Supplier catalogues often arrive as PDFs, while price lists arrive as spreadsheets. The tempting workflow is “give the files to an AI and upload the result.” The safer workflow is to treat this as a data-matching and import-preflight project.
A PDF is a presentation format, not a Shopify product data model. Before any upload, create a reviewable intermediate table and keep a clear line between extracted facts, assumptions, and fields that still need a human decision.
1. Choose one stable identity
Pick the supplier SKU or product code as the primary matching key. Keep it as text so spreadsheet software does not turn long codes into scientific notation or remove leading zeroes.
For each row, record:
- supplier SKU or product code
- product title and a proposed Handle
- variant option names and values
- price and currency
- vendor
- image source reference
- a short source-page or source-file note
Do not match a price list to a catalogue by title alone. If a code is missing, duplicated, or spelled differently in the two sources, put that row in an exception list instead of guessing.
2. Build variants deliberately
Shopify product CSVs represent one product across multiple rows. Give all variants of the same product the same Handle, keep the option names consistent, and give each distinct variant its own SKU when one exists.
Check for the common failure modes:
- one variant accidentally gets a new Handle
- a colour or size value changes spelling halfway through the product
- a blank or reused SKU makes two variants look identical
- an update file contains only one variant from a multi-variant product
Keep a last-known-good product export nearby so you can compare identity before changing existing products.
3. Treat images as a separate mapping
An image printed in a PDF is not automatically a usable Shopify image URL. If you have public image URLs, test that each URL opens directly without a login, redirect chain, or expiring session. Then map the image to the correct product and variant before importing.
Do not silently replace a missing image with a guessed supplier URL. Mark it for review.
4. Validate the structured CSV
Once the table has become a Shopify-shaped CSV, run a local preflight for headers, encoding, prices, handles, variant grouping, image fields, and file size. The browser-only Shopify CSV Import Fixer is useful at this stage because it can flag structural problems before you upload a small test batch.
The checker does not parse supplier PDFs, crawl supplier websites, create products, or write to Shopify. It is a safety net for the structured CSV you have already prepared; ambiguous catalogue decisions still need review.
5. Import in a small batch
Start with a few complete products: include every variant, one or two images, and the fields you actually intend to update. Confirm the preview, import result, handles, prices, option values, and image assignments before expanding the batch.
Keep the original source files, the intermediate mapping sheet, the final CSV, and a short exception report. That trail is much easier to debug than trying to reconstruct why a large catalogue import changed an existing product.
The goal is not “one-click PDF to store.” It is a staged conversion where every uncertain match is visible before Shopify receives the file.
Top comments (0)