Groups represent a powerful yet distinctly different approach to data organization. While sets focus on filtering, groups provide a method for subdividing a single dimension into meaningful categories that better serve your analytical needs.
This functionality addresses a common limitation in enterprise data: the need to reorganize information into logical categories that don't exist in your original dataset. Perhaps you need to segment products by geographical regions, classify items by strategic importance, or group suppliers by performance tiers. Groups give you the flexibility to create these custom categorizations on demand.
The key distinction is that groups create new organizational structures from existing data. For instance, you might want to create a group for premium products, another for budget items, or categorize all products from specific suppliers into strategic partnerships. In this exercise, we'll focus on grouping subcategories to demonstrate the practical application of this feature.
Let's begin by examining our subcategory data. Navigate to the subcategory field, click the dropdown menu, and select "describe" to view the available data points.
Before we proceed with the technical steps, consider this analytical challenge: looking at our current data structure, can you identify a logical way to organize these 17 subcategories into two distinct, meaningful groups? This kind of strategic thinking is essential for effective data visualization—the technical execution is only as valuable as the logical framework behind it.
Upon examining our product subcategories, a clear pattern emerges. These 17 items can be logically divided into two primary categories: electrical and non-electrical products. This classification creates a meaningful business distinction that can drive strategic insights about product mix, supplier relationships, and market positioning.
While some items clearly fall into one category or the other, you may encounter edge cases that require business judgment. This is typical in real-world data analysis, where perfect categorization isn't always possible, but meaningful approximation drives valuable insights.
Similar to sets, creating a group generates a persistent element in your workspace. However, instead of the two-ring icon that represents sets, groups display with their own distinct visual identifier—a paperclip icon that symbolizes the binding together of related elements.
Our objective is clear: create electrical versus non-electrical categories for strategic analysis. This manual process requires deliberate decision-making about each item's classification.
It's crucial to understand that grouping is an intentionally manual process. Unlike automated categorization tools, this approach ensures that business logic and domain expertise drive the classification decisions rather than algorithmic assumptions.
Now let's execute the grouping process. Right-click on the subcategory field, navigate to "Create," and select "Group." This opens the group creation interface where you'll define your new organizational structure.
In the dialog box, start by naming your overall field. I'll call this "Power Type"—a descriptive name that clearly indicates the classification criteria. This becomes the new dimension name that will appear in your data model.
Next, create your first group by clicking "Group" and naming it "Electrical." Notice that accessories appear in this initial grouping location—this is simply where we're starting the creation process, not a classification decision.
Create the second group by clicking "Group" again and naming it "Non-Electrical." Chairs clearly belong in this category, representing our first definitive classification.
Now begin the classification process systematically. Accessories are non-electrical, so drag them into the appropriate group. Machines clearly qualify as electrical—you can either drag and drop them or use the dropdown menu to select "Electrical" for streamlined assignment.
The interface supports multiple selection methods for efficiency. Use Shift+click to select contiguous ranges or Ctrl+click to select non-contiguous items. For example, select art, binders, and bookcases together, then assign them to non-electrical. You can continue holding Ctrl to add envelopes, fasteners, furnishings, and labels to your selection before making the group assignment.
Continue this process methodically: appliances and copiers go to electrical, paper products to non-electrical, phones to electrical, and supplies and tables to non-electrical. Pay careful attention to ensure no items are left unassigned—orphaned items automatically create their own categories, which can disrupt your intended classification structure.
The paperclip icon represents the binding nature of groups—like using a paperclip to organize related documents. Click "OK" to finalize your Power Type grouping and add it to your available dimensions.
To demonstrate the analytical value, create a new worksheet called "Power Type Analysis." Drag Sales to the Columns shelf and Power Type to the Rows shelf, then expand to "Entire View" to see the complete breakdown. Convert this to a bar chart to visualize the distribution between electrical and non-electrical product sales.
This reveals the power of custom groupings: the electrical/non-electrical distinction didn't exist in your original data, but now you can analyze your entire business through this strategic lens. This capability proves invaluable for executive reporting, strategic planning, and market analysis.
To add more granular insight, consider incorporating the underlying subcategories into your visualization. Adding subcategory to the Color shelf shows the composition of each group, while adding it to the Text label provides clear identification of individual components. This layered approach delivers both high-level strategic insight and detailed operational visibility.
The automatic formatting adjusts font colors based on background colors, ensuring readability across your visualization. You can highlight specific subcategories by clicking on individual elements, making this approach highly interactive for presentation and exploration purposes.
An important consideration for ongoing data management: when new subcategories are added to your dataset, they won't automatically join existing groups. This manual assignment requirement ensures data governance and prevents misclassification, but requires ongoing maintenance as your data evolves.
Let me demonstrate this with an additional example. Create a new group called "Alpha" to show alphabetical organization. Select accessories and click "Group," then name this group "A-M" to represent the first half of the alphabet. Use Shift+click to select multiple items for batch assignment to this group.
Create a second group called "N-Z" for the remaining items. You can organize these by dragging items between groups or using the dropdown assignment method. This alphabetical approach might seem arbitrary, but it demonstrates how groups can organize data along any logical dimension that serves your analytical needs.
The resulting visualization becomes increasingly powerful when you layer multiple dimensions. Adding subcategory to color creates a detailed breakdown, while adding it to text labels ensures clear identification. This multi-layered approach transforms simple bar charts into rich, interactive analytical tools.
My preferred analogy for understanding groups is Zoom breakout rooms. Just as meeting participants must be assigned to specific breakout rooms (with unassigned participants remaining in the main room), all data elements must be explicitly placed in groups or they create their own default category.
Here's the complete workflow for creating effective groups: Right-click on your target dimension and select "Create Group." In the dialog box, organize your dimensions into logical groups using drag-and-drop or dropdown assignment. Remember that groups require manual assignment—like managing breakout rooms, you must explicitly place each element where it belongs.
Select contiguous items with Shift+click or non-contiguous items with Ctrl+click for efficient batch processing. Always click "OK" to finalize your groups and make them available as new dimensions.
Understanding when to use groups versus sets is crucial for effective analysis. Groups include all values in organized categories, while sets focus on filtering specific subsets. The visual cues help distinguish them: paperclip icons represent groups, while union symbols indicate sets.
In practice, you'll likely use sets more frequently than groups due to their flexibility and automation capabilities. Sets offer dynamic functionality with top/bottom rules, conditional logic, and automatic updates, while groups always require manual maintenance.
To summarize these organizational tools: Hierarchies streamline your workspace and enable drill-down functionality without constant field manipulation. Groups and sets transform how you can slice and analyze your data—groups for categorical organization, sets for dynamic filtering.
The strategic advantage becomes clear in application: Power Type groupings enable executive-level product portfolio analysis, while subcategory sets (like those exceeding $100,000 across four years) provide operational performance insights. This dual capability supports both strategic planning and tactical execution.
Sets integrate seamlessly with filter functionality, which we'll explore in the advanced modules. The key distinction is scope: filters operate locally within individual sheets, while sets function globally across your entire workbook, making them invaluable for consistent analysis frameworks.
The bottom line: while groups require more manual effort, they provide unmatched flexibility for creating business-relevant categorizations that don't exist in your source data. This capability transforms raw data into strategically meaningful information that drives decision-making across your organization.