Folding Away the Detail: Grouping Rows and Columns in a Financial Model
How row and column grouping in Piskes Sheets lets you collapse supporting detail behind a plus/minus control, and what actually survives when you export the file.
A financial model usually has two audiences at once: the person who built it, who needs to see every line item, and everyone else, who mostly wants the subtotal. Scrolling past forty rows of assumptions to find the number that matters is friction every reader hits. Row and column grouping — sometimes called outlining — solves this by letting you fold detail rows behind a single collapsible header, with a +/− control in the margin that expands or collapses the group on demand.
Using it
Select the rows or columns you want to collapse together — a block of expense line items feeding a subtotal, or a run of monthly columns feeding a quarterly total — and group them. A small control appears in the row or column margin. Click the minus to fold the group down to just its summary row or column; click the plus to bring the detail back. Groups can nest, so you can collapse quarters inside a year, or line items inside a category, at whatever depth your model needs.
Why this is worth calling out specifically
If you've used this feature in Excel under "Group" in the Data tab, you might reasonably assume any web spreadsheet has some equivalent bolted on. It's worth checking rather than assuming: a search of the underlying @univerjs/sheets package that Piskes Sheets is built on turns up nothing for row or column grouping — it isn't part of the open-source component out of the box. Piskes built it as its own feature, in frontend/apps/drive/src/lib/office/outline.ts, rather than getting it for free from the spreadsheet engine underneath.
What survives an export, and why
Grouping is implemented on top of outlineLevel, a property that the .xlsx format itself understands and that ExcelJS (the library Drive uses for spreadsheet export) knows how to write. That matters because not every Sheets feature makes it out of Drive intact when you download a file: charts and pivot tables are both dropped from .xlsx exports today, with a warning telling you so, because there's no clean equivalent to hand off. Grouping isn't in that category — because it's backed by a real .xlsx concept rather than a Drive-only construct, a grouped model opens in Excel with its groups intact, +/− controls and all.
If you're building a model you expect to eventually hand off as a spreadsheet file rather than keep inside Piskes, that distinction is the practical takeaway: lean on grouping freely, but don't build your presentation layer around charts or pivots if a clean Excel handoff is part of the plan.