Applies to Write-Back Server and Write-Back Cloud

Question

How to create a forecast with Write-Back applicable to a relative period?

Forecast is an important management process and every main process in an organization tends to be based on some prediction to support them. When doing forecasting we typically want to focus on the periods that really matter, and this means for instance filling the forecast for the upcoming 12 months or 4 quarters. This is a sliding window forecast in the sense that the forecasting months or quarters are relative to the current date while keeping a consistent length. 

Answer

The attached workbook has the full solution implemented and is using Write-Back public demo environment, for more information on using it check our Public Demo instruction page.

Related Links & Attachments

Forecasting    Write-Back Configuration    Connecting to the Dataset    Sliding Window Forecast.twbx

Step 1: Create an Anchor Worksheet

As with all Write-Back configurations, it is necessary to create a visualization that will be used as a source for data submission. This source visualization must contain a time dimension. As it is a sliding window forecast, where it is possible to have several configuration versions, it is imperative that the visualization contains every month.

Open

Create an Anchor Worksheet

Step 2: Set Up Write-Back for Forecast Configuration

Creating datasets on Write-Back is really easy so we can create a dataset, which is managed by forecast process administrators, allowing to configure how the actual forecast is going to take place.  You can see this is a meta dataset which enables establish rule for the actual business process, in this case the forecasting. The main goal is to associate a forecast version with the corresponding time periods (months) and the year for each of them. 

Drag a new Write-Back extension instance to the dashboard pane. This instance will be linked to the view created in Step 1.

After selecting the .trex file, you need to change the interface to tabular mode. This change will allow you to associate a value to each month, in this case, a year. The Next step is to create the Input Fields. You need to create a Input Field that will allow you to specify which year the month (month) belongs to (year). Additionally, create a field that allows you to identify the version of the configuration (version).

Your configuration should look like this:

• Month on Columns

• Year on Rows

• Version on the Main Form

• One Source Field on Configure Fields

Step 3: Submit Forecast Configuration

Bearing in mind that we are dealing with a sliding window forecast, where the time periods are relative, a configuration is required for each of these periods. It is at this point that Write-Back takes preponderance and allows the management of these configurations in a simple way. With the data set created in the previous step, it is now necessary to submit a form to obtain a forecast configuration. To do this, we need to associate each month with a year and a version.

Forecast is looking at the future so typically you will be looking at subsequent periods of the current year and the next one.  Let's imagine that we are at the end of the 3rd quarter and we have to create a forecast configuration for the next four quarters. In the Write-Back form, we have to associate October, November and December with the current year and the remaining months with the next year. 

To submit a forecast configuration, select a mark on the anchor sheet and insert the year value for each month.

With this configuration it is possible to create several forecast versions without the need to reconfigure the extension.

Step 4: Create a Forecast Configuration View

Given that this type of dashboard is used across several time windows, meta configuration, you can implement a solution to navigate and select among the various forecast configurations. If you want to take it further you can even summarize forecast submissions and using the dashboard as the central place for forecasting process managers. 

As the data entered through Write-Back is written directly in a database of your choice, it makes access to that data very easy, you just need to set up a new connection and voila. You now have access to the data set that has the forecast configurations. Don't forget to filter Write-Back table by IsActive set to True to ensure you only get the latest snapshot provided by the users for each record. 

With the connection to that data set ready, you need to build a visualization that allows you to navigate and select the forecast configuration that you want.

If you look at the table structure, you will notice that the year and month are in separate columns. It is necessary to create a calculated field to combine these fields to create a valid date field.

With the date field created, build a view similar to the following:

Open

Create a Forecast Configuration View

Step 5: Set Up Write-Back for Forecast Submission

At this point, you are able to submit forecast configurations and navigate between them. The only thing missing now is to build a mechanism that allows the users to enter values for the actual forecasting. If you are looking to segregate roles between forecasting submitters and administrators this can be placed on a separate dashboard and when published to the server they can even have different permissions allowing that each user can only view the dashboard for his role. 

Similar to step 2, we can use Write-Back to do this. The view created in Step 4 will be the anchor point for this instance.

Drag a new extension instance to the dashboard pane. Change the UI to tabular view and create two Input Fields, being "Sales Forecast" one of them. Make sure that you include the Version field in the form as well, as this is the field that identifies a forecast configuration.

Your configuration should look like this:

• Year and Month on Columns

• Sales Forecast on Rows

• One Input Field on the Main Form - e.g. Comments

• Version  on Configure Fields

Step 6: Submit Forecast

After you choose the forecast configuration that suits your needs, it's time to fill it up with forecast values. To do that select one mark on the Forecast Configuration view and click on Submit Forecast.

Enter the forecast values for each month and click Submit.

Step 7: Create a Combined View with Real and Forecast Data

The purpose of the forecast is to ensure that the organization meets the goals set out in the initial budget. To ensure that these goals are met, it is necessary to regularly compare the real and what was foreseen so that, if necessary, adjustments can be made.

This could be achieved by creating an integrated view comparing the real with the forecast.

You need to create a new connection to the database in order to access the data. After blending the data, you will have a visualization similar to this:

Step 8: Create a Version Comparison View

Sometimes adjustments need to be made and this will end up having some influence on the forecast. A good way to assess these impacts is by comparing forecast versions.