A stockist table with merged cells
Member lists and stockist pages often merge cells: here Opening hours spans two columns in the header, and Frome spans two rows because two stockists share the town.
By default, a merged value appears once and the cells it covers are empty, which mirrors what you see on the page. Tick Repeat merged cells and Frome is filled into the second row as well, and the header repeats Opening hours over both columns. That second version is the one to use for sorting or filtering, because every row then has its own town.
<table> <tr><th>Town</th><th>Stockist</th><th colspan="2">Opening hours</th></tr> <tr><td rowspan="2">Frome</td><td>Kiln & Co Pottery</td><td>Mon to Fri</td><td>9:00 to 17:00</td></tr> <tr><td>Oak Lane Joinery</td><td>Sat</td><td>10:00 to 14:00</td></tr> <tr><td>Bath</td><td>Marlow Print Works</td><td>Mon to Sat</td><td>8:30 to 18:00</td></tr> </table>
Default: Town,Stockist,Opening hours, Frome,Kiln & Co Pottery,Mon to Fri,9:00 to 17:00 ,Oak Lane Joinery,Sat,10:00 to 14:00 Bath,Marlow Print Works,Mon to Sat,8:30 to 18:00 Repeat merged cells: Town,Stockist,Opening hours,Opening hours Frome,Kiln & Co Pottery,Mon to Fri,9:00 to 17:00 Frome,Oak Lane Joinery,Sat,10:00 to 14:00 Bath,Marlow Print Works,Mon to Sat,8:30 to 18:00
Getting a table out of a page
Copy the HTML
In your browser, open the page source (often right click, View page source), select all and copy. You can also paste just the <table> element, or press Open a file for a saved .html file.
Choose the table
If the page has more than one table, pick from the Table list. Each option shows its number, caption if it has one, and its rows and columns.
Decide on merged cells
Tick Repeat merged cells (colspan and rowspan) to copy merged values into every cell they cover.
Set the separator and save
Choose comma, semicolon or tab, check the preview of the first 50 rows, then copy or download table.csv.
Tables on business websites
Trade association member lists, council registers, show exhibitor lists and wholesale price lists are still often published as HTML tables. Converting the table directly is faster and more accurate than selecting it on screen and pasting into a spreadsheet, which tends to lose merged cells and split long names oddly.
Line breaks inside a cell, written as <br> in the HTML, are kept as line breaks inside one CSV cell, so a two-line address stays in one column. Links, images and formatting are dropped and only the cell text is kept. If the page builds its table with JavaScript after loading, the original source may not contain it; save the page from the browser after it has loaded, or copy the table element from the developer tools.
Questions people ask
How do I extract a table from a website into Excel?
Copy the page's HTML source, paste it here, pick the table and download the CSV. Open the CSV in Excel, or convert it to .xlsx with the CSV to Excel tool if it contains phone numbers or codes with leading zeros. Merged cells can be repeated so each row is complete for sorting.
What do colspan and rowspan do to the CSV?
A cell with colspan covers several columns and one with rowspan covers several rows. The converter places every cell in its correct column. By default the covered cells are left empty; with Repeat merged cells ticked they get the same value, which makes the CSV easier to filter.
Is it safe to paste a whole page's HTML?
Yes. The HTML is read with the browser's own parser into a detached document that is never displayed, and scripts in it do not run. Only the tables are read, so the rest of the page, including any tracking code, has no effect.
Why can't it find the table I see on the page?
Some sites draw tables with JavaScript after the page loads, or lay out data with div elements styled to look like a table. Neither appears as a <table> in the source. Try copying the element from your browser's developer tools after the page has loaded; if it is not a real table there either, this tool cannot read it.
Where your data goes
Your HTML is parsed by the browser on this page without being rendered, and scripts in it never execute. Nothing is fetched from the page's site and nothing is uploaded.
Checked against the tool on