Convert PDF product catalogues to Excel

Catalogue design versus a data table

A seasonal catalogue may put four products in a grid, wrap names under photographs, or hide the SKU in a caption. Those pages are hard to treat as rows. Other catalogues include a dense “index” or “price” section that looks like a PDF table: code, name, pack, price. That is the section Linework is for.

If you try to convert the whole booklet, expect empty marketing pages, partial rows, and a lot of cleanup. Exporting only the tabular section is usually faster than forcing every spread through OCR.

What you can extract

• Product codes
• Descriptions
• Pack sizes
• Prices
• Currency

You will not get a full PIM record: no image assets, no category tree, no automatically generated SEO titles. After review you have a working product table you can take into ecommerce preparation or an internal list. Linework does not import into a store.

When extraction is difficult

Multi-column magazine layouts, text on images, rotated pages and stylised fonts all reduce quality. Selectable-text catalogues still need a correct header: a page title is not a column header. If columns wander between sections, re-map after you pick the right header. Repeated headers on later pages can be recognised even when columns shift, but you should verify the first rows of each new section.

Reviewing OCR in catalogues

Image-heavy catalogues often have little text. Automatic mode will scan sparse pages; “OCR on every page” is the fallback when products are missing. Use clear, upright English scans. Watch pack sizes and prices—OCR commonly swaps digits. Low-confidence OCR rows are flagged, but every row still needs a human check. Limits: 15 MB, 50 pages, 20 OCR pages. Split a 200-page book.

For scan-specific advice see scanned PDF to Excel. For a price-list-shaped catalogue appendix, supplier price list to Excel may match the job more closely.

Choosing which pages to convert

A 80-page catalogue often has a story at the front, photography in the middle, and a tabular index at the back. Convert the index. If you cannot split the PDF, you can still read a shorter file more reliably: extract the price section in a PDF tool you already trust, then open that extract in Linework. The 50-page and 20-OCR-page caps exist because long image books are slow and error-prone in the browser.

Seasonal books change descriptions more than codes. If last year’s spreadsheet still has SKUs, use the new extract to refresh prices and packs, then match on code. Do not assume the visual order of products is a stable sort key.

After you have a spreadsheet

Catalogue extracts are starting data. You will still write titles that fit your store, decide categories, and attach images from a separate asset pack. Linework will not generate those. What it can do is stop you from retyping every code and pack from a printed spread. Keep the export notes sheet; it records whether OCR was used and which pages were unreadable.

If two tables appear on one page—new lines on the left, clearance on the right—review is mandatory. Extraction may interleave them. Use the page preview, not the row order, as the authority. When a section header repeats with different columns, pick the header that matches the block you care about and re-extract.

Privacy and next steps

Your PDF is processed locally in your browser and is not uploaded to Linework for document processing. Read how Linework works or how to extract product codes and prices.

Frequently asked questions