Run a tender and the responses come back looking like they should be easy to compare: a list of suppliers and a list of prices. They are not easy to compare, because the price is the one number that means the least until everything around it is equal.
What actually differs between quotes
Units and packaging. One supplier quotes per piece, another per box of 50, a third per kilogram. Minimum order quantity. A lower unit price at a 10,000-piece MOQ is not cheaper if you need 2,000. Incoterms. An EXW price and a DDP price are not the same price — one of them still owes you freight, insurance and duties. Payment terms. Net 30 versus net 90 is real money at today’s cost of capital.
Normalization: one axis, then compare
Normalization is the unglamorous work of restating every quote on a single, consistent basis — true landed cost for the quantity you actually need, on comparable terms — before anyone ranks anything. Done by hand, it is a spreadsheet rebuilt for every category and a frequent source of expensive mistakes. Done by software, it is automatic, and it is where a tender stops being a guess.
Once quotes are normalized, “cheapest” finally means cheapest — and the award becomes a decision you can defend.
Why it belongs in the engine, not the inbox
Normalization is exactly the kind of step that should end in a decision, not more data. A decision engine collects quotes through a structured portal, normalizes them across units, MOQs, incoterms and payment terms, and scores every supplier on the same axes — so the buyer opens one view and awards from the winning quote, instead of rebuilding a comparison by hand.