Image to Excel Converter

Drop a screenshot or photo of a table — get an editable .xlsx with the rows and columns in place.

Drop a picture of a table here or click to choose a file · or paste a screenshot (Ctrl/⌘ + V)

🔒 Runs entirely in your browser — nothing you type is uploaded or stored on a server.

Drop in a screenshot of a table, a scanned price list or a photo of a printed report, and the page reads it with OCR and rebuilds it as a grid: one spreadsheet row per line of the table, one column per column. Check the result against the picture, fix anything the scan misread, and press Download .xlsx for a real Excel workbook — or Copy for Sheets and paste it straight into Google Sheets.

Also on Txtset: read the text out of any picture · turn CSV into an Excel workbook.

How the columns are found: the OCR engine reports where every word sits on the picture. Words at the same height become a row. A gap between words becomes a column break only when it runs down the whole table — so a two-word name like Ada Lovelace stays in one cell, and a title across the top does not split the columns under it.

It works best on the pictures tables usually come in: a straight-on screenshot or scan with clear text. A photo taken at an angle, handwriting, or a table with merged cells will need more fixing by hand, and the box above the preview is there for exactly that — edit it and the preview and the download follow. Amounts written as 1,250.00 or $12.50 are stored as real numbers so you can add them up straight away.

Your picture never leaves your device. The OCR engine (Tesseract, the same one Image to Text uses) is downloaded once from a CDN, and the reading and the spreadsheet are both done by this page, in your browser.

How to use

  1. Drop, choose or paste (Ctrl/⌘ + V) a picture of a table.
  2. Wait a few seconds while it is read. The first time, the OCR engine downloads.
  3. Compare the preview with your picture and fix any cell in the box above it.
  4. Press Download .xlsx, or Copy for Sheets and paste into a spreadsheet.

Examples

Screenshot of a web page table
Each row and column lands in its own cell, header row bold and frozen.
A price list with $1,250.00
Stored as the number 1250 formatted by your spreadsheet, so it sums.
A ZIP code column holding 02134
Kept as text with its leading zero.
A photo taken at an angle
Rows may run together — straighten it first with Flip Image, or fix the rows in the box.

Frequently asked questions

Is my picture uploaded?
No. The OCR runs in your browser. The only download is the OCR engine itself (about 2–4 MB, from the jsDelivr CDN, the first time), and the page never sends your picture or its text anywhere.
How accurate is it?
On a clean screenshot or scan, printed text is read very reliably and the grid comes out right. Low-resolution images, photos at an angle, unusual fonts and handwriting all cost accuracy. Always compare the preview with the picture before you use the numbers — that is why the result is editable.
Why are two columns merged into one?
If the gap between two columns is narrow, or one row has text that crosses the gap, the page cannot see the break. Put a tab between the values in the box and the preview updates. Cropping the picture to just the table (Crop Image) also helps.
Can it read a PDF?
Not directly — take a screenshot of the page with the table and drop that in.
Does it keep colours, borders and merged cells?
No. It rebuilds the text and the grid; formatting in the picture is not copied. The header row is made bold and frozen with filter buttons.
Which languages does it read?
English, Spanish, French, German, Italian, Portuguese, Dutch and Polish. Pick the language before you drop the picture, or change it afterwards and the picture is read again.