Over the years I managed to build some really good relationships with the people who worked for me in development groups large and small. A few of those relationships I still maintain despite leaving the companies involved years ago and a select few of those relationships have turned into valuable friendships. As I spend more time in selected groups, it is natural to develop a better relationship with some people than others, simply because of mutual respect, dedication, common interests and approaches, and ability to depend on some more than others. As part of the ongoing relationship building, questions get asked that deserve candid answers apart from the official corporate line.
One of the most common questions I get from people is about leaving the current company and moving on to others. Usually, this is not a decision taken lightly unless there are extenuating circumstances such as a move, friction between company or individuals, or lack of performance. Those are different issues than the one of interest here. Where this becomes interesting is when a valued employee asks whether it makes sense to move on to another company for personal or professional reasons. Although the loss of that employee can be deeply felt in the current company, the short answer is that often changing companies is the only way to move up the ladder of promotion significantly, get a noticeable jump in salary or benefits, or learn new technologies. And so, it's often down to me to explain these issue to the person I really don't want to lose, but have to acknowledge moving on is the best thing for them.
Often it's for purely professional reasons that people want to move. They may have spent the last few years working on something like .NET, for example, but want to broaden their own skills and work in a LAMP (Linux, Apache, MySQL, PHP) environment. In a Windows-centric company using .NET, the chances of learning LAMP is minimal (or superficial at best). Moving to a company that is LAMP based is the best thing for that individual, especially if they relish the challenge. Sure, they could stay where they are and become a .NET guru, but most technical people throve on learning new language sand environments, and the only way to realize that is to move to a company that embraces those technologies (even if it means starting at the bottom rung all over again).
Sometimes the move is inspired by a need to accelerate the career path and salary. Let's face it, when you've spent three or more years in one company, especially one that is not growing rapidly, the chances for moving up the management ladder are limited and don't come by often. Also, corporate salary guidelines being what they normally are, salary increases are not going to be dramatic but small, routine increases. Often, by changing companies, a technical person can both move up a level (to a lead or director position, for example) immediately, as well as negotiate a salary increase that is worth the move. (As a general rule, I recommend moving only when the salary increase is 20% or more, but that's assuming no extenuating circumstances). So, by changing companies, the employee gets a big pay increase, a promotion, and a chance to start in a new environment. For some, that's worth leaving the comfort of a long-held job in a solid company.
Of course, there's also the issue of getting stale. Doing the same thing, day after day, can become tiresome whether you're a programmer, a lead, or even a senior executive in the company. When things get boring, some people get complacent and comfortable, and they lose their drive and their edge. Some, on the other hand, get impatient and want to move to something else that will keep them challenged and interested. Since the dynamics of a corporate group are pretty slow to change, moving companies is often the only way out of that rut.
A lot depends on the individuals involved. Everyone is different, but generally there are those who are driven and need to constantly be challenged (and they are the ones who will move to other companies if they can't get those attributes where they are) and then there's those who are more laid back and willing to accept the status quo and be comfortable in a less challenging environment. Development groups need both kinds. But those who move around will seldom be the complacent type. Sure, losing those driven people hurts the current company in the short term, but it is often the best move for the individuals involved. And that causes some angst for executives who have to balance the good of the group with the good for the individual. And that's when those relationships start to factor in, as the equation will usually swing more to the best for the individual. At least it does in my position.