Monte Carlo - When?
The When chart aims to tell you when you can expect the team to finish a specific number of assignment that you are yet to start working on.
Based on how you have performed in the past, the "When" chart can estimate how long it is gonna take you to get 100 items done in the future.
For example, you can use it if you’ve got a product update scheduled for March 30th and you wish to know how many features you can finish by then.
It has an identical structure to the Monte Carlo: How Many (Components) and works in the same way. The only structural difference concerns the significance of the percentiles that line the last chart in the view.
Unlike in the Monte Carlo: How Many chart, they are increasing from left to right, instead of decreasing. This refers to the fact that the further you go in time, the greater the certainty of the 100 work items upon which the simulation is based to be completed.
NOTE: It is important to note that if the characteristics of the team that generated the historical data changes (eg. new team member is added), the simulation produced from the data will no longer be valid.
Controls for this Char
Throughput Mode: if the line chart is not to your liking, there is a histogram view that may prove to be more convenient for you.-> use the Throughput Mode at the top of the Controls to define it.
Percentiles: select the Percentiles to activate the probabilistic view mode of the chart. The filter allows you to add/remove the percentiles from your simulation. For example, if you are interested to see where there is an 85% chance for the prediction to come true, just remove the check marks from the rest of the available percentiles to focus only on your target.
Simulation Input: use the Simulation Input in the Control Charts to define the future time frame for which you would like to generate a forecast.This is the only filter that is slightly different between the two charts.
The Monte Carlo: How Many simulation provides you with the ability to select a start and end date of your future prediction while in the Monte Carlo: When, you’re able to define a starting date and pick any number of tasks for which you want to generate a likely completion date.
Trials: You can run more than 1 million trials -> to simulate trials use the Simulate More Trials button.
Components: switch on/off the components you would like to see.
Filter for Current Chart: the last filter is designed to give you а different perspective by removing any card attribute of your choice from the prediction as well as different stages of the workflow. In combination with the filter for past time frame, it will allow you to make very specific predictions. For example, let’s say that you are about to launch a new marketing campaign and have already prepared the breakdown of the tasks that need to be completed. You’ll probably want to know when the content for the campaign will be ready. To find out, select the specific card type in both filters and the platform will exclude the rest from the simulation, thus giving you a more precise prediction based only on your previous content performance.
The Global Attribute Filter at the top of the menu gives you the ability to run a very specific simulation based on pre-selected attributes from your board setup. For example, you may include just the cards with high priority in the historical data and see how many cards (unfiltered) will be completed for the rest of the quarter.
*This filter is global for all premium analytics and will apply same filter criteria if you choose to switch between the different charts.
You can include and exclude different stages of your workflow from the historical data, to experiment with different possible conclusions. For example, you can remove the requested section of your workflow and predict how many cards can go from an activity column, such as Coding, to a queue column of the process (e.g. ready for code review) in a pre-selected period of time in order to anticipate how many cards will pile up if there is no one to process them.
Note: This article was originally published on the Kanbanize blog