Kanban Analytics part III: Distribution Chart

Very often we need to know how many tasks we have assigned to each person on the team, or we want to know whether we have more tasks in column A than column B. It could also be the case that we want to estimate the relative size of all tasks in one column vs. another. Not only that we want to know the number of tasks, but we may want to actually inspect these tasks and, later on, export the image.

Good news, ladies and gentlemen! The Distribution Chart that comes out of the box with Kanbanize can do all that and actually much more. Feel free to go through the entire article and see for yourself what a powerful tool we have ready to be used without ANY overhead. Prepare yourself for a very long article with a lot of information to digest, but believe us, it’s worth the read.

Kanban Task Distribution Chart1

Before we move on, let us cover the mechanics about how you can interact with the chart.

Kanbantaskdistribution2

  • The first thing to notice is the radio button group which controls the tasks you see on the chart. You can select to see tasks which are currently on the board or tasks that have been moved to the permanent archive. Please note that the panel on the right of each Kanban board is the temporary archive and tasks are only moved to the permanent archive via the archive tasks functionality.
  • The second control is the Dimension selection. You can choose between multiple dimensions like assignee, size, tag, etc.
  • The third control is the unit of measure, which is one of four (Count, Size, Elapsed time, Logged time). We will explain how that works later on.
  • The next one is a swimlane selector. If you have more than one swimlane on the board you can choose the one to be taken as a datasource for the chart.
  • The last thing to note is the column selector. When you are seeing tasks from the board and not from the permanent archive, you can enable or disable some columns. For example, you will often want to know something about the tasks in progress only. In such cases, you just uncheck all other columns and the chart will be updated accordingly.

One of the greatest features of the chart is that you can click on any part of the pie and see a table with the task details below it. Clicking on the icon in the first column of the table opens the task details, which is a very convenient way to review how things are going.

Kanbantaskdistributionchart3

So far so good! Now let’s start exploring the various use cases for the distribution chart and outline in which situations we might be finding them useful.

Scenario 1: Distribution by Count
This is the default and most frequently used unit of measure. As the name suggests, it slices the data by the number of tasks. This configuration is to be used when you want to answer questions like:

  • How many tasks we have in columns X, Y, Z
  • How many tasks are assigned to person X, Y, Z
  • How many tasks are bigger than X
  • How many tasks are high priority and are currently in progress
  • etc.

Distribution by columns
The distribution by columns shows how many tasks we have in each column. You can control which columns are visible from the chart configuration panel on the left. This variation of the chart is actually a different visualization of the Kanban board. If you have hundreds of tasks on the board this pie chart can give you a better comparative visualization of the size of each column.

Distribution by assignee
When aggregated by assignee, the chart simply shows how many tasks are assigned to each team member. You may want to look for the bigger pieces of the pie and try to figure out why these people have more things assigned to them. Try to balance the work across all team members and keep the values low – that is how you will keep your people more productive.

Distribution by size
If you put size labels on your tasks, this might be a very interesting representation for you. You can see whether you have predominantly smaller or bigger tasks. Our recommendation would be to have more smaller and just a few bigger tasks. It is easier to work on a bigger thing, but it is also harder to deliver it, so try to create flow by constantly releasing smaller chunks of value.

Distribution by type
Very often people categorize their tasks into types. Popular examples would be “defect”, “customer request”, “support request”, etc. Usually we want to work more on value-adding activities and spend less on non-value adding activities (defects). With this configuration of the chart you can track the distribution of work across the Kanban board.

Distribution by tag
If you put tags on your cards (e.g. customer, release, milestone, price, etc.) you can slice the data by the tag value. If you are keeping the customer in the tag field, then you can see how many tasks you have per customer.

Distribution by priority
This one is pretty clear – how many tasks we have of low, high and medium priority. The only thing to be careful about here is not to have too many tasks with high priority. Try to label something as important only when it really is.

Distribution by color
The colors in Kanbanize help us to visually indicate different characteristics of a task. To some of us this means priority (red is urgent, green is not), to others it means customer, third consider it size, etc. When you group the tasks by color you can map the result to your context and analyze the data as appropriate.

Scenario 2: Distribution by size
This configuration works in a very similar way as the count distribution, but it shows the accumulated size of tasks. In order to make use of this chart you should be using the “size” field in any of its forms (number, T-shirt size, Fibonacci sequence). The values of all size fields is summed and displayed on the chart. The relevant questions here would be (these are just examples):

  • How much work we have pending in columns X, Y, Z
  • How much work has person X, Y, Z done
  • How much high-priority work we have in our backlog
  • Do I have more work in value adding activities or not
  • etc.

To keep the article shorter than “The Lord of the Rings” we would take a look at the most popular use cases only.

Distribution by columns
This representation of the chart is valuable when you want to see how much work you have in the different columns. Sometimes the number of tasks may be misleading and this is how you double check what’s really happening at the board. For example, if you have 10 small tasks in one column and 3 very big tasks in another, you would be mistaken if you only take a look at the number of tasks. That is why you may want to check the distribution by size as well.

Distribution by assignee
Reviewing the distribution by assignee is also quite popular. If you have people working on a very big scope, they are likely to deliver slowly and your flow would be disrupted. Try to keep the scope for each person on the team small enough so that the result can be shipped in the shortest possible time (a few weeks would be fine in most cases).

Distribution by type
Another use case that is typically important is the distribution by type. Most teams implement some kind of work classification into types. For simplicity, we can assume we always have at least two types: “value adding” and “non value adding”. If you label your tasks with one of these, then with the size distribution by type you will always know how much value you have added and how much you have invested into fixing issues and some other crappy activities which gave you nothing in return.

ebook-blog-banner

Scenario 3: Distribution by logged time
For all of you who want to track your time we provide the log time functionality. We may not ourselves be the greatest fans of counting hours, but we do realize that a lot of people have to do it, therefore we make their lives easier. When you slice the data in the distribution chart by logged time, you present an accumulated view of how many hours have been logged in total for a given column, assignee, type, etc.

Distribution by columns
When you distribute the logged time by columns you see the total logged time for all tasks in the corresponding columns. If a task has been in three different columns and the logged time for each column has been 4 hours, you will see a total of 12 hours on the chart. If you want to inspect how much time has been logged per column, you can review the “Metrics” tab in the task details.

Distribution by assignee
This is probably the most frequently used scenario for the Distribution chart by logged time. You use it to see how much time each person has logged against the tasks they work on. With this configuration you can track how time is being logged and compare different people.

Distribution by type
Since you can directly associate the logged time per type with the amount of money you spend on each type of work, this is a very important view. If you find yourselves investing too much money into non-value adding work, you should probably investigate why that happens.

Scenario 4: Distribution by elapsed time
The charts about distribution of elapsed time are usually a superset of the charts about logged time. All data should be interpreted the same way as above with the difference that we have elapsed time instead of logged working hours. Please note that elapsed time is different from cycle time. The elapsed time is the absolute time that tasks have spent on the board and it does not take into account the “Cycle time configuration” which you can make from the project settings tab.

Conclusion
The distribution chart in Kanbanize is a powerful tool that can be used to digest huge amounts of data in a single glance. Typical use cases like “Logged time per assignee” or “Cumulative size by type” allow us to spot deficiencies in our process and optimize. This was a long long blog post, but we hope it was informative and that it will help you adopting our software.

Happy Kanbanizing and don’t forget to send us your feedback!

 

 

One thought on “Kanban Analytics part III: Distribution Chart

  1. Pingback: Kanban 101 - Tracking progress : Part 2 | Kanbanize Blog

Leave a Reply

Your email address will not be published. Required fields are marked *