Import pipeline
What happens between your file and a product: where data can come from, how several files are joined, how columns are mapped, and how rows are filtered out.
An import is a configuration, not an upload. It remembers where the data came from, how the files fit together, which column means what and which rows to ignore — so the second run is a re-fetch rather than a repeat of the setup.
Where data comes from
File upload, URL, FTP, JSON endpoint or a connected store — and the difference between a product import and a price and stock sync.
Read pageLinking several files
Nominate a base file and join the others on a shared key, including attribute-per-row exports and columns that produce several values per product.
Read pageColumn mapping
Mapping your columns to system fields, which fields exist, and what happens to the columns you don't map.
Read pageFiltering rows
Skip rules: the operators available, how matching and number parsing behave, how rules combine, and why the row-count estimate is approximate.
Read page