Do Google Sheets charts update in real time when source data changes?

Yes. When a cell within a chart's source range changes - either from a direct edit or from a formula recalculation - the chart re-renders automatically, typically within 1-2 seconds. This works across tab references too: a chart on a Dashboard tab pointing at data on a P&L tab will update when P&L values change. No manual refresh is required.

Why doesn't my Google Sheets chart show new rows or columns I added?

Because chart data ranges are static. When you create a chart over `'P&L'!C3:N15`, Sheets locks that exact range into the chart definition. New rows below row 15 or new columns after N are invisible to the chart, even though they sit directly adjacent. To fix it, either edit the chart's source range manually (Chart editor > Setup > Data range), use an OFFSET-based named range, or build a QUERY formula that outputs to a fixed range and point the chart there.

Can a Google Sheets chart pull data from multiple tabs automatically?

Yes. In the chart editor under Setup > Series, you can add series that reference different tabs using the format `'Sheet Name'!A1:A20`. Each series updates automatically when its source cells change. The limitation is that you're defining each series range manually, and those ranges are static - if the data structure of the source tabs changes (new columns, more rows), you'll need to update the chart series ranges accordingly.

What's the best way to build a chart that always shows the last 12 months of data?

Use a QUERY formula on a dedicated ChartData tab to pull and filter the rolling window, then point the chart at that output range. The QUERY reads a start-date assumption from a cell like `Assumptions!$B$3` and returns the last 12 periods dynamically. As you update the assumption or add new monthly actuals, QUERY recalculates and the chart updates. This is more reliable than OFFSET-based dynamic ranges, which carry a volatility cost that compounds in large multi-tab models.

Does IMPORTRANGE work as a chart data source in Google Sheets?

Yes, with caveats. You can point a chart at a range that contains IMPORTRANGE output. Google's help page says open receiving documents check for updates every hour, but does not guarantee an exact refresh interval. For accuracy-sensitive reporting, expose and verify a source-updated timestamp rather than assuming the chart is current.

Google Sheets Charts Auto-Update: What the Docs Miss

The part that breaks financial models isn't the update mechanism. It's the range.

Google's support documentation states: "When you change data in a spreadsheet, charts that use that data will update automatically." That's accurate. What the docs don't explain is that "that data" means the exact range you defined when you built the chart - nothing more. Add a 13th month column to your rolling revenue chart, and the chart ignores it. Extend your P&L forecast by 2 quarters, and your waterfall sits there, frozen at the original row count.

This distinction matters a lot when you're maintaining a quarterly board pack or refreshing a bank syndicate DCF where the data structure changes month to month.

What "Automatic Update" Actually Covers

When you or a formula changes a value inside a chart's source range, Sheets re-renders the chart within roughly 1-2 seconds. This applies whether the change comes from:

A direct cell edit
A formula recalculation (including SUMIFS, INDEX/MATCH, or QUERY pulling from another tab)
An import via IMPORTRANGE or an add-on writing to the sheet
Another cell that feeds a formula in the chart's range

So if your Revenue tab has a live SUMIFS pulling from a transactions sheet, and your Dashboard chart points at that Revenue output range, any new transaction that shifts the SUMIFS result will flow through to the chart automatically. The update chain works.

What doesn't work: the chart growing its own range. Google Sheets caps spreadsheets at 10 million cells, but that limit isn't the constraint here. The constraint is that chart data ranges are hardcoded at creation time. A chart pointed at 'P&L'!C3:N15 will always and only look at 'P&L'!C3:N15, regardless of what you add in row 16 or column O.

Where Live Financial Models Actually Break

The most common failure pattern is a multi-year model where time runs across columns. You build a 3-statement model with FY2023 through FY2027 in columns C through G. Your chart covers that range. Then FY2028 assumptions get added in column H and the chart doesn't move.

The second failure pattern is row expansion. Your contribution margin by SKU chart covers rows 4 through 22 (18 SKUs). You add 3 SKUs in a new product launch. The chart never sees them.

Both are the same underlying issue: static range, dynamic data.

The third failure is less obvious. Charts built on cross-tab references behave identically to single-tab charts - they update when values change within their defined range - but analysts sometimes expect cross-tab ranges to auto-expand when the source tab structure changes. They don't. A chart reading =SUMIFS('P&L'!C:C,'P&L'!B:B,">="&Assumptions!$B$3) in a helper range and then pointed at that helper range will stay current as long as the helper range itself doesn't need to grow.

Making Chart Ranges Dynamic

There are 3 practical approaches, in order of how well they hold up under pressure.

OFFSET-based named ranges let you define a range that expands as data grows. The setup:

// Name this "RevenueSeries" in Data > Named ranges
=OFFSET('P&L'!C4, 0, 0, 1, COUNTA('P&L'!4:4)-2)

This gives you a row that grows right as you add months. Then build the chart against RevenueSeries instead of a hardcoded range. The chart re-reads the named range definition on each update. One problem: OFFSET is volatile, which means it recalculates constantly and can slow down large models. For a board pack model with 8 tabs and 400+ formulas, that cost adds up.

QUERY as chart source data is the pattern I'd reach for first in a real model. Instead of pointing a chart directly at raw data, create a "Chart Source" section fed by QUERY. The QUERY output lands in a fixed range; the chart points at that fixed range; and QUERY handles the filtering and reshaping dynamically.

// On a ChartData tab, cell A1:
=QUERY(
  {'P&L'!A:N; 'Actuals'!A:N},
  "SELECT Col1, SUM(Col4), SUM(Col7)
   WHERE Col1 >= date '"&TEXT(Assumptions!$B$3,"yyyy-mm-dd")&"'
   GROUP BY Col1
   ORDER BY Col1
   LABEL SUM(Col4) 'Projected', SUM(Col7) 'Actual'",
  1
)

Now your chart covers ChartData!A1:C25 (or whatever max output size you've reserved). As the date range in Assumptions shifts, QUERY re-runs and the chart updates. The range stays fixed; the content changes. This is cleaner than OFFSET and doesn't carry the volatility penalty.

Size the output range conservatively large (more rows than you'll ever need) and fill unused cells with "" or 0 rather than leaving them blank, depending on whether your chart should show gaps.

Entire-column references (A:A) are the quickest fix and the one most likely to create performance problems later. Google's Sheets API documentation notes that unbounded column ranges force the engine to scan all populated cells on recalculation. For a chart, this often means Sheets reads 1,000+ blank cells to find your 18 data rows. Fine for a personal tracking sheet; a problem in a model with 50+ cross-tab formulas recalculating simultaneously.

Cross-Tab Chart Data in Multi-Tab Models

Google Sheets lets you build charts that reference multiple tabs as separate series. The setup is in the chart editor under "Series" - you can manually enter a range like 'Revenue'!C4:N4 for one series and 'COGS'!C4:N4 for another.

This works, but the chart editor's range picker defaults to the active tab, so you end up typing cross-tab references manually. The format Sheets expects:

'Tab Name'!C4:N4

For a gross margin bridge chart pulling from 3 tabs (Revenue, COGS, OpEx), you'd define each series pointing to its respective tab. All 3 series update automatically when underlying cells change. All 3 series have the same static range problem described above.

As of June 2026, there's no native way to define a chart series range as a named range that itself references a QUERY or OFFSET formula. The chart editor accepts named ranges, but only static ones. Dynamic named ranges built with OFFSET or INDIRECT don't work in the chart series range field. This is a documented limitation; the QUERY-as-chart-source pattern above is the practical workaround.

The Pattern That Actually Holds Up

For a live dashboard feeding a quarterly board pack, the cleanest architecture is:

Raw data lives on its source tabs (P&L, Balance Sheet, Cash Flow, Actuals)
A ChartData tab runs QUERY formulas that pull from those source tabs, filter by date assumptions, and output clean series data
Charts live on a Dashboard tab and point only to fixed ranges on ChartData
When Assumptions change (new quarter, revised forecast period), QUERY recalculates, ChartData updates, Dashboard charts update automatically

This gives you the "automatic update" behavior the docs describe, applied to a model where the underlying data actually changes shape over time. The charts never need to be edited. The QUERY formulas absorb all the structural changes.

ModelMonkey can write and maintain these QUERY-based chart source ranges directly from a conversation - useful when the filtering logic changes between board packs or when you're reconciling actual vs. forecast periods that shift each quarter. Try ModelMonkey free for 14 days - it works in both Google Sheets and Excel.