How to Scope a QBO Cleanup Before You Quote It

I quoted a cleanup based on a five-minute glance. Client said it was "just a little behind." Twelve hours later I was drowning in unreconciled transactions, phantom balances, and a COA that looked like someone threw darts at a wall. Underquoted by half.

That project didn't go sideways during the work. It went sideways because the scoping was wrong. The 20-minute red-flag scan below tells you whether you're looking at a cleanup, a reconstruction, or something you should walk away from — before you commit to a price or a timeline.

The Mistake Most Bookkeepers Make at the Start

You get access to a new client's QuickBooks Online file. There are 18 months of uncategorized transactions. Bank feeds are disconnected. The chart of accounts has 200+ line items, half of which are duplicates. You start categorizing.

Three weeks later, you realize the opening balances don't tie to anything. There's no prior tax return to reconcile against. The client "thinks" their accountant filed but isn't sure. You're now doing forensic reconstruction, not cleanup, and you quoted for cleanup.

I had a file that looked like a standard 12-month catch-up. Bank accounts reconciled, transaction volume was moderate, client was responsive. Quoted it at 8 hours. Then I opened the reconciliation history and found forced adjustments in six of the previous nine months. Someone had hammered the beginning balances until they matched, which meant every "reconciled" month was unreliable. Undeposited funds had a $9,400 balance from duplicate payment entries. That 8-hour cleanup turned into 28 hours of unwinding someone else's fixes before I could start mine.

That project went sideways because I treated the first hour as production time. I opened the file and started working. The diagnostic step got skipped entirely.

The Red-Flag Scan

Before you commit to a price or a timeline, spend 15 to 20 minutes running through a red-flag scan. This isn't the full diagnostic and scope build, which takes longer depending on how messy the file is. This is triage. It answers one question: am I looking at a cleanup, a reconstruction, or something I should walk away from?

Here's what I check before I touch a single transaction.

Opening balances: do they tie to anything? Pull up the balance sheet as of the start date for your cleanup period. Does it tie to the prior year tax return? If no tax return exists, does the client have a CPA who can confirm a starting point? If neither exists, you're not doing a cleanup. You're doing a reconstruction, and the price is different.

Chart of accounts: how inflated is it? How many accounts are there? Are there duplicates, inactive accounts with balances, or accounts that clearly don't belong (personal expenses mixed into a business file)? A bloated chart of accounts means every categorization decision during cleanup has a higher chance of landing in the wrong place. I've opened files with 200+ accounts where the owner created a new expense account for every vendor — that's not a quick fix, that's a full COA rebuild before you can categorize anything.

Reconciliation status: clean or forced? Go to the reconciliation screen for every bank and credit card account. Which ones have been reconciled, through what date, and are there discrepancies? Unreconciled accounts are expected in a cleanup. But accounts that were reconciled with forced adjustments are a different problem. Someone "made it work" and you'll have to undo their fixes before you can do yours. I once found a checking account showing negative $43,000 — the previous bookkeeper had manually entered expenses but never connected the bank feed. Six months of income was completely missing from the file. The account looked reconciled at a glance, but the numbers were fiction.

Uncategorized transactions: how deep does it go? Check the balances in Uncategorized Income, Uncategorized Expense, and Ask My Accountant. Heavy balances here are expected in a cleanup file — that's the work. But here's the rule I follow: when I see heavy uncategorized balances, I mentally add 20-30% to my time estimate because there are almost always other issues hiding underneath. Uncategorized transactions mean nobody was watching the books, and when nobody's watching, mistakes compound. Finding and fixing them is usually Step 6 in my process, but the volume you see here directly affects your scope estimate.

Undeposited funds: the single best indicator of file health. Check the balance right now. If it's anything other than zero (or a small amount representing deposits in transit), payments were recorded incorrectly. I've seen $47,000 sitting in Undeposited Funds — weeks of payments that were invoiced and recorded but never deposited correctly, making the bank balance and the books completely disconnected. This one account tells you more about the state of a file than almost anything else.

AR/AP aging: does it match the balance sheet? Run the aging reports for both receivables and payables and compare them to the balance sheet totals. If they don't match, someone posted journal entries directly to those accounts instead of using invoices or bills. That's a scope variable you need to know about before you quote.

Bank feeds and connections: were they ever connected? Are the bank feeds currently connected? Were they ever? If transactions were manually entered instead of pulled from feeds, you need to know that upfront because you'll be working without the matching layer that most cleanup workflows assume is there.

Payroll setup: where do the liabilities live? Is the client running payroll through QBO, through a third party like Gusto or ADP, or not at all? If payroll liabilities are tracked in QBO, are the balances reasonable? Negative payroll liability balances almost always mean payments were double-recorded, once through the payroll system and again from the bank feed. This is one of the most common cleanup issues and one of the most time-consuming to unwind.

Why This Changes Your Pricing

Every item in that scan is a scope variable. A file with clean opening balances, a tight chart of accounts, and connected bank feeds is a fundamentally different project than a file with none of those things, even if the transaction volume is identical.

When you skip the diagnostic, you end up quoting based on the number of months behind or the number of transactions. Those are inputs, not complexity indicators. Two files can both have 12 months of uncategorized transactions. One of them takes 8 hours. The other takes 30. The difference is almost always in the items above. For a category-by-category framework that turns the red-flag scan into a defensible hour number, see how to estimate hours for a QBO cleanup.

Running the red-flag scan first means you can make a go/no-go decision with confidence, and if you go, you have the information you need to scope accurately and price the project without guessing.

The Diagnostic Scorecard

After running the red-flag scan, count how many areas showed problems. The total gives you a quick complexity tier that maps to realistic time estimates:

0-1 red flags: Catch-up job, not a real cleanup. The file is fundamentally healthy — it's just behind. Bank feeds are connected, reconciliations are clean (or close), and the COA makes sense. You're mostly categorizing transactions and reconciling recent months. 3-8 hours. These are great projects: predictable, profitable, and fast to deliver.

2-3 red flags: Moderate cleanup. The fundamentals are there but nobody maintained them. Maybe the COA needs trimming, reconciliations have gaps, and there's a pile of uncategorized transactions. You'll need to fix some structural issues before you can do the routine work, but you're not rebuilding from scratch. 10-25 hours. This is the sweet spot for most cleanup specialists — substantial enough to justify a meaningful fee, contained enough to scope with confidence.

4-5+ red flags: Full reconstruction. Significant structural problems that touch almost everything. Opening balances don't tie, the COA needs a complete overhaul, reconciliations were forced or never done, and there are phantom balances across the balance sheet. You're not cleaning up — you're rebuilding the file's foundation before any transaction-level work makes sense. 25-60+ hours. Don't quote flat-rate without thorough discovery. If you're going fixed-fee, pad your estimate or build in a mid-project reassessment clause. These are the projects that eat bookkeepers alive when scoped casually.

Use these tiers as a starting framework, not gospel. A file with two red flags but 500 transactions per month is harder than a file with three red flags and 40 transactions per month. Context matters. But the scorecard gives you a defensible baseline for your initial conversation with the client.

The Takeaway

Build the red-flag scan into every engagement before you commit to anything. Use it to identify your known-good starting point, flag scope risks, and determine whether you're looking at a cleanup, a reconstruction, or something you should walk away from. And once your scan produces a real deliverable, start charging for it — most bookkeepers give this work away free and pay for it in unbilled hours.

This scan tells you what's broken. But turning that into an accurate scope document, a defensible price, and a client-facing proposal is the harder part. That's what I built LedgerClean to do — upload your client's QBO exports, get a diagnostic scan across all of these areas automatically, and use the results to build a client-facing proposal with time estimates and pricing. Try a free scan and see how it compares to what you'd find manually.