Using Kindor to answer engineering productivity questions (Part 1)

Here's a guide of best practices that will allow you and your team to start answering questions that usually arise as you start gathering efficiency and productivity metrics.

This post is written in a way that you can easily identify the challenge you might be facing and the recommended actions to start measuring and improving the most relevant metrics for this challenge.

Challenge:  

"I want to make sure that the team is working according to expectations."

Kindor provides a workload score that lets you assess whether your teams and contributors are having the activity according to their role.

This score is calculated based on each organization's role and it is an excellent indicator to discover activity inconsistencies across peers with the same role in the company.

Common scenarios or questions that can be answered with Kindor:

• How long does it take for a new hire to reach productivity levels similar to their peers?

  
  

• Is there a contributor with significantly lower activity than her peers?

  
  

• Are people with the same title and seniority doing similar work?

Things to pay attention to when using this metric:

• Contributor title and seniority need to be correctly defined for every contributor tracked by Kindor. Failing to do so, could result in comparing people erroneously.

  
  

• As with many other metrics in Kindor, make sure that contributors are registering the work they perform, like completing tasks or tickets in the project management system and well as using Pull Requests to push code to production or intermediate branches.

Observations

• This metric is based on activity medians calculated over a long period of time. The score will only be considerably affected when the contributor has had low activity for a prolonged period of time.

  
  

• If the contributor is the only person with a specific title, he will be compared against himself. This will let you identify time periods where the contributor reduced his activity.

Challenge:  

"I want to evaluate how are delivery has been evolving over time."

Kindor lets you evaluate delivery metrics like number of deployments per day; task, tickets and story points completed per day, week or month; Pull Requests opened and merged in a specific period of time, etc.

You can evaluate this for the whole tech organization or based on a specific team or division.

Common scenarios or questions that can be answered with Kindor:

• Did our delivery increased after growing the team?

  
  

• Which are our most productive days, how does this month compares to the previous one?

  
  

• We went through a team restructure, how did this affect our delivery?

Things to pay attention to when using this type of metrics:

While these metrics can be very useful, they should never be the only metrics to track, since this could give a false perception of progress. It is always recommended to balance this with other relevant metrics that measure quality and or team satisfaction, to avoid burnout or sacrificing quality over speed.

Challenge:

"I want to understand where my team is focusing on and what are the main initiatives we are tackling."

Having a lot of activity main not mean anything unless it is having an impact. Understanding your team focus and the things they are prioritizing will help you not only evaluate the output, but the outcome.

Common scenarios or questions that can be answered with Kindor:

• How much time or effort are we spending on new capabilities?

  
  

• What are the most expensive initiatives we are currently tackling in terms of effort or time?

  
  

• Are we spending too much time fixing bugs?

Things to pay attention to when measuring this:

• Make sure you have a reliable way in your project management tool to register your main initiatives. We recommend using Epics and associate tasks to them to track all the related work.

  
  

• Try to follow a naming conventions or use labels on your Pull Requests, so that Kindor's engine can easily detect the work type performed in each PR.

  
  

• Differentiate Epics related to Feature Development and Epics related to maintenance, tech debt or bug fixing work. We recommend using labels in your Epics to easily identify the work type.

  
  

• When possible, use story points to evaluate tasks effort. This helps to get a more reliable measurement in terms of effort dedicated to a particular initiative.

Challenge:

I want to have a better understanding of my team's capacity and how reliable my team is when committing to deadlines

Another important thing to measure while evaluating a tech team performance, is how reliable they are in terms of delivering features on time.

Probably one of the biggest frustrations for business leaders is that their tech teams are usually late on their deliveries. While this is a common scenario, in order to get better at time estimations, you need to objectively understand the real capacity of your team, the things that prevent them from delivering on time and their historical pattern, to have better ETA projections.

Common scenarios or questions that can be answered with Kindor:

• What percentage of the committed features for this quarter were delivered on time?

  
  

• Are we prioritizing planned work or do we have too much unplanned work that prevents us from delivering on our promises?

  
  

• How many tasks or story points are we able to deliver in a specific period of time?

Things to pay attention to when measuring this:

• To start tracking reliability or "say do ratio" metrics, you need to define deadlines. This can be done in multiple ways:

  
  

  • Defining deadlines for epics or large initiatives.

  • Work under sprints with a pre-defined time period and scope.

  • Define deadlines at the task level.

    
    

• Make sure your tasks are being marked as completed as they are finished, try to avoid marking everything as complete at the last day of the sprint or forgetting to update an Epic status after it was completed.