Problem
Excel Not Responding on a Big File — What to Do Instead
The progress bar is at 40%, the window has gone pale, and Windows is offering to close the program for you.
The short answer
Excel loads the entire workbook into memory. Past a certain size that stops working, and no setting fixes it. Open the file in something that reads from disk instead — in DataTray, Data Explorer — take the rows and columns you actually need, and export those to Excel.
Why it hangs rather than refusing
A spreadsheet is not a data file. Excel builds a live object for every cell, plus formatting, formulas, a dependency graph and an undo stack. A 300 MB CSV can become several gigabytes of working set once it is a workbook.
When that exceeds available memory, Windows starts paging to disk and everything slows by orders of magnitude. The application has not crashed — it is thrashing, which looks identical from the outside and is often slower than a crash would be.
Check this before you buy RAM
32-bit Excel cannot use more than about 2 GB, no matter what the machine has. It is still widely deployed in managed environments, often on hardware with far more memory than it can reach.
File → Account → About Excel tells you which you have. If it says 32-bit, that is your ceiling, and the answer is a different application or a smaller file — not more hardware.
What will not help
- Closing other programs. Frees a few hundred megabytes against a multi-gigabyte problem.
- Disabling automatic calculation. Helps a formula-heavy workbook; does nothing for a large data table, which has no formulas.
- Saving as
.xlsb. Smaller on disk, identical in memory once open. - Waiting. Once paging starts, the finish line moves away from you.
What does help
Do not put it in a spreadsheet. Open it in a reader that pages against the file — file size stops being the thing that decides whether you can look at your data.
Then take a slice. Filter to the rows that matter, drop the columns that do not, and export that. The result opens in Excel instantly, and it is the file you actually wanted.
Or split it. By row count, by size, or by a column's distinct values — one pass writing a folder of parts with the header repeated in each. Twelve regional files are more usable than one that will not open.
Or convert it. If the file is CSV and stays large, converting to Parquet makes it dramatically smaller and faster to query, and DataTray reads it either way.
The failure that is worse than hanging
If the file has more than 1,048,576 rows, Excel may not hang at all. It opens the first 1,048,576, warns you in a dismissible dialog, and looks completely normal.
Your filters work. Your totals return numbers. The numbers are wrong. A hang is at least honest.
When Excel is still the right tool
For everything after the extraction. Pivot the slice, chart it, model it, format it — that is what a spreadsheet is for, and it is very good at it. The mistake is asking it to be the thing that opens a multi-gigabyte export in the first place.
Every other free way to do this asks you to upload the file first — and a file too big for Excel is usually too big to upload comfortably as well.
Do this in one pass with — Data Explorer
Read-only, always. Open a CSV, Excel workbook, Parquet directory or JSON file of any size and look at the rows — then switch to Columns for a full data profile: what every column actually holds, where the nulls and the mixed types and the outliers are, a data-quality read on each one, and a data dictionary you can hand to someone. A third tab takes SQL if you want it, and never asks you to write any if you don’t. Nothing here modifies the file; the only things written are the exports and reports you ask for. Learn more about the Data Explorer.
Windows 10/11 (x64) · Free · No account