Dashboards can consist of visualizations that are based on different datasets which have some common parameters, such as company or country. In this case, you can create one filter per dashboard that will control different visualizations.
This article describes how to create such a filter and covers the following questions:
- How to add a common filter for several visualizations based on different datasets?
- Which parameters can be used as a common filter?
- Changing the list of filter elements: common across all datasets vs. all elements.
In general, a filter can be added to any dashboard to control for some parameters of the visualizations, for example, company or indicator.
However, there are cases when a dashboard contains visualizations built on different datasets which have some common parameter and you need to add a single filter for these visualizations. For example, you may have a "company monitor" dashboard that presents information about a company from many different sources and you want to add a filter by company so that a user can choose a company from the list and see charts for this company.
How to add a common filter for several visualizations based on different datasets?
Once you've created a dashboard and added visualizations to it, click "Parametrize" button in the top menu.
Then, click a chart that you want to set a filter for and select one of parameters that connect a chart to the header of the dashboard. Do the same for the next visualization. Note, that you need to select a paratemter that is common across visualizations: it can be either a region or a company.
As shown on the screenshot above, there are two possible ways to filter a visualization by company:
- Clicking a purple box will create an individual filter for this specific visualization. If you click another visualization, this box will be of another color indicating that it will create a separate filter for each visualization.
- Clicking a red one will create a single filter for the two visualizations. If you click another visualization, this box will still be red indicating that this will create a common filter for all visualizations with this parameter.
Click "Apply" button to save the filter or click the parameter name at the top of the dashboard to give a name to the filter.
Once everything is set up, you'll end up having one drop-down list at the top of the dashboard that will filter several visualizations.
Which parameters can be used as a common filter?
You may have many different parameters that are common across datasets. For example, different datasets may have "Indicator" column with the same set of variables. However, if you try to filter visualizations built on different datasets by "Indicator", you will have multiple filters (drop-down lists). There are only two types of parameters that allow to create a single filter for visualizations sourced from different datasets. These are region and company. These two parameters have a central catalog in the platform that allows one to automatically merge different datasets by these parameters.
The system automatically recognizes region and company parameters in datasets when they are onboarded to the data hub. For the company dimension to be recognized correctly, the dataset should have two columns: company and ticker (symbol on a stock exchange).
To check if dataset parameters can be used as common filters, open a dataset, click "More actions" menu and select "Edit details" (note that this feature is available only to owners of a dataset, DataHub managers, or users with "Edit" access to the dataset). Scroll down to the "Dimensions" sections and check if a parameter have "Company" or "Region" in the "Special" column. If it does, then you can use this parameter to create a single filter for multiple visualizations built on different datasets with such parameter.
Changing the list of filter elements: common across all datasets vs. all elements
Once you've created a filter, you can change the list of parameter values to choose from. To do it, click the box with parameter name at the top of the dashboard. In the dialog that appears, you'll be able to select one of three options:
- Common across all datasets;
- Display all; and
- Selected members only.
If you select Common across all datasets, then it will show only companies that are present in all datasets connected to the filter. So, some companies that are present in one dataset but absent in another one will not appear in the drop-down list.
If you select Display all, then all parameter values from all datasets will be present on the list. In this case, if you select a company that is present in one dataset but is absent in another one, some visualizations may become empty.
If you select Selected members only then you will be able to manually pick those companies that you want to appear on the list.