Creating a hierarchy
- Hierarchy types
- Creating a hierarchy
- Editing a hierarchy
- Rename the root level of a hierarchy
- To find out more...
It is possible to create data hierarchies on Dimension columns. The different levels of hierarchy can then be explored in the graphs using these dimensions.
For example, on a dimension containing a list of cities, you can create a first level of hierarchy grouping cities into departments, then a second level grouping departments into regions.
On a Dimension (time) column, DigDash Enterprise automatically creates three date hierarchies:
- The " Date" hierarchy with three levels: Day (level 0), Month (level 1) and Year (level 2),
- The "Month Year " hierarchy with 4 levels: Day (level 0), Month (level 1) Quarter (level 2) Year (level 3).
- The "Week Year " hierarchy with two levels: Week (level 0) and Year (level 1).
For example, on a graph displaying sales, clicking on one of the years will display the sales for each quarter of the selected year (you must first indicate that you wish to navigate the hierarchies: see the page Configure interactions ) :
Column chart showing sales by year :
Column chart after clicking on the year 2018. Sales are displayed for each quarter of 2018.
For a Dimension (geographical) column, DigDash Enterprise proposes the geographical hierarchy with the best match among those available (hierarchies corresponding to existing maps in the Map Manager). In the case of matches with the same score, the first is chosen.
Hierarchy types
Different types of hierarchy are available. The type of hierarchy you can create depends on the type of dimension selected: discrete or continuous.
- Discrete dimension:
- Automatic grouping : group hierarchy based on other dimensions in the data model. Each level corresponds to a dimension in the data model.
- Manual grouping : group hierarchy in which each level value is entered manually.
- Child-father grouping : hierarchy based on two dimensions with tree relationships.
- Intelligent grouping (clustering): grouping the members of a dimension according to certain measures.
- Continuous dimension :
- Hierarchy : hierarchy whose definition is based on formulas. For example, age groups.
- Common to discrete and continuous dimensions:
- Drill path : path used to structure the user's navigation in dimensions which have no particular link. For example, Year/Store/Product.
- Simple set : set used to group elements according to a specific need. It can be used as a filter in the dashboard.
- Combined set : combination of simple sets.
Creating a hierarchy
To create a hierarchy :
- In the list of columns, select the dimension for which you want to create a hierarchy.
- In the Hierarchies section of the properties panel, click on the + button and then, in the pop-up menu which appears, select the type of hierarchy you want to create.
- Consult the pages dedicated to the different types of hierarchies by clicking on one of the links in the list of hierarchy types above.
- By default, the root level containing all the members of the dimension is displayed for the hierarchy, for example, in the dashboard filters. If you do not wish to display this root level, uncheck the Show root level box.
Editing a hierarchy
To edit an existing hierarchy:
- Select the hierarchy in the Hierarchies section of the properties panel.
- Click the Edit button
. - Modify the hierarchy as required. See the page dedicated to each type of hierarchy for more details.
In the case of an automatically generated geographical hierarchy, you can edit it as described below.
Automatically generated geographical hierarchy
| Parameter | Description |
|---|---|
| Group | Name of the hierarchy group. Click the |
| Renaming undefined groups | Groups that do not match in the hierarchy are grouped in Other. You can rename this group as required. Click the |
| Disable approximation | By default, when there is no exact match between a member of the dimension and an element of the geographical hierarchy, a match is made with the closest element. This is an approximation. Checking the Disable approximation box allows only exact matches to be retained. Members without an exact match are then added to undefined groups (Other by default). |
| Manual selection | Check this box to select a hierarchy other than the one found automatically:
In the case of a score of less than 100%, you can click the View report link to view details of the exact matches or matches not found. See the Geographic Data Report paragraph below for more details. |
Geographic data report
When using a geographic dimension with an automatic hierarchy, it is possible to encounter association problems between the members of the dimension and the geographic levels of the map.
There may be multiple causes: dimension with missing members, incorrectly spelled members, incorrect map, inappropriate map hierarchy, etc.
The geographic data report displays details of the matches between the data in the data model and the geographic data in the map, with the data found/not found (via the approximate or exact search).
The Level data tab allows you to view all the geographic level data for the selected map.
You can:
- check the exact wording expected by the map
- identify differences in spelling or naming
- confirm that the correct level of geographical hierarchy is being used...
Example
In the report above, there is no match found for the New Scotia region.
To find out why, go to the Level data tab and search for Scotia.
The search returns Nova Scotia.
Rename the root level of a hierarchy
If the Show root level box is ticked, this level is displayed by default under the name Root.
You can change this name to make the display more relevant to dashboard users in particular. To do this:
- Enter the name of your choice in the Root level name field.
Click the
button if you want to translate the name into different languages. See the page Translating the name of an object page for more details.
For example, for the Geographic hierarchy, we enter County as the Root Level Name.
When the hierarchy is selected in the Studio, in the Dashboard Editor or on the dashboard itself, County is then displayed instead of Root.
➡ 