Financial Sustainability
The cost benefit method: five steps that establish what readiness costs, what transport collects, and whether the system a community wants can be produced and sustained.
The decision rubric has set the constraints and the direction and the budget is the mission expressed in numbers. The question is whether the system you want can be produced and sustained.
The method separates two different things, what readiness costs and what transport collects, and forces them to meet in a single, board-defensible decision. It uses ordinary, well-established financial tools: net present value, internal rate of return, and breakeven.
The five steps run as two tracks that converge.
The expense track: what readiness costs
- Production analysisHow many, and when?
- Production expensesWhat does readiness cost?
The revenue track: what transport collects
- Transport analysisHow many transports?
- Transport revenueWhat will you actually collect?
Cost benefitGo or No-go.
The two tracks run in parallel and either may be done first. They meet in the cost benefit analysis. A no-go is accepted or prompts a design change, the system is rescoped, and the analysis is run again.
Before the steps: setting the standard#
The method sizes and prices a system against a standard. In nearly every system that standard is an operational response time (defined in Measurement & the Score), so the standard comes first. Either the system has been given one or it is selecting one.
A response-time standard is a matrix of tiers, not a single number. The tightest allotment (8:59, for example) applies to the highest-priority calls in the densest areas and the allotment lengthens from there.
The response-time goal can be adjusted. The tool for setting it is the marginal utility curve built in step 1: response-time compliance plotted against unit hours, read against each tier's goal. If the curve levels off before the goal is ever reached and no management or personnel failure explains it, the goal is too high and should be moved. If the goal is only reached far out on the curve, the community is spending future-improvement money to buy a number with no demonstrated clinical meaning. The curve does not dictate the standard; it prices each candidate standard so the community can move the dial deliberately. An inherited 8:59 tier may be unreachable (the curve never gets there), reachable but extremely wasteful, or fine by happenstance. The curve is what tells you which.
The standard and the budget are decided together. The five steps set the ceiling: the most the community can sustain over the planning horizon, given what transport collects and what the community is willing to subsidize. Choosing a response-time standard is choosing how much of that ceiling to spend on readiness and how much on everything else.
The conversation usually starts with the standard already fixed (on scene within so many minutes of the request being initialized) and asks only how many ambulances it takes to comply. ODS asks the prior question: is the tier clinical? For most call types, no outcome difference between an 8:59 and a 10:59 tier has been demonstrated (Hansen et al., 2025).
An illustration: at an 8:59 urban tier, compliance takes eight units on the street. At 10:59, it takes four. That exchanges two minutes of operational response time for real funding toward system enhancements.
A standard arrived at this way is changed only explicitly and on the record.
The expense track: what readiness costs#
1 · Production analysis: "How many, and when?"
Production analysis sizes the readiness the system needs: how many unit hours, and when. It starts from demand, measured across the 168 hours of the week rather than assumed, and works out the units required to cover it by time and place. It sizes the capacity to respond; it does not count past responses. Nothing in this step is in dollars, which enter in step 2.
The number of unit hours is set by their marginal utility. Each added unit hour buys a little more compliance with the standard and the gain shrinks as more are added. Plotted, that is the marginal utility curve: compliance on one axis, unit hours on the other. What makes it bend is concurrency (how often calls and commitments overlap), measured from the system's own call and commitment records. At some point the gain stops, because availability is no longer what drives performance. The fewest unit hours that meet the goal are the best economy. Whether that is optimal production takes an analyst's evaluation, because the curve can be erratic and other variables can be involved.
The curve describes expected demand. Capacity for demand above it is sized separately.
2 · Production expenses: "What does readiness cost?"
The full cost of producing the readiness sized in step 1: qualified people, serviceable equipment and supplies, and the overhead to keep them deployable. Costing readiness is how a busy system finds out whether it is losing money.
ODS costs this bottom-up. The number is built from the real components of the system, from an ambulance's acquisition cost down to a single IV catheter. Every item is priced and the total emerges from the parts.
What the cost picture is for and what it is not. The procuring entity knows what each delivery model costs at retail prices, adjusted for scale (reasonable volume discounts) and holds that as its baseline. The baseline guards against failure on both sides:
- Overpaying: a price well above what the model should cost.
- Underpricing to insolvency: a bid so thin the producer risks collapse, leaving the community to absorb the failure. Guarding this floor is an affirmative duty of the procuring entity, because the community, not the bidder, bears the consequences of a failure.
The baseline is not a tool to police producer margins. If a producer builds a genuine competitive advantage, the EMS equivalent of Southwest hedging its fuel, that advantage is theirs to keep. Beating the retail-scale baseline through real efficiency is what a well-designed system should want. The community is entitled to a fair price measured against a known baseline and not to a producer's margin.
The revenue track: what transport collects#
3 · Transport analysis: "How many transports?"
A multi-year forecast of billable work, because under fee-for-service a system is paid only to transport. This is deliberately separate from the demand and readiness analysis in step 1: demand drives cost, transports drive revenue, and the gap between them is where the risk is.
4 · Transport revenue: "What will you actually collect?"
Expected transports run through the fee schedule, the payer mix, and collection estimates. Four factors (transports, fee schedule, payer mix, collections) turn call volume into dollars. The same call volume can be solvent or underwater depending on who is covered. Revenue is expected collections, not billed charges.
The decision: where the tracks converge#
5 · Cost benefit: "Go or No-go."
The expense track and the revenue track meet in one analysis, expressed in standard financial terms: net present value, internal rate of return, and breakeven over the planning horizon. Where the community has committed a subsidy to the system, that subsidy is counted with transport revenue. The output is a decision a governing body can defend and it has two outcomes:
- Go. The system as scoped is sustainable.
- No-go. The system as scoped is not sustainable.
The method is iterative: a no-go prompts a design change, the system is rescoped, and the analysis is run again.