Having gone through hundreds of web startups, I wasn’t always able to predict which ones will become a success and which ones will die out. But one big obstacle to the success of a service I was always able to identify: unnecessary complexity. If I weren’t able to figure out what the service/product is about in the first 10 seconds, or if it took me more than one minute to start using it, I knew something was wrong.

I’ve inadvertently done a little experiment here on FranticIndustries. First, I had a form in the sidebar for those who like to receive articles from FI via e-mail. Then, for visual reasons, I have removed the form and left only a link, which then lead to the subscription form. A couple of days ago, I’ve put the form back where it originally was.

Although most readers prefer to subscribe by RSS, a certain number did subscribe via e-mail when I had the simple form. In the period when I removed the form - approximately 3 weeks - I received zero new e-mail subscribers. In the three days after I’ve reinstated it, my e-mail subscriber count started rising again.

The lesson learned is simple: although it was still flabbergastingly simple to subscribe with a text link - all you had to do was click the link and then enter your e-mail address - this extra step was enough to completely stop people from subscribing. The simplicity and the luxury of being able to just type your e-mail address into the already visible form field and press a button was the tipping point that made a huge difference.

This small-scale experiment can easily be compared to many popular services. One great example is Digg. Digg’s voting system is as simple as it gets. You can only vote up, and it’s done via an always visible button with a vote count that refreshes the second you vote. Had it been any more complicated than this, I’m sure Digg would have been a failure. I’m certain that the developers of Digg considered adding a more complex voting system, for example letting users grade stories from 1 to 5. They did well by not going down that route. Anyone who tries to make a Digg clone should know this.

It’s important to understand that this simplicity has little to do with users’ intelligence or knowledge. The Internet has made us all lazy conformists with an attention span of about half a second. With most things on the Internet being free and instantly available, Internet users have acquired a nasty habit: they want it now or they don’t want it at all.

What can a developer of a web service or a product do to avoid the mistake of complicating things too much? How do you keep it dead simple; so simple you can be absolutely sure that no user ever gave up on your service because he was annoyed by too many steps? Here’s a couple of tips:

• Never have a splash page with no other purpose except being an introduction to something
• Always provide a full-featured demo of your service right on the front page
• If you must require a registration, make it a one step process and require only necessary info
• Make sure your registration confirmation arrives instantly. If it doesn’t, the user will forget about you
• Once registered, the users should be automatically logged in to your service every time they visit it via a cookie
• Explain concisely and clearly what your service is about and put it on a prominent spot on the front page
• Additionally, always have a visible “about” or “faq” link which explains everything the user needs to know in more detail
• Always reduce the number of steps to complete any task on your service to the absolute minimum

These does not ensure the success of a product or service, but they will surely help prevent (unnecessary) failure.