How to increase team productivity several times
As often happens in life: a new manager comes and sets the task to increase the number of implemented features by 2 times for the next sprint. Developers, of course, try, work, stay for the weekend, carry out the set plan, but then for some reason everyone leaves.
Most people see productivity based on the paradigm: the more you produce, the more you do, the higher your productivity . But true productivity is a function of two elements: the result obtained and the resources expended. Maintaining a balance between outcome and resources is often quite difficult. However, this balance is the whole point. It leads to a balance between short-term and long-term.
Any manager, be it a development team leader or department manager, should take care to increase the productivity of their team. But how to do it wisely? Is it possible to increase the productivity of the team, for example, 10 times?
')
Today I will try to tell my point of view on this question. If you're interested, welcome under the cat!
First you need to understand what productivity is and how it is measured in your team. In this case, you can use completely different indicators. They can be quantitative and qualitative.
As an example of quantitative indicators one can take, for example, the number of features delivered to a client per unit of time. Someone is still evaluating the work of programmers in lines of code (yes, this still happens today), for someone it’s important the number of fixed errors, the number of autotests and so on.
To establish quality indicators need to find standards of relationships. For example, the current performance of the complex and the requirements for iron N. Then the metric will be an increase in the productivity of the product by 25% or an acceleration of the execution of long-term tasks for the same 25%.
A good metric is Cost per unit (cost per unit of production). A unit of production is everything that has at least some meaning for the user (functionality, error correction, performance enhancement, etc.). It can be measured in terms of per person, product, project, etc. These are all the metrics described above, but expressed in money.
For many, the important metric is Cycle time (change delivery time to the client). It's one thing when you roll out new features and changes every day, another thing is once a month or even less.
It is impossible to speak about the productivity of the team in isolation from the quality of the supplied product. Of course, you should have a whole set of metrics that evaluate the quality of your product. Let's give some examples of such metrics:
So how to increase team productivity?
We will look at the product as the result of the work of the team. Everyone knows the Pareto principle, from which it follows that 20% of the product's functionality covers 80% of user needs. The rest of the functionality is either very rarely used or not used at all. It is very important to spend the time of the team on the necessary and important features, so get rid of the old and unnecessary code, refactor, simplify the code and its support. In the future, this will greatly enhance the team’s effectiveness
It is incredibly easy to fall into the trap of activity, in the cycle of affairs and events, spending more and more effort on making some new useless features, changing the design, etc., in order to realize at once that nobody needs it . You can be a very busy person, not being effective.
Let's deal with the team . It's no secret that certain developers at times, 10 times more productive than their colleagues. Your job as a leader is to build a strong team and select only the best candidates. Hence the fact that you need to say goodbye to the frankly weak, reducing the productivity of the entire team members. Constantly ask yourself the question: would you have hired this person for this position, knowing what you know now? If not, he should not take it.
However, do not chop off the shoulder. There are cases when the productivity of an individual team member is low, but when he / she is in a team, the productivity of the whole team increases! It is important in a team to have a person who would raise the overall morale of the team. Let him even do less than the others, but he will rally the team and increase the overall result.
Consider the organizational and process points . You, as a manager, are required to follow the following process:
Removing once the bottleneck, it will come out in another place. By removing a new bottleneck, you will get it again, most likely on a smaller scale. At some point, you will realize that finding a bottle has become too difficult for you and eliminating it is more expensive than a bonus from the result. It is time for you to experiment in processes! Look for best practices, try to shift them to your team, adapt! Do not be afraid of failures, not all best practices take root in specific teams. It should draw conclusions and move on.
You need to try to automate everything for which automation in your team is reasonable. No one will argue that in the overwhelming number of projects one should use CI / CD to quickly deploy and deliver a new version of the product to the client. Autotest now does not use only a lazy leader. You yourself can and must come up with what is most sensible to automate for your particular team.
Well, and the final rule for managers and all who want to develop!
Get out of your comfort zone! Beware of "boiled frog" syndrome. They say that if you throw a frog in hot water, it will immediately jump out. But if you put the same frog in water at room temperature and gradually heat the water to a boil, the frog will not try to get out and eventually just boil. I do not know how true this bike is with respect to frogs, but something like this I occasionally observe among managers and employees. People tend to gradually get used to unacceptable things that would shock them if they saw them with fresh eyes.
Grow, grow, succeed!
Most people see productivity based on the paradigm: the more you produce, the more you do, the higher your productivity . But true productivity is a function of two elements: the result obtained and the resources expended. Maintaining a balance between outcome and resources is often quite difficult. However, this balance is the whole point. It leads to a balance between short-term and long-term.
Any manager, be it a development team leader or department manager, should take care to increase the productivity of their team. But how to do it wisely? Is it possible to increase the productivity of the team, for example, 10 times?
')
Today I will try to tell my point of view on this question. If you're interested, welcome under the cat!
First you need to understand what productivity is and how it is measured in your team. In this case, you can use completely different indicators. They can be quantitative and qualitative.
As an example of quantitative indicators one can take, for example, the number of features delivered to a client per unit of time. Someone is still evaluating the work of programmers in lines of code (yes, this still happens today), for someone it’s important the number of fixed errors, the number of autotests and so on.
To establish quality indicators need to find standards of relationships. For example, the current performance of the complex and the requirements for iron N. Then the metric will be an increase in the productivity of the product by 25% or an acceleration of the execution of long-term tasks for the same 25%.
A good metric is Cost per unit (cost per unit of production). A unit of production is everything that has at least some meaning for the user (functionality, error correction, performance enhancement, etc.). It can be measured in terms of per person, product, project, etc. These are all the metrics described above, but expressed in money.
For many, the important metric is Cycle time (change delivery time to the client). It's one thing when you roll out new features and changes every day, another thing is once a month or even less.
It is impossible to speak about the productivity of the team in isolation from the quality of the supplied product. Of course, you should have a whole set of metrics that evaluate the quality of your product. Let's give some examples of such metrics:
- the number of errors found by clients during the period after the delivery of the new version (example of external metrics)
- the number of errors found in the testing department after the transfer of functionality or correction in verification (an example of an internal metric).
So how to increase team productivity?
We will look at the product as the result of the work of the team. Everyone knows the Pareto principle, from which it follows that 20% of the product's functionality covers 80% of user needs. The rest of the functionality is either very rarely used or not used at all. It is very important to spend the time of the team on the necessary and important features, so get rid of the old and unnecessary code, refactor, simplify the code and its support. In the future, this will greatly enhance the team’s effectiveness
It is incredibly easy to fall into the trap of activity, in the cycle of affairs and events, spending more and more effort on making some new useless features, changing the design, etc., in order to realize at once that nobody needs it . You can be a very busy person, not being effective.
Let's deal with the team . It's no secret that certain developers at times, 10 times more productive than their colleagues. Your job as a leader is to build a strong team and select only the best candidates. Hence the fact that you need to say goodbye to the frankly weak, reducing the productivity of the entire team members. Constantly ask yourself the question: would you have hired this person for this position, knowing what you know now? If not, he should not take it.
However, do not chop off the shoulder. There are cases when the productivity of an individual team member is low, but when he / she is in a team, the productivity of the whole team increases! It is important in a team to have a person who would raise the overall morale of the team. Let him even do less than the others, but he will rally the team and increase the overall result.
Consider the organizational and process points . You, as a manager, are required to follow the following process:
- eliminate the bottleneck in your current processes and team,
- set change feedback,
- repeat this process an infinite number of times.
Removing once the bottleneck, it will come out in another place. By removing a new bottleneck, you will get it again, most likely on a smaller scale. At some point, you will realize that finding a bottle has become too difficult for you and eliminating it is more expensive than a bonus from the result. It is time for you to experiment in processes! Look for best practices, try to shift them to your team, adapt! Do not be afraid of failures, not all best practices take root in specific teams. It should draw conclusions and move on.
You need to try to automate everything for which automation in your team is reasonable. No one will argue that in the overwhelming number of projects one should use CI / CD to quickly deploy and deliver a new version of the product to the client. Autotest now does not use only a lazy leader. You yourself can and must come up with what is most sensible to automate for your particular team.
Well, and the final rule for managers and all who want to develop!
Get out of your comfort zone! Beware of "boiled frog" syndrome. They say that if you throw a frog in hot water, it will immediately jump out. But if you put the same frog in water at room temperature and gradually heat the water to a boil, the frog will not try to get out and eventually just boil. I do not know how true this bike is with respect to frogs, but something like this I occasionally observe among managers and employees. People tend to gradually get used to unacceptable things that would shock them if they saw them with fresh eyes.
Grow, grow, succeed!
Source: https://habr.com/ru/post/453026/
All Articles