The problem

What they were dealing with.

The client managed product data across more than a hundred supplier PDFs with no consistent format, no central list, and no way to quote accurately. A project system update required bulk upload, and the existing data was nowhere near ready.

Outcomes

What changed.

- 8,000+Products structured and validated

- 100+Supplier tables extracted from PDF

- -87%Time spent looking up a product when quoting

- 7 daysFrom raw PDFs to upload-ready dataset

How we worked

Extract, generate, validate, load.

We extracted every product table from the supplier PDFs, normalised field names across formats, resolved duplicates, and validated the output against the import schema of the project system. The final dataset was structured for direct bulk upload, no manual re-entry required.

Where it is now

Where it sits now.

The inventory dataset is live in the project system and maintained by a small offshore team. Current supplier catalogues are processed through the same pipeline as they arrive, keeping the data current without operational overhead.

📌 PDF catalogue to Excel inventory (Auraiya)
🏢 Blyx By Blynd Data
📍 Auraiya