Sheet column mapper
Moving a sheet is mostly a sorting problem. Some columns carry across untouched, some describe a transaction that the new platform regenerates for you, and a few mean something different from what their name suggests. This tool sorts the header row into those three groups.
Paste the header row, not the data. If your sheet came from a hosted document, selecting the first row and copying it usually keeps the tab separators.
Migration reading
Paste the header row to see which columns still carry useful information, which have to be produced again on the new platform, and which are commonly misread.
Why the header row, not the data
The header row is where the structure lives. Two sheets can hold the same products with completely different column layouts, and the layout is what decides how much work the move takes. Reading the whole file would also mean copying your data into a web page, which this tool deliberately does not ask you to do.
Paste one row. If the tool reports that the line looks like product numbers rather than column names, it is reading a data row and the sorting will be wrong — the message tells you which of the two cases it thinks it is looking at.
What the three groups mean
Carries across
Link, product number, name, brand and category describe the listing rather than the purchase, so they survive a platform change unaltered. The link column is the one that needs work before the move, not during it: rebuild it first, confirm it opens, and the rest of the group follows without further attention.
Has to be produced again
Price, size, colour and photos describe a transaction. A stored price is not a quote, a stored size is a label rather than a measurement, and photos are produced by whichever platform receives the order. Re-entering them is not duplicated effort; it is the part that has to happen on the new platform regardless.
Read differently than you think
Batch codes are the clearest example. They appear in sheets as though they were a grade, and they are not one. This group is flagged separately because getting these wrong changes a decision rather than costing a click.
Where this tool stops
- Abbreviated headers. A column headed only by initials may not match anything. It lands in the unrecognised list, which is a statement about the tool rather than about your column.
- Non-Latin headers. A small set of common Chinese column names is included, but the dictionary is short and will miss regional wording.
- Multi-row headers. Only the first non-empty line is read. Sheets with a merged title row above the real header need that row removed before pasting.
- Nothing is uploaded. The reading happens in the page, and the data row is never requested.
The order that works
Sort the columns first, then rebuild the links, then decide where to send them. Doing it in that order means the platform comparison happens once, against a confirmed list, instead of repeatedly against a list that keeps shrinking.
- The legacy link resolver — rebuild and read the addresses.
- The platform ledger — where the confirmed addresses can go.
- Moving a whole sheet — the longer walkthrough of the same job.