Skip to content
Free · Runs in your browser

HTML to CSV converter

Turn the tables on a web page into CSV files. Paste a table you copied, or drop a saved page, and pick the table you want.

Works with
HTMLMHTMLCSV
Opens inExcelGoogle Sheets

Drop an HTML file here

A saved web page (.html, .htm or .mhtml), up to 100 MB. It stays on your computer.

  • No sign-up
  • No content uploaded
  • Every table on the page
  • Arla Foods logo
  • Bacardi logo
  • British American Tobacco logo
  • Everest Detection logo
  • Golden Acre logo
  • Handlangers logo
  • Inspired Learning Group logo
  • Kantar logo
  • MUFG logo
  • NielsenIQ logo
  • PTI Digital logo
  • RethinkFirst logo
  • Starbucks logo
  • The Specialist Works logo
What you get

Web tables as clean rows.

  • Your file stays on your computer

    The HTML is read and the CSV is written in your browser. Its contents are never sent to Mammoth or anyone else.

  • No footnote marks in the numbers

    Marks such as [1] and hidden text are left out, so 15,292 [14] is written as the number 15292.

  • Merged cells filled in

    A cell that spans two rows fills both, so every row can be sorted and filtered on its own.

How it works

How to convert HTML to CSV

Each table on the page becomes a CSV. Its header becomes the first line, and each row of the table becomes a line.

  1. 1

    Add the table

    Copy a table from a web page and paste it, or drop a page you saved as .html or .mhtml.

  2. 2

    Pick the table

    The largest table opens first. Choose another from the list if the page has more than one.

  3. 3

    Download

    Save the table you picked as a CSV, or every table in one .zip file.

Your options

Other ways to do it

Python's pandas library reads web tables too, with read_html. The difference is what comes out the other end.

  • Python with pandas

    A script you run yourself

    • Footnote marks removedNo
    • One header rowNo
    • Every table at onceYes
    • No code to writeNo
  • This page

    This converter

    A CSV in one step

    • Footnote marks removedYes
    • One header rowYes
    • Every table at onceYes
    • No code to writeYes
Other ways

How to convert HTML to CSV in Python

The pandas library reads every table on a page with read_html, and writes one to a CSV. Here is the whole script.

Want a spreadsheet instead? The HTML to Excel converter puts each table on its own sheet. For a table in a PDF, use the PDF to CSV converter.

Python with pandas
import pandas as pd

tables = pd.read_html("page.html")
print(len(tables), "tables")
tables[1].to_csv("table.csv", index=False)

Pick the table by its number in the list. A table with a two-row header is written with two header rows, and footnote marks such as [12] stay in the cells.

Your file stays private

The converter runs inside this page. Your HTML is read and turned into CSV on your computer, and its contents are never sent to Mammoth. We record only what kind of data it was, such as a price list, and its common column names, never a value.

How we handle your data

Once the page has loaded, it works with your internet switched off.

Every month

Copying the same table every week?

Mammoth can take it from there. Save each week's table as a CSV in a folder and Mammoth adds it to one table, ready to filter, total and download.

Try Mammoth free

Weekly rates

Week 45rates-week-45.csv
Week 46rates-week-46.csv
Week 47rates-week-47.csv
DriveDropboxOneDriveSFTP
All weeksUp to date
Nov 03Euro0.8612
Nov 03British pound0.7598
Nov 10Euro0.8641
Nov 17Euro0.8597
Nov 17Japanese yen154.21

Illustration. Every new file in the folder is added to the same table.

Questions

HTML to CSV questions

Is my file uploaded anywhere?
No. The page reads the HTML and writes the CSV inside your browser. Its contents are never sent to Mammoth or anyone else, and once the page has loaded it works with your internet switched off.
How do I get a table off a web page?
Select the table on the page, copy it, and paste it into the box. Or save the whole page with Ctrl+S (Cmd+S on a Mac) and drop the file on the box. Both .html and .mhtml files work.
Can I paste a web address instead?
No. The converter works only on what you give it, so it never fetches anything from other sites. Copy the table or save the page first.
What if the page has several tables?
The converter opens on the largest table, and you can pick another from the list. To get them all, download every table as its own CSV in one .zip file, each named after the table's caption when it has one.
What happens to footnotes and merged cells?
Footnote marks such as [1] on Wikipedia are left out, and so is text the page hides. A cell that spans two rows fills both rows, and a two-row header becomes one, such as City proper Population.
How are numbers written?
When every value in a column is a number, the thousands commas are taken out, so 37,194,000 is written 37194000 and reads as a number anywhere. A column with text in it, such as 98,964 (2024), is written as it is.
What about links and pictures?
A link keeps its text, without the address. Pictures are left out.
Is it free?
Yes, with no account and no email. Files can be up to 100 MB.