The more complex the world becomes the less we can depend on individual experts. 

For the last year or so I’ve been collecting headlines about the Euro. There were great expectations that it would unravel. At a minimum, Greece would leave. Michael Schuman, writing for TIME, said in May 2012 that “we really have to ask if the game is up.” Richard Quest, CNN's foremost international business correspondent, extolled that, “The wheels are coming off the wagon. The fat lady is about to sing.” In June 2012, George Soros said that there were three months to save Greece. Nobel prize winner Paul Krugman confidently stated that, “Greece will leave the Euro.”

Who knows anything these days? It seems that in a world that is becoming more and more complex, the smartest people in the room don’t have much of a clue.

The geek intelligentsia hated the Samsung phablet. According to Farhad Manjoo, “Smartphone industry blog Boy Genius Report called the Note 'the most useless phone I’ve ever used,' adding: 'You will look stupid talking on it, people will laugh at you, and you’ll be unhappy if you buy it.' Gizmodo argued that the Note 'isn't just designed poorly -- it's hardly even designed for humans.'" Manjoo goes on to note that he was derisory about the phablet himself.

Samsung sold 10 million phablets in 2012.

What on earth is happening? How could these experts be so totally wrong? And this isn’t an exception. In a growing number of instances where I’ve tracked a major prediction I’ve found it way off the mark.

The world is very, very complex. It’s increasingly hard to predict what will happen in the future. It is increasingly necessary to be adaptable, flexible. We need to be driven by data and evidence, not the opinion of experts, no matter how well-known they are.

We should be constantly researching and never trust a single piece of data. If an interesting finding comes through, we need to compare it with other findings. We need to focus on actual behavior, not on what people say. We need to test and rapidly iterate. We need to be able to make small changes on a continuous and rapid basis.

At all costs, we web professionals should avoid opinion. I regularly do tests with web professionals asking them about design and content choices. In most tests, these professionals get 80% of the questions wrong. Yes, given a choice between two design options, thousands of web professionals choose the wrong one. But how do we tell what the right one is? We test it with real customers.

“By the end of the campaign our 240 a/b tests lifted the donation conversion rate by 49%!” That’s according to the Barack Obama web fundraising team. They raised $250 million.

Some organizations say they can’t afford to test. That they can’t afford to observe the customer. The reality is that in this complex, ever-changing world, you can’t afford NOT to test. The intelligent are full of doubt. Only the idiots are certain.