“Exploration is about formulating hypotheses or best guesses; confirmation is about rigorously testing preliminary conclusions. Confirmation turns best guesses into sure bets. As in scientific discovery, the less we know about a phenomenon, the more openended our questions. As relevant knowledge builds up, we become more precise about what we seek to learn, and we start to anticipate (more and more accurately) what we will find. Because hypothesis-testing experiments (for example, taking a new job on a provisional basis) are usually more costly than exploratory experiments (for example, working on a side project without leaving one’s job), we prefer to defer the former until we have solid data suggesting that we are going in the right direction. Variety for its own sake is not enough. In fact, a prolonged exploratory phase can be a defense mechanism against changing, and it can signal to others that we are not serious about making change. A true experimental method almost always leads to formulating new goals and new means to achieve them. As we learn from experience, we have to be willing to close avenues of exploration, to accept that what we thought we knew was wrong and that what we were hoping to find no longer suits”