Questions ministries actually ask

That is the expected input. The parser is written for real exports: byte-order marks, mixed date formats, embedded newlines, ragged rows, duplicate headers, and columns called things like "Mbr #". Import is preview-first, so you can see exactly what would happen before anything is written.

Matching runs on email, then phone plus surname, then name plus date of birth. Fuzzy name matching is deliberately not used, so two siblings on a shared landline stay two people. A matched member is updated rather than duplicated, and a blank cell means "not provided" rather than "delete what you know".

Scoring runs as soon as members exist, so the triage board is populated immediately after a first import. The integrity report needs contribution and disbursement data to be meaningful, which is usually the second thing a ministry loads.

No. Scoring is a sum of rule weights with no model, no training data, and no learned coefficient anywhere in the calculation. The reasons are shown with their exact weights so you can add the number up by hand and disagree with a specific rule.

The full rule set with weights and rationale is published to administrators as a product feature rather than a debug screen. Changing weights is a deliberate operation, because a score that quietly means something different next month is worse than one you disagree with consistently.

Nothing, and that is correct — it advises rather than commands. What the design tries to earn is the right to be believed: interruptions are capped at one a day, dismissed signals stay dismissed until the facts get materially worse, and every number can be checked.

No, and be wary of software that says it will. Auxilium measures and records. It makes certain patterns visible early — money in with nothing out, denials citing guidelines that postdate a member, a share ratio drifting — but a determined bad actor with administrative access can misuse any system. What changes is how long it takes anyone to notice.

No. Health care sharing ministries are exempt from state insurance regulation, and no software changes that. Measuring against the 80.0% medical-loss floor is meaningful precisely because you are not required to meet it.

Then you will find out privately, before someone else finds out publicly, and the report will show which months and categories moved it. The public page is opt-in; nothing is published because the software felt like it.

If it already tells you which members are quiet, which claims have stalled, and what share of contributions reached medical costs this month, you may not need to. The comparison pages are honest about where the alternatives win.