Insights
The real cost of the cheapest line item
Unit price is the number that appears on the quote, so it is the number that gets compared. It is also the one most likely to be misleading, because it is the only part of the cost that arrives with a decimal point attached.
The rest of the cost is real but arrives later, attached to something that does not look like procurement.
Published
Failure rate is a price
A consumable that fails one time in fifty does not cost its unit price. It costs its unit price plus the labour of the failed attempt, plus whatever the failure interrupted. On a production line, the interruption is usually the largest of the three by an order of magnitude.
This is why a specification exists. It is not procedural overhead — it is the record of what was learned the last time something failed.
Documentation you do not have is a price too
An undocumented material in a regulated environment is either unusable or a liability, and which of the two you discover depends on when someone asks. Reconstructing provenance after the fact costs more than collecting it at the point of purchase, and sometimes it cannot be done at all.
Where the documentation was never part of the quote, it was never part of the price either. It just was not free.
Supplier management is staff time
Every supplier relationship costs a share of somebody's week: chasing, reconciling, re-explaining. Twelve suppliers for twelve product lines is not twelve times the admin of one, but it is not the same as one either, and the difference is paid in the time of people who were hired to do something else.
The unit price on each of those twelve quotes may be the best available. The total is still worse than a consolidated alternative that looked more expensive line by line.
What to compare instead
Compare the delivered cost of the outcome: material, freight, documentation, the expected cost of failure, and the administrative cost of holding the relationship. Most of those are estimates rather than quotes, which is uncomfortable — but an approximate figure for the right quantity beats an exact figure for the wrong one.