I am sure you have heard it in all versions, subversions and perversions, but one simple universal workplace truth seems to be don’t tolerate poor performance, and if not outrightly eliminate poor performers, do ease them out over a reasonable but fast period of time. The yawning gap between a bottom performer and the top performer is perhaps nowhere more so prominent as in a human-skill based knowledge industry like software. Over the years, various productivity studies still continue to point a gap of anywhere between 1:10 to 1:20 or even more between the top programmer and the bottom one. The exact number doesn’t really matter.
What is important is to understand that success of a software endeavor howsomuch is dependent on very smart people, it finally needs a great team to deliver goods – no task is trivial enough to be single-handled performed by the star performer on a sustainable basis, nor is anyone smart enough to effectively comprehend every bit of information created by mankind. This doesn’t trivialize individual human talent, nor is meant to belittle those huge efforts that make a star performer a star performer. It only highlights the nature of the beast – to deliverÂ a non-trivial piece of software with reasonable complexity and business criticality, one must have a team comprising of team members that complement each other’s skills.
We all know a great team that works likeÂ Swiss clockwork doesn’t happen by accident. Throwing a bunch of people together, howsoever individually competent, in a blender doesn’t ensure that the result is sweet juices of creativity, teamworkÂ and performance. Instead, all one might get is a bloody mix of team dysfunctions, bruised egos, competing team members constantly on the prowl to backstab others…and so on.
So, how does one go about building a great team. Having the right intent, high bar (higher than what people believe that are capable of achieving)Â and zero-tolerance for poor performance is a good start. Often, it is easier said then done. There are also fair arguments that when you build a team six people, you can find the smartest people, but when your team needs sixty people, try doing the same. I have managed three fairly large teams: two of them were over 110+Â people and one of them had some 190+ engineers working on really complex softwares such as core routers and SoftSwitch, and I can tell you the mindset one needs to manage such large endeavors of human creativity are radically different from what I would need when working on a six-people teams. For one, there is no way you are ever going to get all smart people withÂ top talent – it is just impossible to hire people of equally hire calibre in such numbers on any team, any place on earth. People often don’t admit it, but I am going to hold on to my arguments. There is always a performance curve (“Bell Curve”) in any random distribution. My favorite quote here is from Mali, our HR Manager at Philips, “There is a Bell Curve even at Bell Labs”. Wider the bell curve, bigger is the gap between the top and a bottom performer that I talked about earlier. In a six-people team,Â it is possible to cherry-pick team members that leads to a very narrow bell curve. In a larger team, the curve starts to flatten out and widen at the ends. This is not by intent, but because of a combination of various factors:
- difficulty in finding large number of experienced people with required skills in the given domain
- you can’t always start with a flat staffing (meaning, achieve 100% staffing in the initial stages of the project) – there is not enough work in the beginning, and during the peak, everyone is stressed out and later in the project, there is overstaffing. A Rayleigh’s curve works reasonably well
- the bad economics of having an all-star team – even Manchester United can’t get all the best players in the world to play for it – they might go bankcrupt before the kick-off itself !
- inevitable dilution in hiring standards over time as hiring responsibilities get delegated, often to new managers who might not have the same level of organizational understanding as the original team
- many smart people might not want to join a large team for the fear of becoming just another face in the crowd !
- a large project is more like a marathon than a hunderd meter dash, and is often late, forcing people to forgo weekends and vacations for a prolonged period of time. Many people, irrespective of their talent, might not want to get stuck in such situations – so this already limits the gene pool to work on and work with
There are many more reasons, and even if you don’t agree with all of these reasons, it is incredibly difficult to build a large team. Coming to smaller teams, it is similar challenge, and many of the reasons mentioned above might very well apply there. However, when it comes to absolute numbers,Â I look at it this way: what is easier – raising a loan of $50,000 or $3million ? The fact is, everything else equal, you can almost always go a far better job of building a smaller team than a larger team. But even that doesn’t guarantee scintillating performance by our small team. The fundamentals don’t really change. If any, the stakes are even higher for there no safety nets – the team can’t handle inefficiencies, and probably everyone in the team is a multi-skilled multitasker (as opposed to an individual in a larger team where there is a higher level of ‘vertical differentiation’ and ‘horizontal differentiation’). If one person leaves the team, the impact is far bigger than a similar percentage attrition in a larger team that no HR metric on attrition can capture. And the same goes for low performance – poor performance is far more lethal in a smaller team than a comparable percentage of poor performers in a large team. You would expect a lowerÂ tolerance for poor performers in any team, more so in a small team.
And it is for this very reason, that I was surprised toÂ readÂ Managers stick with poor performers rather than hire new faces. It is unbelievable and very shocking that a good 70% among us would rather put up with an existing poor performer than risk a new hire ! I think most reasons there don’t make sense, and I think ‘denial’ as a reasonÂ is probably a bigger contributor than is generally credited for. Irrespective, I hope like hell this data is an anomaly. For if it is not, I see more serious challenges ahead. By not getting rid of the poor performers even in these tough economic situations, a clear message is being sent that not just condones, even promotes, that level of performance. The teams are as it is not performing to their potential. Everyone in the team knows who those poor performers are, but the fact that a lower performance is being tolerated clearly send mixed signals and confuses people, especially the committed, hard-working performers. Some among them might feel insulted and likely to leave, thus leaving the team in an evenÂ bigger mess. I think managers of those teams are doing the biggest disservice to their organizations: one one hand they are tolerating poor performance and on the other hand, their condoning behavior could be disenchanting higher performers.
With this attitude, I think it surely is going to be one long slowdown. For those companies, surely.