“Internal controls” sounds like something written for a hospital finance department, not a church with sixty giving units and a treasurer who also teaches third grade Sunday school. The term survives because auditors need one, but the idea underneath it is not complicated: money should never depend on the honesty of exactly one person, because that is unfair to the person and unsafe for the church.
Every congregation, no matter how small, has three places money moves through hands: it comes in, someone records it, and someone spends it. Internal controls are just the habits that put more than one set of eyes on each of those three moments. None of what follows requires software, a finance committee, or a budget line. It requires a short list of small commitments a board of five can make in one meeting.
Start with what the money is actually vulnerable to
Most church money problems are not embezzlement. They are honest mistakes that nobody catches because nobody was positioned to catch them: a check deposited but never recorded, a double payment to the same vendor, a designated gift that quietly ends up in the general fund. Controls exist to catch mistakes as much as to prevent theft, and framing it that way makes the whole conversation less awkward with the people who will actually carry it out.
The classic embezzlement risk is real too, and it is worth naming directly rather than dancing around it: it almost always happens to the person everyone trusted the most, precisely because nobody thought to check their work. That is not an argument against trusting your treasurer. It is an argument for never asking trust to do a control's job.
Nobody counts the offering alone
This is the single highest-leverage control in the whole list, and the easiest one to skip when a church is short-handed. Two people who are not related and do not live together count the offering together, every week, and both initial the total before anyone leaves the room. Rotate the pairs so the same two people are not always together. It takes ten extra minutes and it removes the one moment in the entire process where cash exists with no record of it at all.
- Count in a room with the door closed, not spread across a desk in the open.
- Total cash and checks separately, and write both figures down before counting is “done.”
- Deposit the same week — money sitting in a drawer is money nobody can vouch for.
Separate who records money from who reconciles it
The person who enters a gift into your giving records should not be the same person who checks those records against the bank statement each month. If one person does both, an error in either direction has no way to surface. This does not require a second staff member — it can be the same volunteer board member every month, as long as it is never the person who did the recording.
The reconciliation itself does not need to be sophisticated. It needs to happen every month, on a set date, and it needs to actually compare two independent sources: what the church recorded as received, and what the bank shows as deposited. A gap between those two numbers is the first thing any control system is designed to surface.
Spending needs the same two-person habit
The same logic that applies to money coming in applies to money going out. One person should not be able to both approve a payment and cut the check for it. In a small church this often means: any check above a set amount — pick a number the board is comfortable with, even if it is modest — needs two signatures, and every expense over that same threshold gets a receipt on file, not just a memory of what it was for.
Credit cards deserve the same scrutiny as checks, and often get less. If your church has a card, know who holds it, keep the statement in the monthly report the board reviews, and require a receipt for every charge before it is approved — a rule that costs nothing and closes the single most common gap in small-church spending controls.
Give the board something real to look at every month
A board that hears “we're doing fine” once a year at budget time cannot actually oversee anything. Monthly, the board should see: income against budget, expenses against budget, the bank reconciliation, and any check or transfer above the threshold you set. It does not need to be elaborate. It needs to be the same report, on the same schedule, so a gap is visible the month it appears rather than the year it is discovered.
This is the piece a weekly and monthly admin rhythm earns its keep on — controls that depend on someone remembering to run a report eventually get skipped during a busy month, which is usually the month they were needed most.
Write it down, even briefly
A one-page policy that says who counts, who records, who reconciles, who signs checks, and at what dollar amount a second signature kicks in, is worth more than a much longer document nobody reads. The point of writing it down is not legal protection. It is that a written policy survives a treasurer leaving, a pastor changing, or a well- meaning volunteer deciding to “just handle it” because they were in a hurry one Sunday. Review it once a year and update it when a role changes hands.
Reviewing the policy is also the natural moment to look at giving trends and confirm year-end statements matched what donors expected — a mismatch there is often the first sign a control gap exists somewhere upstream.
What this does not require
None of the above requires a finance committee of accountants, an outside bookkeeper on retainer, or expensive software. It requires four or five people willing to follow the same short list of habits every week and every month, and a board willing to actually look at what those habits produce. SundayBridge keeps the giving record and the reconciliation-ready history those habits depend on — recording, editing, and year-end statements — but the counting, the second signature, and the monthly report itself are decisions your board makes, not something any software does for you.
The honest limits of any control system
No list of controls eliminates risk entirely, and it would be dishonest to imply otherwise. Two people can still collude, and a board can still rubber-stamp a report it does not read closely. What controls actually do is raise the number of people who would have to be complicit, or inattentive, before a real problem goes unnoticed — and for almost every small church, that is the entire, realistic goal.