Crunch dashboards can be created and edited in a variety of ways to suit different workflows. Here, we’ll go through those different workflows to give you ideas on how you can make best use of Crunch dashboards to suit your situation.
Entirely in the GUI
Crunch dashboards can be created by saving analyses to a dashboard container in Tables & Graphs mode, and then laid out via simple drag-and-drop and customized via edit panels. No scripting is required to create a dashboard in this way, and no level of technical familiarity is required.
In the GUI but with time-saving using duplication and script edits
Dashboards commonly have repeated or very similar elements, and it can be a time-saver to create and customize a single tab or tile in the GUI and then duplicate that tab/tile as many times as needed. Some very basic script editing then allows you to modify those duplicates as needed - for example, to make them focused on different countries, or different brands, or different markets etc.
In the GUI, then duplicated to other datasets/Views
Syndicated data products usually involve sharing separate datasets (or we would recommend Dataset Views instead) to the subscribing clients where each should have its own dashboard. These dashboards are commonly complete duplicates of each other, so significant time can be saved by creating whole new dashboards in those datasets or Views from the script of the original. If modifications are required, they can be made either by tweaking the script or by using the GUI. Dashboards created via script look and act just like those created in the GUI, so you can make further edits to a scripted dashboard in the GUI just as you would have for a manually created dashboard.
A dashboard created entirely from script in an automated pipeline
Want to have a dashboard on every Crunch dataset by default? The script that defines a dashboard can be generated programmatically and is therefore ideal for building into an automation pipeline. The Crunch Automation commands that create a dashboard can be triggered via the API, meaning that a dashboard which follows certain standardization rules (e.g. all variables in the dataset get their own frequency distribution graph tile) can be generated and delivered to the researcher at the same time as the rest of the dataset.