Giving feedback

Management is all about people. A good manager ensures that his team is continuously improving. Improvement demands that weaknesses and gaps be identified and suitable actions be taken to address them. For action to be taken, you need to apprise people of their weaknesses without demotivating them and that is non-trivial. Giving feedback is one of the most important and delicate tasks that a manager needs to perform.

When?
When to give feedback is something that many get wrong. In most organizations giving feedback is an annual ritual that most employees and managers dread. If you, as a manager or a team member, feel anxiety as feedback time approaches then you should reevaluate the way your current feedback process has been constituted.

Giving feedback is a continuous process. You should give feedback at an opportune moment as soon as you evaluate a task. This helps in more ways than one. Firstly, the task and experience is fresh. It is so much easier to discuss and take away actionable items at that time rather than after several months have elapsed and entire process becomes dependent on recollections - where there is bound to be significant disconnect. Secondly, continuous feedback is easy to assimilate and act upon as compared to when a whole lot is dumped on the receiver once a year. It provides ample time for learning and course correction. This is what truly provides an opportunity to learn from mistakes. Thirdly, employees tend to finish tasks with higher quality when they know that they would be evaluated immediately.

What?
Feedback is of two kinds - positive and negative. Most managers tend to provide only negative feedback. While the underlying intent might be to point out areas of improvement, it is equally important to let the team members know what they are doing well and encourage them to keep doing it. Giving positive feedback, encouraging team members, and helping them find their strengths is just as important as helping them find and improve their weaknesses.

How?
Useful feedback is direct. It addresses the actual issue instead of beating around the bush. Being direct does not mean being judgmental or focusing on recipient's personality. Instead it should be considerate, encouraging and focused on specific behavior that needs to be addressed. E.g. If you were to tell an employee that they are poor at software design, they would become defensive and would argue hard to defend their designs. After all, no one intentionally does a bad design - they are doing what they believe is the best. Instead, if you were to nudge them towards a better design approach and let them come up with a better design and later on just ask them to consider more approaches and possibly read up on few good software design books, they would feel good about it and work enthusiastically to design better. Same applies to feedback about one's conduct. Telling a person that they are rude is much less likely to produce desired results than discussing a specific instance and asking them if situation was handled a little differently, could the results have been better.



For feedback to be effective, you need to think it through before entering a feedback session. If you as a provider, do not have any specific message before you enter the meeting, it is highly likely that the recipient would also part without any specific message.

Two additional things that influence the effectiveness of the feedback session are sincerity and sensitivity. Each individual is unique and if you need to communicate the same message to two different individuals, you may have to fine tune it to match their personalities. You have to be sensitive to their needs and reactions. Lastly, insincere feedback that is not aimed at helping others improve is easily detected and quickly discarded by all.

Myth about multitasking

True masterpiece from Scott Adams that shows one of the ills plaguing all levels of contemporary software development. Numerous researches have shown that human brain is not effective at multitasking. In fact multitasking slows us down since multitasking makes us deal with irrelevant information as well.

Yet, all of us have been seduced by the lure of multitasking - overestimating our capabilities and trying to do two(pun intended) much at the same time. Responding to emails while attending meetings, preparing plans while sitting through architectural discussions, watching tv while taking a call from home, we have all tried our hands at it in many forms.

Reflecting on my experiences, I do not remember much of meetings/conversations during which I checked emails and my email responses at that time weren't great either. Unless we stop thinking of the "other things" we are bound to miss the finer details of our current task. That would also explain why reading in the bathroom is not multitasking unless you want to concentrate on finer details of using the bathroom ;-)


So, is multitasking overrated? How does one manage high pressure day(and night) job with new information flowing in constantly?

The answer to those question lies in how you adopt multitasking. Effective human multitasking is not very different from computer multitasking. A processor gives "appearance" of handling multiple tasks at once. However, at any given moment only one task is being executed - not a single cycle is spared for another task. Only a higher priority task can preempt the current task. If current task is interrupted (waiting for input/resource etc) then next higher priority task is picked up for execution.

In a very similar fashion, instead of attempting multiple tasks at the same time, we should prioritize and pick up highest priority task first and then give it our complete focus. Periodically, new information reaches us (new tasks, events) which should be used to adjust our priority and we should continue with execution of highest priority task. Undivided attention to a task is the only way of getting a task done with good quality. This is especially true of high priority tasks where there is little room for iterations for refinement and delays are just as bad as poor quality.

In short, only way of multitasking effectively is to deal with multiple tasks at the same time but not in the same moment.

As a manager, it is extremely important to understand this reality. It is all too easy to put a lot of tasks on a team member's plate. This might render them ineffective or affect the quality of their delivery. If a team member reports little progress on multiple tasks or cites one task as reason for lag in another, it is a strong indication that you need to sit with them and help them prioritize and execute. It might take a few iterations/weeks before team gets into the groove of prioritizing and then executing with focus but it is worth the effort, after all this is indeed manager's highest priority task - building a highly productive team.

Another one:

When religion is a blasphemy

"Scrum is my new religion" - exactly the kind of statement you'd expect from a real life incarnation of the PHB. Congratulations dear PHB, you have hatched the plot (oopss ... plan) to successfully "manage" to bring motivational levels of Alice to match those of Wally. It would not be too long before your entire staff would be required to wear leaf hula skirts, chant "bum for Scrum", and bow before the mighty Scrum lord for salvation.

Sadly, if you replace Scrum with any fad/buzzword/new tool probably you can relate the story to someone you already know. The PHB's are more common (and more weird) than I'd like to admit.

The task of a manager is to work with a team of people who have different ideas and view points, being able to align these varied perspectives towards a common goal, and facilitating the achievement of these common goals.

A real danger of having religious fanatics lead the pack is that they'd imagine the problems (goals for the team) that do not exist and then fit their religious practices in to solve the imaginary problems. If it weren't so sad, it would be amusing to see how disconnected from reality these fanatics can be. I've seen a manager working hard to improve productivity (and karma) of his team by forcing them to convert to vegetarianism.

If the manager has adopted a religion, how would s/he be able to appreciate and respect different ideas and view points? They'd end up ignoring valuable and enriching inputs that could have solved real problems effectively. Even if these inputs are not directly applicable they still serve a very useful purpose - stimulating a discussion and getting a stronger buy-in from the stakeholders by eliminating doubts in their minds.

Some managers get seduced by the seemingly easier way of getting the required alignment by forcing the entire team to convert. The only bigger blasphemy that a manager can commit than adopting a religion is to force his team to convert. An easy way to rationalize this approach is by convincing yourself that you know best and eventually the less knowledgeable team members would see the advantages once the benefits start pouring in. Having been on both sides of the fence, I can tell you from experience that it never works. This forced conversion violates the ideas and principles that team members truly believe in. They are never motivated enough to put their heart and soul in their work and consequently the imagined benefits never pour in.

Another issue with forced conversion is dealing with voices of dissent. Force conversion requires that one way or the other, these voices should be silenced - whether by reassigning them to grunt work, or by lowering their ratings, or taking them out of way of the implementation of new practices/processes. This might push the smartest (even if they are not the voices that were silenced, they are smart ;-) ) of the lot out of the organization - they'd always find a better job no matter how tight the market is. In knowledge industry, losing human capital can be a major setback, esp productive smart people sacrificed on the altar of experimenting with a fad to solve a problem that does not exist is a criminal waste.