A Checklist for Reviewing a Contractor's Books

You can't reperform everything, so what do you actually check?

Sample the unusual, verify the balance sheet ties to real statements, and write down every error as a rule you can reuse. Skip redoing the whole job.

Redoing every category the contractor already touched defeats the point of hiring them in the first place. You'd be back to doing your own books, just slower, with an extra invoice attached.

The fix is a review by exception: skip the transactions that look exactly like last month's, and spend your attention on what's new or what's been recategorized since you last saw it. Tie every reconciliation back to the actual bank and card statements instead of trusting a spreadsheet that claims it balances. Confirm anything still outstanding, an uncashed check or a deposit that hasn't landed, actually gets tracked instead of quietly disappearing. Then hold the whole set of financials up against last quarter and against what you actually know happened in the business that month. If the numbers don't match your gut, that's the thread to pull.

Knowing which lines are worth a second look matters more than touching every single one of them.

Start with the balance sheet

Start with the balance sheet, not the income statement. It's the part most likely to hide a real problem, because errors there compound quietly instead of showing up as an obviously wrong number.

Ask whether the reconciliation actually happened, not whether someone told you it did. A contractor handling day-to-day categorization work will often keep a separate reconciliation workbook, something that maps every bank and card balance back to the books line by line. Good. Open it. Don't just take its word for it, though. Pull the actual bank statement and check that the ending balance in the workbook matches the ending balance on the statement, for every account, for every month you're reviewing.

This catches more than fraud. It catches the mundane stuff: a duplicated deposit, or an account that quietly stopped being reconciled three months ago because nobody flagged it.

Build the checklist from your own mistakes, not a template

Templates you find online are a decent starting skeleton. They won't catch what actually goes wrong in your business, because they weren't built from your business. Some reviewers start from a SOP cleanup template instead of a blank page, then edit it hard based on what they actually see go sideways in real accounts.

Here's the better habit: every time you find a real error, write it down as a specific rule, not a vague reminder. Skip vague notes like “check categorization.” Write something concrete instead, like “confirm every transaction over $500 tagged as Office Supplies has a matching receipt attached.” Specific enough that two different reviewers would catch the same thing.

The check	Reperforming the work	Review by exception
Transactions	Redo every category the contractor already touched	Sample what's new or what's been recategorized since you last saw it
Reconciliation	Trust the workbook that says it balances	Pull the statement and match the ending balance yourself
Outstanding items	Assume an uncashed check sorts itself out	Confirm it gets tracked instead of quietly disappearing
The checklist itself	A template downloaded from online	A rule written the day you found the error it describes
Rules that stop firing	Stay on the list forever	Get pruned before the list turns into forty items

Then keep score, the same instinct behind logging things in an error and exception tracker instead of a mental list. Does that rule keep flagging real problems six months later, or did it only ever catch one weird thing that happened once? A rule that never fires again probably wasn't a pattern. Keep the ones that do. A checklist that only grows and never prunes turns into forty items nobody actually reads before signing off.

When is this not worth turning into a formal rule?

Not every mistake earns a line in the checklist. If a contractor mistypes a vendor name once and catches it themselves before you even see it, that's just someone doing their job well. Writing a formal rule for every single hiccup turns review into busywork and buries the checks that actually matter under noise.

Honestly, nobody has a clean formula for this yet, at least not one proven across enough cycles to trust blindly. One rule of thumb: if you'd bet money the same mistake shows up again in a different disguise, write the rule. If it feels like a genuine one-off, note it somewhere and move on. You'll know more in six months than you do today, and that's fine. This is a discipline you're building, not a finished system you're implementing.

What ‘more complex’ actually means as you hand off more accounts

Right now, if the checklist works, it might just be because the client is simple, a single new business with no other assets. That's worth saying plainly, since it's tempting to credit the process for something that's really just the account being easy.

As the account stops being simple

Complexity shows up in a few different ways. More accounts to reconcile, sure, but also assets with their own quirks, inventory or intercompany transfers being the usual culprits. Volume matters too, since more transactions means more chances for something unusual to slip past a reviewer moving fast. A checklist built for a brand-new business with one bank account will miss things on an account with three entities and a line of credit.

So when you hand a reviewer a more complex account, go back through the checklist item by item and ask what breaks. Usually it's the sampling logic. “Review anything unusual” means something different when there are far more transactions to skim for what counts as unusual. Widening the checklist as accounts grow is exactly the kind of gap an operations gap checklist is built to catch before it becomes a real miss.

A short note on tools versus judgment

Software checks help. Something like a built-in reconciliation tool will flag mismatches faster than a human scanning a spreadsheet ever could, and there's no reason to turn that down.

But even teams running dedicated bookkeeping review software describe it as a lot of trial and error to get the checks tuned to a specific business. The tool narrows where you look. It doesn't replace deciding what's worth looking at, and it definitely doesn't replace the judgment call of whether an outstanding item is fine to leave open another month or needs a phone call today.

Buy the tool if it helps. Don't mistake owning it for having solved the problem. If you want to see how we build and maintain systems like this for operators who don't have a spare hour a week, that's worth a look too.

Knowing which lines are worth a second look matters more than touching every single one of them.

Common questions

How do I review a contractor's bookkeeping work without redoing it myself?

Use review by exception instead of reperforming every task. Sample new or unusual transactions rather than every entry, check any changes made to prior categorizations, tie every reconciliation to the actual bank and card statements, confirm outstanding items are tracked, and compare the financials against prior periods and what you actually know happened in the business. This catches real errors without requiring you to redo the whole job.

What should I check first when reviewing a bookkeeper's reconciliation?

Start with the balance sheet. Confirm it's actually reconciled by pulling the real bank and credit card statements and checking that the ending balances match what's in the books, for every account and every month under review. A reconciliation workbook is a good sign, but don't take it on faith. Verify it directly, since quiet reconciliation gaps tend to hide there longer than in the income statement.

How do you build a bookkeeping review checklist instead of using a generic template?

Start from a basic exception-review process, then add a specific rule every time you catch a real error, written narrowly enough that another reviewer would catch the same thing. Track whether each rule keeps catching real problems over time and drop the ones that only ever flagged a one-off. Expand the checklist further as you hand off accounts with more assets or higher transaction volume.