Indicator Crosswalk
Illustrative, for a fictional community health nonprofit called Cedar Line Community Health. One program, Q3 FY2026, one participant record, $412,000 of program spend. The definition column is quoted from each agreement rather than paraphrased.
| Report | The funder's own definition | Figure | vs internal | Cost per unit |
|---|---|---|---|---|
| Internal dashboard | Distinct individuals with at least one contact | 1,000 | 100% | $412 |
| A-01 Federal award | Unduplicated individuals with a billable encounter | 812 | 81% | $507 |
| A-02 State contract | Households receiving service | 612 | 61% | $673 |
| A-03 Family foundation | Individuals with three or more service contacts | 447 | 45% | $922 |
| A-04 Corporate funder | New clients whose first contact fell in the period | 289 | 29% | $1,426 |
| A-05 Community foundation | Service encounters delivered | 3,146 | 315% | $131 |
| A-06 Health trust | Participants completing all six sessions | 214 | 21% | $1,925 |
| A-06 Health trust | Maintaining the clinical measure six months after completion | no filter | - | - |
Nothing here is a data quality problem. The six people-based figures span 214 to 1,000, a factor of 4.7, and spend is identical in every row, so cost per participant spans $412 to $1,925 by exactly the same factor. Every definition is reasonable on its own terms: the federal award reports billable encounters upward to its own program office, the state contract's eligibility rules are household-based, the family foundation is funding sustained engagement rather than a single visit, the corporate funder is funding growth, the community foundation funds volume, and the health trust is funding a course. The last row is the one that matters most: the participant record has no six-month follow-up field, so that indicator cannot be produced. It says so, and it does not get approximated.