If you buy this coffee machine you will want to die

The fact is that many websites and web businesses cause this feeling in their users.

And I can tell you exactly why and exactly how to fix it.

But first, have you ever bought something without reading any reviews and got horribly stung? We bought this coffee machine for our office before reading any reviews. Looks nice, doesn't it? Oh yes. And I'm sure it makes perfectly nice coffee too. I wouldn't know because I never actually managed to get a drop into a cup.

On the upside, it did introduce me to my all-time favourite Amazon review:

Click to read this whole review in all its glory.

Click to read this whole review in all its glory.

Look! The product looks good: smart and shiny.

But it's badly designed. Why? Because the experience of using it really does make you want to die. (At least then you wouldn't have to clean up 2 litres of spilt coffee.)

I want to know:

  1. How does a product this bad make it to market? 
  2. How does a well-meaning designer make such a pig's ear out of a product's core function? Ie. getting halfway decent coffee near and ideally into my face?

I can think of three ways:

  1. The product designer's goals don't match the user's goals 
    Perhaps their goal was to make an attractive sculpture of a coffee machine. Perhaps it's a modern art project; a scathing commentary on an over-caffeinated world (see how much less annoyed you get with the product as your caffeine-dependence reduces.) Or a secret plot by Starbucks: "Trying to save money by making coffee in the office? Not with this machine you won't, sucker!"
  2. Nobody tested the product 
    Bravado: "Hey, we've made _loads_ of coffee machines… what could go wrong?"
    Time-pressure: "No time to test – it has to be on the shelves yesterday!"
    Or perhaps the company just don't test their products. (After all, if you chuck enough turds at the wall, some of them are bound to stick...)
  3. Someone tested it and found it broken... but didn't do anything about it
    I reckon they were unable to report the problem because the acute caffeine withdrawal rendered them unconscious. Or maybe they couldn't bear to break the product designer's heart. Or — more sinister and more likely — maybe company culture discouraged the bearing of bad news.

Are you testing your stuff?

If not, why not? If so, why not more?

Nobody can logically argue against testing stuff with users. Disagree in the comments, if you like. But here's why I believe this:

  1. It's the single most valuable activity you can do to improve your products, marketing, sales, etc.
  2. The only reason people don't test is fear. They dress it up in other clothes, but it's fear underneath.

I'm going to focus on websites now, but these excuses apply across the board. Here are some of the outfits you'll see fear wearing:

"We don't have the time"
Really? It can take as little as an hour or so to watch some users, but you're happy to risk weeks of selling nothing because your website doesn't work. It doesn't add up.

"We don't have the budget"
Daft. You can test most websites for absolutely nothing. Just collar a stranger in a coffee shop and ask if they have 5 minutes to look over this "sketch your web developer sent" in return for a coffee. (Are you feeling the fear now, after the suggestion of actually talking to people?)

"We don't have the skills"
OK, OK, there's a grain of truth to this one. Skilled professional usability researchers will get more out of a test. But good old Pareto's Principle means that you'll get 80% of the results just from showing up and being moderately switched on.

"User testing only finds problems: we want to add positive things, not just fix negative things"
Balderdash. This is a myth made up by fear. Yes, user testing does identify where some of your users will lose their way, lose the plot, or lose the will to live. It gives you insights, not solutions. What you do with the insights is up to you


My mission is to do more user testing and get the world in general doing more testing. More on this will definitely follow.

So come on: what's stopping you from testing?