How does a purchase order actually arrive?
Most purchase orders arrive as PDFs by email. Sometimes a supplier sends an Excel file, a Word document, or a printout from their own ERP system. Larger trading partners use EDI or a supplier portal, but that is the exception rather than the rule. The result: your purchasing department receives dozens or hundreds of documents per week in formats that differ by supplier. Item descriptions, units, prices, reference numbers: they are all there, but rarely in the same place and rarely phrased the way your system expects.
Why manual entry is a hard bottleneck here
The problem with purchase orders is more specific than with other document types. A supplier uses their own item codes; you have internal codes in your ERP. A supplier writes '12 units box 250 ml'; your system knows that as '1 carton'. Prices are sometimes exclusive, sometimes inclusive of discounts. The employee entering the order has to make all these translations in their head. That takes focus and time, and every error that slips through leads to a mismatch later between the receipt and the invoice. Those mismatches then have to be resolved manually, if they are caught at all.
What document AI concretely solves when processing purchase orders
Document AI reads the incoming document, extracts the relevant fields (supplier, order number, line items, quantities, prices, delivery date), and translates them into the structure of your ERP or purchasing module. The translation of supplier item numbers to internal codes is set up once as a mapping; dottle applies it consistently. Discrepancies, such as a price that does not match the agreed purchase price, are flagged for human review. The employee approves, corrects where needed, and confirms. The manual entry disappears, but control stays with the buyer.
When does automating purchase orders not pay off?
To be honest about the trade-off: with a small and stable supplier base, say ten suppliers each sending a monthly order in a fixed format, automation is not urgent. The setup time does not outweigh the time saved. The same applies when the source data is structurally poor: incomplete item descriptions, missing references, orders arriving as scans of scans. Automation does not fix a broken purchasing process. It makes a good but labour-intensive purchasing process faster and less error-prone.
When does automating purchase orders pay off?
The strongest business case comes when three things converge: high volume (dozens to hundreds of orders per week), high variation in layouts or suppliers, and a system in which you need to register the orders regardless. Wholesalers, manufacturers, and distributors purchasing from many different suppliers are the typical candidates. Once an employee spends more than an hour a day entering and checking incoming orders, it is worth examining what automation costs and delivers. Two weeks is enough to run a pilot and measure the actual time savings.