Supplier Price List to CSV — SKU, Price and Source Evidence

Turn a supplier's text-based PDF price table into CSV and JSON product records. Extract SKU, description, unit and decimal price, with the original page and row recorded for each result.

Try the built-in fictional demo by submitting {}. It returns three parsed products and one uncharged review record. No sales call or setup meeting is required.

Pricing

$0.002 per parsed product row delivered to the dataset ($2 per 1,000 rows). Review records, failed documents, duplicate documents within the same batch and the built-in demo have no product-row charge. Platform usage is included under the current listing's event configuration. Your Apify plan and spending limit apply; check the current pricing tab before running.

Examples: 100 delivered rows cost $0.20; 500 cost $1; 1,000 cost $2. These are buyer usage prices, not seller earnings. This tool emphasizes normalized supplier records and evidence, and may cost more than a generic PDF table extractor.

Input

{
  "pdfUrls": ["https://your-supplier.example/catalog.pdf"],
  "currency": "USD",
  "numberFormat": "decimal-dot",
  "maxPages": 10,
  "maxRows": 2000
}

Replace the illustrative URL with a direct public HTTPS PDF you have permission to process. Choose either pdfUrl or pdfUrls. Use up to five URLs; duplicate URLs and identical document bytes within a run are skipped. The row limit is shared across the batch, including review rows. The page limit applies to each document.

For PDFs with both list and net prices, use columnMap to choose the price column explicitly. For example, {"sku":0,"description":1,"price":3} selects the fourth column. A recognizable SKU/price header is still required.

Automated output

• Dataset: successfully delivered product rows, accessible through the Apify API or exports.
• PRODUCTS.csv: the same delivered rows, with spreadsheet formula protection and document identifiers.
• SUMMARY: parsing counts, limits, delivery count and whether the spending limit stopped output.
• DOCUMENTS: per-document outcomes and self-service troubleshooting in batch mode.
• REVIEW: ambiguous, conflicting or incomplete records, without a product-row charge.

The service returns results and diagnostics automatically. It does not include manual conversion, supplier contact, data correction or a compatibility consultation.

Supported scope

Text-based PDFs with visible table borders and recognizable SKU, description and price columns. Supported currency choices: USD, CAD, AUD, EUR and GBP. Decimal dot and decimal comma are explicit options. A dollar sign alone does not establish USD.

Scans, handwriting, arbitrary borderless layouts and password-protected files are unsupported. The extractor does not run OCR or a language model. It does not guess missing prices or repair document contents. Extraction completeness is not independently established; inspect the source before a consequential import.

Limits: 10 MiB per PDF, five URLs, up to 20 pages per PDF, 5,000 total candidate rows, a 16-second per-document processing deadline and a 90-second outer worker deadline. Large or slow documents may fail with a document diagnostic; processing restarts require a new run. Exports persist in the caller's Apify run stores according to their retention settings. Do not submit a sensitive document through a publicly accessible URL.

Common questions

Why are there no rows? Check the document report: scans need a text layer, borderless tables are unsupported, and ambiguous price columns need an explicit map. Invalid or private URLs are rejected.

Why are rows flagged? Missing currency, zero/non-numeric prices, missing SKU/description and conflicting SKU records are examples. Review records are returned separately and are uncharged.

Does the spending limit apply to CSV too? Yes. CSV contains the rows actually delivered; it does not expose rows withheld by the billing limit. The default $0.25 limit may deliver at most 125 charged rows at the current price. Raise the caller's limit deliberately for larger jobs.

Can an AI agent use it? Yes, via the existing Apify API and the documented JSON input. Keep the caller's Apify token server-side and pass it in an Authorization header. Do not put it in a public webpage or example URL.

Free sample and automation kit

Inspect the supplier PDF sample and download the automation kit.

Compare a fictional PDF with its CSV/JSON output before running the Actor. The free kit includes a Python API client and an n8n setup guide for starting a document revision once and collecting the saved run. No email or sales call is required to download it.

The sample demonstrates selectable-text tables with visible borders; scans and borderless PDFs are unsupported. The Python client passed offline tests using simulated responses. The n8n guide is a setup recipe, not an activated or executed workflow.