap.

You fail!

You may have heard this in a similar way already in the past:

“Test often, fail fast, fail cheap” (I think I heard it first time in scrum training), or “Fail Early, Fail Fast, Fail Often” e.g. discussed in a blog from, and it really is a  favorite saying in Silicon Valley as stated by However, the Lean methodology take this concept one step further with the statement: “Think big, act small, fail fast; learn rapidly“, as dzone explains in their blog.

So this is for the lean start-ups, but it is still all about failing. Here is the thing: failing is not nice, actually it hurts, each and every time – especially in case it is you who fails. But the more you train to get over it, the less it will hurt each time. However, I would like to streamline things a bit more and go to: Fail fast, fail cheap, as found e.g. from zurb. But by the end of the day you will fail. Each and every time you do something you will fail, somewhere. We are humans and we are not perfect, thank god (or who/whatever you say thank you to).

However, I do have a problem with the “fail often”, as I am not from the UX masochist division. To my understanding, failing often is just plain stupid and it shows a lack of understanding of what UX actually can do for your business. Why should we fail fast? Well, first of all it reduces costs if we spend only a little money and time on something; so we fail cheap – the sooner, the better. But how do we fail as fast as possible to fail as cheap as possible and ultimately to reduce costs?

Here is the thing: You fail faster if you force yourself to find the failure.

Dude, is there anything else than failure?

Yep there is. So far  we were just talking about failure. And here is the bridge for the gap. I am big fan of Steve Krug and his book “Rocket surgery made easy” (I am a fan of his other books as well, like “Don’t make me think” and the George M. Prince’s reprint of “The Practice of Creativity“). The next paragraph is a spoiler:

He basically describes a system in the loop tested every 2-4 weeks and he compares it against the old traditional big usability studies. I did something like this before the book was on the market, simply because I had to fit tests into scrum development cycles (2 participants every 2 weeks, it was a mobile app project) and no budget was foreseen for that, but kudos to Steve, he tied things nicely with each other and I still found many eye openers in this book; I love to recommended this book just because he describes these ideas so nicely; I actually gave it as present to other interaction designers (yes plural).

So this is about the “test often”. The other thing is about test early. Steve describes this in his book as well. Test whatever you have, competitors webpage, your old webpage, sketches on napkins (the uglier the better and no excuses). Read my lips (ok this line): NO EXCUSES! Today there does not exist anything like “too early” when it comes to testing and design.

Yay, finally: “Test early”.

So and here we are: Test early, test often; fail fast, fail cheap.

From a process point of view, the most important thing is that you get a foot into the door, preferably even before the coder gets her first energy drink, which to my experience is usually before she even touches the keyboard 🙂  (i.e. run your studies before people start coding). Once your usability studies are part of the design or development cycle (even in waterfall, not only in agile) you have good chances to lower the development costs (ok there are more obstacles, like the developer having the keyboard and not you, or (wannabe) design managers with king-size egos, but this is for a different post).

Good news everybody

You do not need to be a usability professional. You can learn it just by doing – and failing. You will not be as experienced as a senior, but also they started with their first test. However, the most important requirement for becoming a usability tester is about the mindset. You have to be able to fail and let people shoot down your idea one after another. Yet you have to ask them to continue beating you up (ok there is some masochism involved, maybe) until the hour is done. But your product will be better than all of the competitor’s solutions, simply because they care a s**t about usability studies. You know better now.

In addition, the tests are ridiculously cheap. At best they cost your working time and some incentives or compensation for the 2-3 study participants every second week. So the only thing you need to become a usability tester is common sense and discipline. Both are sometimes hard to find in every profession as there is also something called “ego”.

Wait wait wait, it costs my working time. I cost a s**tload of money

I found a great explanation on that: In Tomer Sharon’s book “It is Our Research” there is one short interview with Ben Albert, Director, Design and Usability Center, Bentley University, US. His example is dead simple: It is like asking for the way if you do not know it. It takes only a few minutes of your traveling time, but you will reach your destination. If you do not ask, chances are quite high that your journey will take longer and there is even the chance that you will never arrive.

You may also think it this way. Your team with 10 people (manager, developer, testers, graphics dude or dudette, trainee, marketer, copy editor, sales…) and one person goes with 25% of her time to prepare and run the test. This is 2.5% of the team’s time spend for usability testing. This loss of work time is hidden in noise, and an engineer should understand the term “noise”.

So act today, order Steve’s book (whichever format, and I am not an endorser of Peachpit publishers) and reserve now a half-day slot every 2 weeks. Now. Yes now. Immediately. In Finland we would say “nyt heti”.

So this is what helped me in the past a lot; and these explanations usually are an eye opener to all of my students and clients. I actually got new clients just by talking about this topic.

Happy testing.

PS: I did the badge after watching a tutorial video from Chris Spooner.