Relations Tab: 06. Visualizing Relations Data

back to list of topics           see also   
 
 
 
Use the Visualizer to gather additional insights on a set of objects from the Relations Tab. Any set of objects can be loaded into the Visualizer from the Relations Tab, from a list of Brands most over-indexed in skincare conversations, the Location Types that take up the largest share of hard seltzer conversations, and more. To visualize a more curated list of objects (whether they be objects across different object types, objects of particular relevance to a brand, etc.) create a worklist. Once created, choose ‘Current Worklist’ from the drop-down menu next to Posts and Users (image 1).
 
 

To open the Visualizer window, right-click an object and select ‘Visualizer’. This will load the selected object and all objects above it into the Visualizer, unless there are more than 50 objects in which case the top 50 will be loaded. If the object selected is within the top 10, the top 10 will be loaded.

From the Visualizer window, select custom metrics for the graph using the menus at the top right of the window, or choose between two presets by right-clicking on the graph and selecting ‘Filter Preset’ (image 2).

 
 

Preset 1, Frequency vs. Change in Frequency: Sets the y-axis to Raw Penetration and the x-axis to Raw Penetration - rel chng 12 mn. This option is ideal for analyzing objects’ shares of focus object conversations and how they have changed relative to their shares the year before.

Preset 2, Unique Identifiers, Freq & Sgnfnc: Sets the y-axis to Penetration Pctl and the x-axis to Raw Penetration. This option is ideal for analyzing objects’ shares of focus object conversations and simultaneously comparing these shares to shares of conversations for all other similarly categorized objects.

Outside of the presets, the custom metrics and assortment of graph types at the top right of the Visualizer screen offer a multitude of ways to visualize and represent data. The presets provide an accessible entrypoint, but utilizing additional features may help to paint a more robust picture of the data.