Builder Lesson 40 of 66 ⏱ 10 min ▶ Video

Aggregate grid and pivot grid

Lesson summary

Charts turn a summarized report into a picture. These two visualizations go the other way: they take your raw detail rows and do the grouping, subtotaling, and cross-tabbing for you — with totals that stay attached to the report and recalculate as users filter. The aggregate grid is a grouped, subtotaled table. The pivot grid is a true cross-tab. One video covers both, because the setup is identical and the payoff is the same: real, dynamic numbers on raw data.

By the end of this lesson

  • Tell an aggregate grid from a pivot grid — and know when the default data grid is already enough
  • Build a grouped, subtotaled aggregate grid whose totals are attached to the report and recalculate when users filter
  • Build a multi-dimension pivot grid (cross-tab), and understand that saving stores its current state, not just its configuration

You'll need

  • A report with raw detail rows and at least one numeric field — the thing you'll group and total. If trimming and tidying a raw dataset is new, Dataset prep and Criteria & prompts cover the prep this lesson assumes
  • Composer role on the app, with a report open in the Composer — see Your first report

Background

Most of Module 8 has been about charts — taking a summarized report and drawing it. The aggregate grid and the pivot grid are different animals. They stay as tables, and they work on your raw detail data — the un-summarized rows straight out of your app. You do the grouping and the totaling inside the visualization, and the numbers it produces are part of the report, not a label painted on top.

That last point is the whole reason these exist, so it's worth being precise about it. Two ways to put a total on a grid:

  • Presentation-layer totals — an extra row drawn under the data. It looks right, but it's just decoration: sort or filter the grid and the total sits there unchanged, no longer matching what's above it. The default data grid's totals work this way (more on that in a moment).
  • Attached totals — totals that are part of the report's calculation. Filter the grid and they recalculate; export it and they follow the filtered rows. The aggregate grid and pivot grid work this way.
Aggregate grid
A grouped, subtotaled table. Group by one or more columns, subtotal and grand-total your numbers, control format and layout per column. Best when you want a clean, foldable detail report — a financial statement, a grouped line-item list.
Pivot grid
A true cross-tab. Drag fields into rows, columns, and data; expand multiple dimensions; show multiple values. Best when you want to read one number across two axes at once — amount by category and period.

Which one — including "none of the above"

There are three rungs here, and you climb only as far as you need to:

Default data grid
Can group and total from Config. Fine when a simple, static grouped list is all you need.
Aggregate grid
When you need real, dynamic subtotals and per-column control.
Pivot grid
When one number needs to read across two axes at once.

One more distinction worth nailing before you start: the Simple Pivot in the Data Prep dropdown is a different thing from the Pivot Grid visualization here. Simple Pivot reshapes the report's own data (a transpose) so it charts cleanly; the Pivot Grid is a rendered cross-tab widget with its own configuration and saved state. Data Prep tools draws that line in full — and notes you pick one path or the other, not both on the same report.

Haven't watched the video yet? It's about ten minutes and it builds both grids on one finance report — a grouped, subtotaled income statement in the aggregate grid, then the same data as a cross-tab in the pivot grid. This page follows the same path and adds the parts the video points to the knowledge base for: where each control lives, and the default-grid totals trap that makes these visualizations necessary.

Not sure which of the two you need for the report in front of you? Email [email protected] with the app and what you're trying to show. Real human, same business day reply.

Do it

  1. Part A — Open the aggregate grid

    With your raw report in the preview, click Visualizations and choose Aggregate Grid. Instead of a chart palette, you get a sidebar where every field is a card. Click a card's action menu and you'll see options the default grid never offered — set Column Width and Column Format right here, per column, no need to go back to the dataset level. (For a click-by-click version of everything below, the help center has How to create a new Aggregate Grid report.)

  2. Group, then subtotal

    On the field you want to group by — category in the video — open the action menu and choose Group By. You can group by more than one column; one is plenty to start. Then open the Settings accordion and turn on Auto Expand Grouped Columns so the groups open by default instead of arriving collapsed.

    Now the subtotal. On your numeric card (amount), the action menu now shows the grouping-aware options — SubTotals and Grand Totals. Choose SubTotals → Sum, and while you're there set Column Format to Currency. Notice what just happened: things you used to set at the dataset level — the aggregate, the data type, the format — you now set inside the visualization.

  3. Place the subtotal where you want it

    By default the sum prints in the group header, in parentheses next to the category name. The aggregate grid lets you move it. Back on the amount card:

    • Align by Column — set to false and the subtotal aligns underneath the actual amount column it's calculating, rather than riding up in the group heading.
    • In Group Footer — set to true and the subtotal drops to the bottom of each group instead of the top.

    These are presentation choices — pick whichever reads best for your report. The Heading Colors and Conditional Highlighting options (shade a row when a value crosses a threshold, or when two columns compare a certain way) live in the same sidebar; the full control list is in the options reference, and the color-rule walkthrough is here. Click Save & Apply.

  4. See that the totals are real

    Open View Full Document and page through. Now do the thing that breaks the default grid: apply a filter to a couple of categories. The subtotals and grand totals recalculate to match what's left — because they're attached to the report, not painted on. Export respects the same filter. This is the difference the whole lesson turns on: your end users can slice the grid and the numbers stay honest.

  5. Part B — From Simple Pivot to the Pivot Grid

    Reset back to the raw grid and let's look at pivoting. There are two layers, and it's easy to conflate them:

    • Data Prep → Pivot (Simple Pivot) is a transpose — a data-prep step that rearranges the report's own rows and columns. Pick one column, one row, one data field, and the data field defaults to a count. If that simple transpose is all you need, you're done; no visualization required.
    • The Pivot Grid visualization is the feature-rich version — and where this part of the lesson goes.

    The line between them matters enough that Data Prep tools spells it out: Simple Pivot reshapes the report data itself; the Pivot Grid is a rendered cross-tab widget with its own configuration, totals, and saved state. You use one or the other, not both on the same report.

  6. Build the cross-tab

    Clear any Simple Pivot, then open Visualizations → Pivot Grid. You get a field chooser with areas for Rows, Columns, Data, and Filter. Drag your fields into place — the video puts category and subcategory into Columns (so you get multiple column expansions), name into Rows, and amount into Data. The preview builds as you drag.

    By default a data field counts. To make it useful, right-click the amount value and set its Summary Type to Sum, and its Format Type to Currency. You can drop multiple fields into Data; right-click any of them to hide it again. There are checkboxes for showing column/row totals and grand totals, for putting data-field headers in rows, and — a feature plenty of organizations need — Show Data As Text, which lets a pivot cell display text instead of a number. Click Save & Apply. (Click-by-click: How to configure a Pivot Grid report.)

  7. Saving a pivot saves its state

    Open View Full Document. Right-click a column to expand it; click a dimension like Category to filter to just the values you want — the totals and grand totals adjust as you go. Here's the part to understand: when you save the pivot grid, you're saving the state it's in at that moment — which dimensions are expanded, which filters are set. Users open it in exactly that starting state, and from there they can keep expanding, filtering, and re-slicing live, with every numeric value recalculating dynamically. You're handing them a configured starting point, not a frozen picture.

Make it real

Two ways the same finance report pays off, plus the three buttons you'll use to move between these views.

A grouped income statement (aggregate grid)

The classic use: an income statement grouped by category, with each line item under it and a subtotal per category — actuals only, last month, currency-formatted, subtotals in the group footer. Because the totals are attached, a viewer can filter to OPEX alone and the subtotal re-sums to just those rows. That's a report a finance team can hand around and trust, not a static export that goes stale the moment someone sorts it.

The same numbers as a cross-tab (pivot grid)

Now read the same amount across two axes at once: category (and subcategory) across the top, name down the side, summed in the cells. One glance answers "how does this line item compare across categories" — a question the grouped list can't answer without scrolling. Save it expanded to the level most viewers want, and let the rest drill from there.

Config, Visualization, and Reset Grid — how to move around

Three controls in the preview toolbar, and mixing them up is the most common confusion:

  • Config reopens the configuration for the visualization you already built — come back here to tweak a subtotal or a format.
  • Visualizations changes the visualization type — clicking it starts you on a different viz, so don't use it for small edits.
  • Reset Grid drops you back to the raw default data grid, clearing the visualization.

If you're stuck

The stumbles, in the order they tend to show up. Most are about which layer you're working in.

My totals look right, but they go wrong when I sort or filter

You added them on the default data grid through Config, where totals and subtotals are presentation-layer — an extra row that doesn't move with the data. That's exactly the limitation the aggregate grid removes. Rebuild the grouping and subtotals in Visualizations → Aggregate Grid and the totals become attached: they recalculate on filter and follow the rows on export.

My pivot value shows counts, not the sum I wanted

A data field in the pivot grid defaults to Count. Right-click the value, set Summary Type → Sum (or Avg, Min, Max), and set the Format Type while you're there. The count default catches everyone the first time.

Hierarchy gives me an error

A hierarchy super-header only spans report fields that are next to each other in the field list. If the fields you picked aren't adjacent, it errors — reorder them so they sit together, then add the hierarchy. Full steps are in the options reference.

Display Full Records made a huge, slow report

Display Full Records removes pagination in the library and in dashboard widgets — it's meant for a small result set, or a widget you've sized to show every row cleanly. On a large dataset it's the wrong tool; leave pagination on, or trim the data with criteria first.

I pivoted in Data Prep but wanted the Pivot Grid visualization (or vice versa)

Different layers. Simple Pivot (Data Prep) transposes the report's own data; the Pivot Grid (visualization) is a rendered cross-tab widget. You can't run both on one report — pick the path that matches the result you want, exactly as Data Prep tools describes.

My saved pivot opens differently than I left it

Saving a pivot grid stores its state — the expansions and filters in place when you saved. If it opens collapsed or filtered unexpectedly, re-open it, set it to the starting view you want users to land on, and save again from there.

I can't drill down from the aggregate grid or pivot grid

That's expected — neither the aggregate grid nor the pivot grid supports drill-down. They're built for grouping, subtotaling, and cross-tab summarizing in place. If a viewer needs to click through to detail, build the drill-down on a standard report or chart instead — see Drill-downs.

None of these match my situation

Email [email protected] with the app, the report, and a screenshot of what you're seeing. Real human, same business day reply.

7-day free trial — no credit card

Built lean. Priced fairly. Supported by humans.

Full access to all features. No credit card required.

Prefer no subscriptions & full control? Self-hosted from $4,995 one-time →

Click once to extend to 14 days — need more time? Just reach out.

25+ years building BI tools Support from the team that builds it Available in US & EU regions
DashboardFox mascot