ROI is one of the fastest ways to lose credibility with commissioners if it is presented as a headline number without defensible logic. Most public systems have seen inflated “savings” claims that can’t survive scrutiny, especially when multiple programs influence utilization at once. A commissioner-ready approach treats ROI as an evidence chain: what changed, why it changed, what cost it avoided (or replaced), and what assumptions sit behind the calculation. This article sits within Return on Investment & Value for Money and aligns with the discipline in Cost vs Outcomes so claims are realistic, attributable, and auditable.
Oversight expectations that shape ROI and value-for-money
Expectation 1: Transparency of method and assumptions. State agencies, counties, and Medicaid plans commonly expect ROI narratives to show the method (what was counted), the unit costs used, and the assumptions made (attribution, time horizon, comparison group). “We saved millions” without method typically triggers challenge rather than confidence.
Expectation 2: Evidence that outcomes improved without displacing risk. Funders look for value, not just reduced spend. ROI is weakened if utilization falls because people couldn’t access services, if restrictive practices increased, or if avoidable harms rose. A defensible case shows safety and quality alongside spend impacts.
What commissioners actually mean by ROI in community services
In practice, commissioners look for one of three value-for-money stories: (1) cost avoidance (preventing expensive events like ED visits or inpatient days), (2) cost substitution (shifting care to a lower-cost setting with equal or better outcomes), or (3) productivity (achieving more completed work per dollar without reducing quality). Each requires different evidence. The most common failure is blending them into one number without separating what is truly attributable.
Operational Example 1: Attributable cost-offset modeling that separates “influence” from “control”
What happens in day-to-day delivery
The program builds a simple cost-offset model tied to the pathway it can plausibly influence. Staff define a cohort (e.g., people receiving a specific intervention), track baseline utilization for a defined lookback window, and measure utilization after intervention over a defined follow-up window. The model uses a small set of utilization events (ED visits, inpatient days, crisis placements, jail bookings where relevant) and applies locally agreed unit costs. Crucially, staff record “attribution rules”: which events the program can reasonably claim to influence and which are shared with other system factors. The model is reviewed monthly with commissioners and adjusted when assumptions are challenged.
Why the practice exists (failure mode it addresses)
This exists to prevent overclaiming and to make ROI defensible. Most systems have multiple initiatives running at once; if a program claims credit for all utilization reduction, the claim fails. Separating “influence” (contributing factor) from “control” (directly tied to the intervention) protects credibility and supports honest negotiation about shared impact.
What goes wrong if it is absent
Without attribution rules, ROI becomes a marketing number. Commissioners quickly challenge it: “How do you know it was you?” When the program cannot answer, confidence drops, reporting burdens increase, and renewal conversations become adversarial. Internally, teams may chase “savings” rather than safe outcomes, risking inappropriate diversion or unsafe step-down to make numbers look good.
What observable outcome it produces
An attributable model produces clearer, audit-ready ROI discussions. Evidence includes cohort definitions, baseline vs follow-up utilization tables, documented unit costs, and a written assumptions log. Over time, it supports stable contracting because commissioners trust that the program will not overstate impact.
Operational Example 2: Unit-cost and intensity tracking that links spend to real delivery
What happens in day-to-day delivery
The service tracks the unit cost of delivery in a way that reflects operational reality: staffing hours by role, travel time (if applicable), supervision and QA time, and non-labor costs required to deliver the model safely. Intensity is recorded at the individual level using a simple banding approach (e.g., standard, enhanced, high-intensity), with clear criteria such as frequency of contact, on-call activity, and multi-agency coordination load. Leaders use this to produce a “cost per completed outcome” view—such as cost per stabilized case at 30 days, cost per successful step-down, or cost per avoided ED visit with documented rationale.
Why the practice exists (failure mode it addresses)
This exists to prevent two common distortions: assuming all cases cost the same, and assuming outcomes are achieved without intensity. In reality, complex cases require more resources; if the funding model ignores intensity, either quality drops or staff burn out. Unit-cost visibility allows programs and commissioners to design realistic rate structures and avoid accidental underfunding of high-need cohorts.
What goes wrong if it is absent
Without unit-cost and intensity tracking, programs can’t explain why costs rise or why outcomes vary. Commissioners may assume inefficiency, while the real driver is higher acuity or growing system pressure. Internally, staff may ration time in ways that reduce quality—shorter contacts, less follow-up—leading to higher returns and ultimately higher system cost.
What observable outcome it produces
The outcome is stronger value-for-money defensibility: the service can show what it costs to deliver safely and what results that spend produces. Evidence includes intensity banding logs, staffing-hour summaries, unit-cost calculations, and outcome rates stratified by intensity band.
Operational Example 3: ROI assurance through case-audit sampling and “no perverse incentive” checks
What happens in day-to-day delivery
Each month, the program samples a small set of cases used in ROI reporting (including both “successes” and returns). Auditors verify whether the service actions that underpin the ROI claim actually occurred: timely follow-up, medication access verification, completed referrals, and documented risk decisions. The audit also checks for perverse incentives—such as pushing people away from ED without adequate safety planning, or prematurely closing cases to make outcomes look better. Findings feed into governance, and the ROI model is updated if audit results show weak attribution or incomplete execution.
Why the practice exists (failure mode it addresses)
This exists because ROI reporting can drift into performative compliance. If ROI is tied to contracting, the system is vulnerable to gaming (intentional or accidental). An assurance layer protects both commissioners and providers by proving that claimed outcomes reflect real practice and that utilization reductions are not achieved by shifting risk onto other parts of the system.
What goes wrong if it is absent
Without assurance, a single adverse event can collapse confidence in the entire ROI narrative. Commissioners may conclude the program reduced utilization by reducing access or increasing restriction. The response is often tighter oversight, more reporting burden, and reduced flexibility—making it harder to deliver good outcomes.
What observable outcome it produces
Assurance audits produce an audit trail that strengthens renewal and scaling decisions. Evidence includes sampling logs, verified proof points, documented corrective actions, and stable or improving safety indicators alongside utilization outcomes.
How to present ROI without overpromising
A credible ROI presentation is usually a short pack with: a clear cohort definition, a small set of attributable utilization outcomes, the unit costs used and their source, an assumptions register, and a quality/safety dashboard to demonstrate no harm displacement. The strongest presentations also include a “range” rather than a single number—showing conservative and optimistic scenarios—so commissioners can see the program is not selling certainty where none exists.
ROI that commissioners trust is not about producing the biggest number. It is about making the causal story, the assumptions, and the assurance process clear—so value-for-money is defensible and improvement can continue without argument over credibility.