Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Reverted from v. 10

📽️ Using Data Rollups

Rollups is a simple feature that allows you to roll data up on a higher level of aggregation. It gives us the capability to take the data source or data-mart which exists within Pricefx and apply further aggregation to get only specific information we need.

A great feature of Rollups is that it can be used as a data source for Analytics too. 

We access Rollups on the Analytics Menu.

Widget Connector
overlayyoutube
_templatecom/atlassian/confluence/extra/widgetconnector/templates/youtube.vm
width600px
urlhttps://www.youtube.com/watch?v=a3XfTmmNqFg
height300px

Visualisation

When you navigate to Analytics, specifically in the rollups section, you'll notice tiny input tables for entering group buys and measures to gather data. Previously, navigating and viewing all the labels and details were slightly difficult. To address this, with the new release, two improvements were made.

Firstly, you can now expand the left side area to make it larger.

Secondly, we've enabled advanced settings for the table, allowing you to freeze, sort, change the order, and perform other actions on columns.

Image Added

🗒️ How To

Data Rollups and How to Use Them

Info

All our how-to guides and video tutorials contain screenshot of our demo system and the information that you see will differ depending on your organizational configuration and the version of Pricefx software that you are running.


This How-To Guide explores the rollups feature in Pricefx Analytics.

Rollups is a simple feature that allows you to roll data up to a higher level of aggregation. It gives you the capability to take the data source or datamart which exists within Pricefx and aggregate information in there.

It creates a data table specific to what you're looking for, and a great feature of Rollups is that it can be used as a data source for Analytics too.

Expand
titleCreating a New Rollup

Create a New Rollup

We access Rollups on the Analytics Menu and we can then create the rollup.

Image Added

So for example, if you want to see a quick overview of all your information in order to get an understanding of the country in which you are selling the most, you could build up a chart but then you'd have to work through all the data fields in the data table. Instead, you can create a rollup, that allows you to apply specific groupings and measurements based on only the information you need.

Image Added

So here you'll go ahead and select the data source, which for this example is Standard Sales Data and you want to group information by the country and maybe you also want to group information by its customer segment and you can then define the measurement so for example in this case you want to see what the quantity you have sold in there so you can select based quantity. You could also filter by specific products or customers, or add generic filters if you want it, but for this example, leave those settings to their default.

Image Added

You can check how it will be aggregated, so whether it's a sum, if it's a contribution or you can also enter a function. Let's say you want to see an average instead of a sum, so select that option from the drop-down button and then click on the apply and refresh button and you basically get a summary over all the different countries you’re selling in and the different customer segments for each country.

Image Added

You can also use the pivot function, so in this example, you'll pivot the country group, and when you apply and refresh you have them as an additional dimension.

Image Added

But you can see that it's a little difficult to read the column titles, so let's click on the customer segments column, click on the options, and AutoFit All Columns. 

Image Added

So now you can clearly see the customer segments in the first column and the Base Quantity by country across the remaining columns.

Image Added

You can only pivot by a single grouping at a time, but whereas Excel is limited to 1 million lines of data, Pricefx can handle a virtually unlimited amount of data and the processing is far easier.

So let's remove the pivot and apply and refresh.

So now that you have the data the way you want it, you can export it to Excel or CSV for use outside the system, you can save the Rollup to come back to later, or you can save it for use as a data source in the analytics charts.

Image Added

So let's save this Rollup as SSDCustomerVolumeAVG and you can decide to keep it private to use yourself, public for anyone to use, or Open for all (View and Edit) which will allow other users to view and make changes to it.

 For now, keep it private.

Image Added

This functionality is intensively used for creating dashboard items, and you'll quickly demonstrate how to access the data source to create a chart, and if you want to add elements to a dashboard for easy access, you can view our Dashboard tutorials for creating or customizing new dashboards.

 But for now, let's refresh the browser and head over to create a chart.

Image Added

On the Analytics menu, click on Data Analyzer, and let's create a Heatmap. In the Data Source field select your SSDCustomerVolumeAVG data, place Country on the X-Axis and Customer Segment on the Y-Axis, and for the Color Axis, use Base Qty (AVG), then apply and refresh and you can now see the Heatmap displayed.

Image Added

If you want to use this chart in a dashboard you can now save it for use later on.

Image Added
Expand
titleEditing a Rollup

Edit a Rollup

If you realize that you're not happy with the data displayed as averages, you can make some changes to the rollup. You can head back to the Rollups option on the Analytics menu and load the rollup to edit it.

Image AddedImage Added

You can now update the aggregation to Sum, instead of average.

Image Added

If you want to exclude a few countries from the rollup, you could click on the Filter button and then create a rule to display countries that are not Australia or USA, for example.

Image AddedImage Added

Click apply and you can now see that both countries have been excluded from the data. However they've only been excluded in this view and not from the underlying data, so if you want to exclude them in a chart you'd have to add filters in the chart to do that.

Image Added

You do not need to save the Rollup under a new name. Let's call it SSDCustomerVolumeSUM and keep it private.

Image AddedImage Added

Now if you refresh the browser and head back to the heatmap, you can recreate the heatmap and see the sum of the base quantity displayed instead of the average. If you want to exclude the countries, you can now add a generic filter with the necessary conditions, place the Country on the X-Axis, the Customer Segment on the Y-Axis, and Base Qty on the Color Axis, then apply and refresh. 

Image AddedImage Added

Now if you wanted to add this to a dashboard for easy access, you simply click on the options, click Add to Dashboard and in the dialogue window, you can select the dashboard, enter the label that will appear in the dashboard, and then select from the range of filter options and submit.

Image Added

Here we'll select My Sales Dashboard, call it Customer Volume Sum and hit submit.

Image Added

If you then navigate to the Dashboards, you can select My Sales Dashboard from the list.

Image Added

Now you can see the Heatmap with the label you've given it and you can then stretch/enlarge it from the corner of the chart, to see a better view. You can access the options to view it in a dialog window, you can export it, hide it, etc…

Image Added

So, that covers the main feature of Rollups in Pricefx Analytics. You should now be comfortable using Rollups to aggregate information, export it, and save the rollup for use in creating charts and adding them to dashboards.

If you need more information why not watch the accompanying video tutorial.