Excel Won't Open My Large CSV (1M+ Rows) — What to Do

Double-click a 2-million-row export and Excel either truncates it at 1,048,576 rows, freezes, or "helpfully" reformats your IDs and dates. The file is fine — the tool is wrong for the job. Google Sheets is not much better with its cell-count caps.

You do not need a database for most oversized CSVs. Preview the shape, split or sample it into openable parts, filter to the rows you need, and convert only the slice Excel can handle — all in your browser with RividTech, so the data never uploads. For format-loss details, see Excel to CSV (and Back) Without Losing Data.

Step 1: diagnose the size and shape

Before splitting anything, measure. Run Data Stats for row/column counts, empty cells, uniques, and type inference. Open a slice in CSV Preview to check headers and whether the first rows are real data or export junk.

• Over 1,048,576 data rows — must split or filter; Excel physically cannot hold it.
• Under the limit but slow — wide files (50+ columns) or long text fields freeze Excel; drop columns first.
• One column of commas — delimiter problem, not a size problem. Fix with Fix Delimiter.

Ragged rows and empty headers also crash imports — confirm structure with Validate CSV.

Step 2: split into openable parts

Split CSV breaks a large file without Excel:

• By row count — e.g. 500k rows per part keeps every chunk safely under the Excel cap with headroom.
• By column value — e.g. one file per region, month, or status, so each part is meaningful on its own.

Keep headers in every part (the tool does this), name parts clearly (orders_2026-01.csv), and never re-save a part in Excel before converting — Excel may silently change dates and leading zeros.

Step 3: sample when you only need a look

Often you do not need all 3 million rows — you need to check a mapping, debug an import, or build a pivot template. Sample Data takes head, tail, or random rows:

• Head (first N) — best for schema checks and building formulas.
• Tail (last N) — best for recent data in append-ordered exports.
• Random — best for representative testing before a full run.

Step 4: shrink columns and filter rows

A 40-column export where you need 6 columns is 85% waste. Drop the rest with Column Tools (rename, drop, reorder), then filter in Data Process — e.g. only status = paid in 2026. A smaller, narrower file opens fast and converts cleanly.

Dedupe before counting: Remove Duplicates on the right key (order ID, email), and normalize labels with Find & Replace.

Step 5: convert only the slice Excel can open

Once a part is under the limit and clean, convert it:

• CSV to Excel — CSV slice to .xlsx for Excel users.
• Excel to CSV — back to CSV after edits, without Excel's save quirks.
• CSV to JSON — when the destination is an app or API, not a sheet.

Converting a file that still has delimiter or encoding issues bakes the damage into the workbook — clean first using How to Clean Messy CSV Files.

What about TSV, JSON, and SQL exports?

The same playbook applies: preview TSV with TSV Preview, flatten API JSON with Flatten JSON, and generate load scripts with CSV to SQL when the data belongs in a database instead of a sheet.

Getting started

Stats, then preview, then split: Data Stats, CSV Preview, Split CSV, Sample Data, and CSV to Excel. Your rows stay on your device the whole time — see our Privacy Policy.