Compress PDF on Ingest and Report Size Savings in 2026 (Signed Invoices)

When ingesting an invoice PDF, compress it before applying a digital signature. Record the byte counts of both the original and compressed files, their difference, and content hashes. If a signed PDF arrives, keep the original bytes unless a policy allows modification. Compression should be treated as a state transition with a receipt, showing what entered, left, and the storage bytes avoided.

Do not infer success from response codes or rounded percentages. For invoices generated from order data, the sequence is render, compress, validate, sign, archive. Keep the original order ID and renderer version in protected metadata. For inbound signed invoices, retain the received object as the record. Never substitute a smaller rendition for the original signed file.

Count the physical bytes retained, including both objects. The compression method should be lossless, removing unused objects, consolidating repeated resources, compressing eligible streams, and rewriting file structure. Lossless structural cleanup is the conservative option for digitally generated invoices, with gain depending on the source's production method.

Define acceptance checks: render representative pages, extract expected invoice fields, verify page count, and reject smaller results when checks fail. For Node.js, compress the PDF on ingest using a narrow interface. The TypeScript function accepts a compressor supplied by the application, rejects empty output, keeps the original if compression fails, and returns durable metadata with integer byte counts and SHA-256 hashes.