Direct comparison
Cost-Per-Use vs. Cost-Per-Unit
Why unit price alone misleads supply budgeting: a worked example of a cheaper item costing more per successful use once failure rate is counted.
Written and maintained by CASRAI Editorial Board
Last updated
Ask CASRAI · free to try
Ask about Cost-Per-Use vs. Cost-Per-Unit
Ask your first 2 questions free below. Subscribers get 150 a day for $29 a month.
An AI assistant specialized in research administration. It cites the sources behind every answer, labels web answers and says when it can't answer.
Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.
Works on this site and inside Claude, Cursor and the AI tools you already use.
Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays free to read, with no account and no card.
How do Cost-Per-Unit, Cost-Per-Use compare side by side?
The table below compares Cost-Per-Unit, Cost-Per-Use across 9 procurement-relevant dimensions, from what it measures through relationship to total cost of ownership (tco).
Side-by-side comparison
| Dimension | Cost-Per-Unit | Cost-Per-Use |
|---|---|---|
| What it measures | The purchase price of a single item, as it appears on an invoice or catalog listing. | The fully-loaded cost of one successful use — item cost divided by the number of times it actually accomplishes its purpose without failure, waste, or rework. |
| Formula | Unit price (or case price ÷ units per case). | Unit price ÷ first-pass success rate (successful uses ÷ units consumed, including discarded or failed units). |
| What it captures | Only the acquisition price of the item. | Acquisition price plus the effect of defect rate, waste rate, and rework tied directly to using the item. |
| What it deliberately excludes | Failure rate, waste, rework, and any use-associated cost beyond the sticker price. | Costs unrelated to the item itself — general facility overhead, staff time that would be spent regardless of which product is used. That's the territory of a full Total Cost of Ownership analysis for capital equipment, not this metric. |
| Best for comparing | Identical or near-identical items from different vendors at the same known failure/waste profile — a genuine apples-to-apples price check. | Consumables where the products being compared have materially different failure, defect, waste, or rework rates — which is most real substitution decisions. |
| Time horizon | None — a single transaction price, true at the moment of purchase. | A representative usage period (a month, a quarter, a defined lot) long enough that the observed failure/waste rate is stable, not a fluke of a small sample. |
| Data required | The invoice or catalog price. Nothing else. | Unit price plus a measured or vendor-reported first-pass success rate for that specific use case — generic manufacturer defect specs aren't a substitute for how the item performs in your facility's actual workflow. |
| Main risk of using it alone | Selecting the item with the lower sticker price that, once failed/wasted units are counted, actually costs more per successful use. | Spending tracking effort on low-value, low-variability items where unit price alone was already a reliable enough proxy. |
| Relationship to Total Cost of Ownership (TCO) | — | The same “don't stop at sticker price” logic TCO applies to capital equipment over a multi-year ownership horizon, applied instead to a single consumable item over its use cycle. See the Total Cost of Ownership (TCO) dictionary entry for the equipment-level version of this idea. |
Common questions
Common questions about Cost-Per-Unit vs Cost-Per-Use
Can you show a worked example of cost-per-unit giving the wrong answer?
+
Illustrative composite, not real vendor pricing — use it as a template, not a sourced quote. A facility is choosing between two single-use consumables for the same formulary slot. Product A is priced at $0.80/unit; on this facility's own usage data, 20% of units are discarded before or during use (sterile-field breach, packaging defect, or in-use failure) and a replacement unit has to be opened to complete the same procedure, so the first-pass success rate is 80%. Product B is priced at $0.95/unit with a 97% first-pass success rate. On unit price alone, A looks $0.15 cheaper. Cost-per-use tells a different story: A is $0.80 ÷ 0.80 = $1.00 per successful use; B is $0.95 ÷ 0.97 ≈ $0.98 per successful use. B is the cheaper option once failure is counted, despite costing more per unit — at 10,000 procedures a year that's roughly $200 in B's favor, before even counting the staff time each failed-and-reopened unit costs with A.
When is cost-per-unit alone good enough?
+
When you have solid reason to believe the failure/waste rate is essentially identical across the options being compared — for example, two lots of the exact same SKU from the same manufacturer, or two items with a long track record at your facility showing no meaningful difference in first-pass success. In that case the lower unit price is genuinely the lower cost, and calculating cost-per-use adds precision without changing the decision.
How do you get a first-pass success rate if you don't already track one?
+
Start with a defined sample: pull a representative lot or a fixed time window (a month is usually enough for a moderate-volume item) and have staff log every unit that's opened but not successfully used — packaging defect on opening, in-use failure, contamination, or any reason a second unit had to be opened to complete the same task. Successful uses ÷ total units opened in that window is your first-pass success rate. It doesn't need to be perfect to be useful; even a rough estimate from a two-week sample is enough to catch a failure-rate gap large enough to reverse a unit-price ranking.
Does cost-per-use include staff labor?
+
Only the labor directly tied to redoing a failed attempt — the extra time to open, prep, and use a replacement unit after a failure. It does not include the baseline staff time the procedure would take regardless of which product is used; that's a real cost, but it's identical across the options being compared and doesn't change the ranking, so most cost-per-use calculations leave it out to keep the comparison focused on what actually differs between the products.
Going deeper
