What is a Should-Cost Analysis?
A should-cost analysis is a process led by the procurement professional that answers a simple question: given what raw materials cost today, what should this part cost? It tracks that figure against what the part cost at the last negotiation, so any gap becomes visible and measurable.
Unlike a zero-based cost model, a should-cost analysis shouldn’t try to model the supplier’s P&L from scratch, which is always contentious. Instead, it uses one or two objective inputs—such as the price of a raw material, energy, and labor—and treats everything else as constant. This keeps the model fair and practical, giving buyers competitive pricing that still covers suppliers’ margin and overhead.
Should-cost analyses are simple by design, as their goal is directional accuracy and buy in from the supplier. You will never know your supplier’s true overheads or margin, and you should not expect to. What you can know is the published price of the raw material and what it has done since the last negotiation. For example: If steel was $1,000 a ton in January 2025 and is $950 today, the finished steel part should logically cost less. That directional signal is enough to prevent margin and price creep and put the procurement professional in the driver’s seat at the negotiating table.
How Do Should-Cost Analyses Change Supplier Conversations?
Without a should-cost model, the negotiation dynamic is one-sided. The phone rings, resin is up 10%, and the supplier wants 5-6% on the finished part. That call almost always comes. The call when that same material drops almost never does—and most buyers rarely notice, because their job is firefighting, expediting, and responding to the line. Negotiating prices is one task among many.
We call these silent slopes: raw material costs fall, but the price on your invoice does not. A should-cost model fixes this by making the movement visible in both directions, catching price creep on the way up and margin creep on the way down.
Should-cost analyses also change the negotiation itself, through what you might call a “jujitsu” approach. Rather than arguing against the supplier’s numbers, you use them. You build a first-pass model—say, 50% material cost, 50% gross margin—and send it to the supplier. They will disagree. They will tell you the steel ratio is 60%, not 50%, and you forgot the labor. Take their corrections as-is. Now it is their model, built with their numbers, and they have no choice but to accept the result the next time the raw material moves. The good news is that, going forward, this revised model will work to your advantage, by continuing to prevent price and margin creep.
