How TableRune works
TableRune opens your spreadsheet inside your own browser tab and does the work right there. Nothing is uploaded. This page walks through what happens from the moment you pick a file, what Free and Pro each include, and how you can check the privacy claim for yourself in about a minute.
What happens to your file
Most "online" spreadsheet tools upload your file to a server and send back a result. TableRune has no such step, because the whole engine ships to your browser instead of your data shipping to us.
The file is read, not sent
When you pick a file, your browser hands TableRune the bytes and TableRune registers them with an analytics database (DuckDB, compiled to WebAssembly) that runs inside the page. There is no upload, because there is no server-side copy to upload to. Excel, OpenDocument, and Parquet files are decoded the same way, on your machine.
Every action becomes a step
Each cleanup or reshape you pick is recorded as a step in a pipeline. TableRune's engine turns that pipeline into SQL and runs it over your data, in the tab. The Steps list shows the exact query each step ran, so the work is inspectable rather than a black box, and you can undo or redo your way back through it.
The grid is a viewer
Results appear in a virtualized read-only grid that only renders the rows on screen, which is why big files scroll smoothly. You can select cells and ranges, move with familiar spreadsheet keys, and copy a range out with Ctrl/Cmd+C to paste into Excel or Sheets. You do not type into cells: changes are made through the operations, so every change stays a repeatable step.
Step by step
1. Open your file
- Drop in a CSV, an Excel workbook
(
.xlsx,.xlsm,.xls), an OpenDocument sheet (.ods), or a Parquet file. - For a workbook, pick which sheet to use. TableRune guesses which row holds your column names (Excel exports often have a title banner above it) and shows the guess highlighted so you can correct it before importing.
- Column types are detected for you, and the data appears in the grid.
Apple Numbers files (.numbers) cannot be read: export to Excel or CSV from
Numbers first, then open that.
2. Clean it up and reshape it
Operations are grouped into six sets, and you fill in a small form rather than writing code:
- Clean: remove duplicates, trim whitespace, standardize case, fill missing values, tidy dates, emails and phone numbers.
- Transform: rename, drop, split or combine columns, change a column's type, add a calculated column with a formula.
- Filter: keep or remove rows by a condition, keep the top N, sample N rows, drop blank rows or rows missing a value.
- Aggregate: group and summarize, count distinct values, pivot and unpivot.
- Combine: merge with another file on a key, append one file under another, or compare two files to see what is only in each.
- Sort: sort rows, or apply a custom order.
Made a mistake? Ctrl/Cmd+Z undoes a step and Ctrl/Cmd+Shift+Z redoes it.
3. Look at the result
Scroll the grid, or build a chart from your data to see the shape of it. Charts are computed from an aggregated query over all your rows, on your device, and viewing them is free.
4. Take it with you
- Export the cleaned data as CSV or Parquet. The file is written by your browser straight to your computer.
- Or select a range in the grid and copy it into your spreadsheet app.
- Save the pipeline as a named workflow and reload it later onto a new file, so next month's version of the same messy report is one click instead of twenty.
- With Pro, export a chart as a PNG or SVG image to drop into a deck or a document.
There is no Excel (.xlsx) export: data comes out as CSV or Parquet.
What Free includes, and what Pro adds
Free is not a trial. Every cleanup, reshaping and charting feature is in it; Pro raises the size ceiling and unlocks the two things you take away from the app.
Free
- Datasets up to 500,000 rows.
- All six operation groups, with nothing held back.
- Build and view charts.
- Up to 3 saved workflows.
- Export to CSV or Parquet.
- No account, no sign-up, no email address.
Pro
- Datasets up to 10 million rows.
- Unlimited saved workflows.
- Chart export as PNG or SVG.
- Use on up to 3 of your own devices.
- Subscribe. Pick Pro from the pricing section and check out. Payment is handled by Lemon Squeezy, our Merchant of Record.
- Activate. Open your license page, paste the key from your Lemon Squeezy receipt, and click Activate. Pro unlocks on that device right away.
Pro re-verifies with our licensing service every few days and keeps working in between. Moving to a different machine? Sign the old device out from the license page to free its slot. Changed your mind? There is a 14-day refund window (see the refund policy).
Where the row limit actually lives
The row ceiling is not a number in a JavaScript file. It sits inside the compiled WebAssembly engine, which checks your signed license itself before a pipeline runs. We mention it because it also tells you something about the architecture: the thing enforcing your plan is running on your device, next to your data, rather than a server that had to see your file first.
Check it yourself
We say your data never leaves your browser, and you should not have to take our word for it. Because everything runs on your device, you can watch for the upload that is not there.
- Open your browser's developer tools (press F12, or right-click the page and choose Inspect), then click the Network tab.
- With the Network tab recording, open TableRune, load a file, run a few operations, and export the result.
- Read the list of requests. You will see the app's own code and its engine load, and then nothing that carries your file.
Why this shows it: a tool that processed your data on a server would have to send it there, and a 40 MB spreadsheet is not a subtle request. There is no such request, because the query engine came to your data instead.
The things that do cross the network
Being straight with you, three things do, and none of them carries your data. The first is the application itself: the page, its code, and the query engine, some of it
served from our own asset domain (assets.tablerune.com). That traffic goes one
way, to you.
The second is licensing, and only if you buy Pro. Activation and the periodic re-check send your license key, your device's public key or its fingerprint, and a signature. That is the entire payload: no file names, no column names, no rows. On the Free tier TableRune makes no licensing calls at all, because there is no license to check.
The third is feedback you choose to send. If you use the feedback form, we receive what you typed, your email address if you supplied one for a reply, the page you sent it from, and a truncated copy of your browser's user-agent string. Nothing else is attached.
There is no AI feature and no cloud processing option, and that is a permanent decision rather than a postponement: the full reasoning is in the Privacy Policy.
Where your work is kept
All of it on the device you are using, so closing the tab is not the end of your session.
- Your open dataset and its steps are saved into your browser's private file storage, so a reload restores where you were. If that storage is unavailable (a private window, for example), the app still works and simply does not restore.
- Saved workflows are just the list of steps, not your data, and live in this browser's local storage.
- A Pro license token is stored locally too, and is verified on your device by a signature check.
- Clearing TableRune's site data in your browser removes all of the above from this device.
What you need
A desktop browser: Chrome or Edge on Windows, macOS or Linux is what we test against. Working with millions of rows in a tab needs memory and a precise pointer-and-keyboard interface, so phones and tablets are sent to a short explainer page instead of the workspace. There is nothing to install, and no account to create.
Questions, or something did not work the way this page describes?