I've made the case elsewhere that churches should watch their vital signs weekly, not yearly. This guide is the practical companion: if you're actually building a dashboard — in a spreadsheet, a reporting tool, or a custom build — which numbers earn a spot on the screen, how is each one defined, and where does the data come from?

Two rules before the list. First: fewer is better. A dashboard with forty metrics is a report nobody reads with better fonts. Twelve is honestly the ceiling; your first screen should carry five or six. Second: every number needs a trend. A single week's value is trivia. This-week-versus-the-same-week-last-year is information.

Attendance (four numbers, not one)

1. Adult worship attendance. Adults in services, all campuses, all service times. Source: your check-in or head-count process, entered into your church management system. Caveat: count consistently — decide once whether volunteers serving during a service count, and never change silently.

2. Kids & students attendance. Tracked separately, because a flat total can hide a growing kids ministry inside a shrinking adult one — two stories needing opposite responses. Source: kids check-in, which is usually your most reliable attendance data anywhere in the building.

3. Online attendance. Pick one definition (concurrent devices, one-minute views, whatever your platform reports), label it on the dashboard, and stop arguing about it. The trend matters; the theological debate about what "counts" doesn't belong on a screen. Source: your streaming platform's analytics.

4. Second-visit rate. Of first-time guests four to six weeks ago, what share came back? First visits measure reach; second visits measure the church. Source: guest records in your ChMS — this one only works if first-time guests are actually recorded, which is itself a finding.

Giving (three numbers, only one of them dollars)

5. Weekly giving vs. budget. Total gifts against the weekly budget need, with year-to-date alongside so one soft week doesn't cause panic. Source: giving platform or ledger. Caveat: normalize for month length and December before comparing anything.

6. Giving households. The count of distinct households giving in the last 90 days. Dollars can look healthy while households quietly decline — fewer people carrying more weight is fragility dressed as stability, and this is the only number that shows it. Source: giving platform, de-duplicated by household.

7. Recurring-giving share. Percentage of giving that arrives on autopilot. Higher means steadier summers and saner cash-flow planning. Source: giving platform.

People (three numbers)

8. Group participation rate. Adults in a small group ÷ average adult attendance. The best predictor I know of whether this year's guests are still around in two years. Source: group rosters in your ChMS — purge dead groups first or the number flatters you.

9. Distinct volunteers this month. Not slots filled — people who served at least once. Reach of the serving culture. Source: scheduling tool or rosters.

10. Volunteer load. How many volunteers served three-plus times this month. This is your burnout early-warning system, and almost nobody watches it. Rising load with flat distinct-volunteers means the same shoulders are carrying more. Source: same scheduling data, different question.

Money & operations (two numbers)

11. Budget-vs-actual flags. Count of budget lines running more than 10% over year-to-date pace. When I audit churches, over-budget areas are almost never decisions — they're the absence of anyone looking monthly. Source: your accounting ledger's budget report.

12. Days of cash on hand. Liquid reserves ÷ average daily operating expense. The board-level health number that turns "are we okay?" into an actual answer. Source: balance sheet. Caveat: exclude restricted funds, or the number lies comfortingly.

The caveats that keep a dashboard honest

  • Trends over values. Every tile should answer "compared to what?"
  • Label your definitions on the screen. "Online attendance (1-min views)" prevents the meeting where two leaders argue about a number that means different things to each of them.
  • Don't dashboard what you won't discuss. A number nobody will act on is decoration. Start with the five or six your leadership team will actually talk about weekly.
  • Garbage in, garbage faster. A dashboard on top of inconsistent check-ins just delivers wrong numbers sooner. Fixing the capture habit is part of the build, not a prerequisite you can skip.
  • Numbers feed discernment; they don't replace it. The shepherd who counts the sheep is still a shepherd.

Getting the numbers onto an actual screen

Every source above — ChMS, giving platform, ledger, streaming analytics — can produce these numbers today. The problem is that assembling them weekly by hand is a real job, and it's the first job that gets dropped. That's the problem dashboards solve: not new data, but the same data, current, in one place, every week without anyone building the packet.

If your team wants to own that screen — your data path, your definitions, your church's brand, with the code in your hands — that's exactly what a custom church dashboard build is. And if you're still deciding what belongs on it, the wider church operations guide shows where these numbers fit in the whole operating picture.