Talent wins games, but teamwork and intelligence wins championships.
There is no 'i in 'team', but there is a 'u' in 'suck'...
-Dave the Platypus
I am often called upon to interview people for various technical jobs and to assess the skills of those already in the organization. The goal of each assessment (interview or otherwise) whether directly verbalized or not is to uncover the "superstar" that is just waiting out there to be hired or promoted or even simply recognized. Or, more accurately, to decide if someone is a mere mortal or if he/she will ascend to the ranks of "demi-god" within the organization.
There is a problem, though. The "demi-god" we are all looking for doesn't exist. The quest itself is the mythical snipe hunt.
"What?" you may say. "I know oh so many superstars! How can you say he doesn't exist?"
Calm down there, Sparky. Let me explain. I am not trying to demean talented individuals. I like to think of myself as one, and while I can be as self-deprecating as the next egomaniacal geek I am not so stupid as to say something like "talent doesn't exist nor does it count". The point I am making - and I do have one - is that the process of trying to identify the so-called "superstar" is impossible because it is based on a faulty comparison, that the labeling of someone as a "superstar" is unfair and that ultimately identifying any person consistently as a superstar has unintended side-effects to your team as a whole.
It would probably help if I identified what I consider a "superstar". It is not something I arrived at on my own, but rather an accepted industry definition - details may differ, but as a whole the concept holds water no matter where it is applied. Generally speaking, superstars are people who consistently over-perform to expectations and most often by some consistently high order of magnitude compared to the other team members. When a project is in crisis-mode, they are the people we look to swing in to action and save the day. When we have a particularly difficult problem to solve, superstars are the people who come up with the solution faster and better than anyone else. When we want to provide an example to the rest of the team, superstars are the people we point at and ask "why can't you be more like him?" They are the impossible standard made possible.
The basis for that definition comes from the concept of "The 10x Engineer", an idea with roots growing from a couple of vaguely scientific studies back in the 1960s and founded completely on an awful and misleading comparison. There is a presumption that the "superstar" is someone who is roughly worth 10 of the average team members. Before you go asking how I came up with the 10x figure (see - I know you!), let me lay on you an interesting article which explains the origin of "The 10x Engineer" through an analysis of those studies and courtesy of someone named Shanley, I am not entirely sure who she is, but the article rather impressively tracks the history of the superstar performer (not directly by name, but rather through the definition of "The 10x Engineer" and how that concept came to be applied) and how overall the idea is not sound. Summarizing, the thought that a superstar outperforms the norms by a factor of 10 actually comes from reverse thinking. The original studies which started the superstar myth actually says that poor performers can consume as much time and resource as 5, 10 or even 20 good performers. And remember, at this point we are comparing poor performers to average or good ones - we have not yet even introduced the concept of the superstar.
I shouldn't need to point this out, but the studies clearly state that we are comparing "good" performers (or normal team members) to those who perform poorly. To extrapolate that good performers outshine poor ones by a factor of 10 seems like a rather wild leap to a conclusion. It is much like the syllogism "God is love and love is blind, therefore Stevie Wonder is God." I know, silly example, but how is that different that the leap we are making comparing good and poor performers?
Further, when we make this comparison we are also clearly not talking about output but consumption. In most fields we measure performance by output - how many papers someone wrote, how often they met deadlines or how well the given task was performed. People who perform poorly drag down the rest of the team because they consume an inordinate amount of resource and time to arrive at any outcome, not just the right one. I am not sure anyone would argue with that logic. What they potentially use when compared to other functional team members is between 5 and 20 times the available (fill in the commodity, time or resources) whatever.
At this point most sane people would interpret that to mean that we should be looking at cutting poor performers to create a bigger pool of that "whatever" for those who can make the most of it. The idea over time became twisted, though, and rather than being used to address the allocation of resources, it instead became a measure for our best and brightest. The original comparison is a far cry from saying that one of your team has 10 times the output of the others, and when we make that leap, we establish the boundaries for the definition of a superstar. In that unfair (and misleading) comparison is the origin of our trouble with the concept.
While it is perfectly appropriate - and desirable - to recognize personal achievement within a team, calling someone a superstar actually has adverse effects that most leaders don't consider. The biggest adverse effect is that it places undue expectations on the team, on the superstar and on the team leader. We start to expect that the superstar will always outperform (by our definition, a factor of 10 at a minimum), and by calling out the team member as a superstar, we set up the expectation that we are wanting more - perhaps more than can normally be expected and possibly more than the superstar is capable of delivering. Setting goals is one thing, and stretching to reach them is what growth is all about. Creating the superstar mythology on a team sets the bar too high for mere mortals to reach and inevitably leads to disappointment when "the chosen one" doesn't perform as expected.
Then there is the effect that labeling a team member has on the rest of the team. We call that effect "demotivation". It sends the message that if you aren't able to hit your goals as a part of the team, don't worry! Our superstar will fly in and save the team's collective bacon. It also to some extent demeans the efforts the entire team has put in on a project, making the people who comprise the team feel less like a team and more like meat for the grinder - with your superstar being the grinder.
And finally there is the effect it has on the ego of the superstar. It says that you are god-like in your ability to get things done, and therefore you can do no wrong. On some level it suggests that the superstar can take unnecessary and even unwarranted risks, that he or she can redirect resources and efforts on a whim, and that the rest of the team are merely supporting players. Granted, this is not an inevitable outcome, but it is a common one. I have worked with so many people over the years that have found their place in the organization in this way, and in almost every case it has lead to catastrophic results like product failures (numerous bugs, late launches, inconsideration of customer needs, teams crashing and burning just to name a few) and even in one case the implosion of a company. And in the post-mortem for such failures you can (most of the time) trace the root cause back to the crowning of an individual as the superstar that would provide the leadership and guidance to the organization. That they did, just not in any kind of desirable way.
While the superstar individual is a long-term myth, the concept of a superstar team is not. Having a diversity of thought and skills on a team not only brings balance, it also ensures that each team member can feel the superstar love based on what they contribute to the whole. As Jon Galloway of the Herding Code podcast says: pairing "step back and think" devs with "crank a lot of pretty good code out" devs is a recipe for a good team. So that builds a good - maybe even a great - team, but how does that help an individual team member feel the love? Well, back in my day job, my teams vote on who they think the MVP for each release is. Rather than me as a manager bestowing a crown upon the brow of someone, their peers do it for me. It provides motivation for team members who want the recognition to find ways to contribute more. It gets the team to think about how each member works.
And, in the end, it helps me find the "meat" in the team, the ones who are taking cycles away from the rest without giving back. Occasional pruning makes the team stronger. It makes my team superstars, and in an already over-ego drenched world it provides sanity, competency and yes, even success to a group of people who love what they do but don't necessarily need to be "personalities".