QuickBooks Cleanup Pricing: How to Scope Fixed-Fee Engagements

The cleanup you quoted at ten hours and are still working in month three isn't bad luck. It's a scoping failure — and scoping is fixable.

---

You quoted the cleanup at ten hours. It felt generous at the time: a Shopify client, about eighteen months of history, deposits booked straight off the bank feed, a clearing account that had never once hit zero. Ten hours at your rate. The client said yes on the call.

That was March. It's May, and the file is still open. This is the engagement that gives QuickBooks cleanup pricing its reputation — every fix surfaces two more problems, the second sales channel nobody mentioned showed up in week four, and there's no far edge in sight. You've stopped opening the file on Fridays because opening it costs you money.

Most firms process this exact experience into a rule: cleanups are unquotable — bill hourly and hope.

That's the lie this post exists to take apart. Cleanup risk feels unbounded because you priced it before you measured it. Measure first, and a cleanup becomes one of the most quotable engagements a firm can sell: the variables are countable, the labor is boundable, and the quote survives contact with the file.

Why Hourly Is the Worst Way to Price QuickBooks Cleanup

Be generous about why the lie feels true. If you quote a cleanup blind — off a phone call and a glance at the P&L — the scope genuinely is unknown, and hourly billing is the rational response to an unknown. The mistake isn't hourly. The mistake is quoting blind, and hourly is the symptom.

But look at what hourly does to a cleanup specifically, because the incentive mismatch runs in both directions at once.

On your side: the ten-hour estimate anchored the client's expectations even though it wasn't a quote. At hour thirty, you're choosing between billing three times the anchor and eating the difference — and most firms eat it, which is how e-commerce work quietly destroys firm margins. Worse, hourly punishes competence: your fifth cleanup takes half the hours of your first, and hourly pays you half as much for being better at it.

On the client's side: they're buying an open meter attached to a problem they can't see. Every status update reads as the number going up. They start second-guessing hours, delaying answers to your questions — which adds hours — and the engagement ends with a clean file and a client who resents how they got it. That's a bad setup for the retainer conversation you actually want to have.

Hourly makes a bad month your loss and a good month the client's suspicion. A cleanup needs a model where the number is known before the work starts — which requires knowing the scope before the quote. That's a diagnostic problem, not a pricing problem.

The QuickBooks Cleanup Pricing Formula: Verdict × Months × Defects

Here's the method. Three variables, all measurable before you commit to a number: what's wrong (the diagnostic verdict), how far back it goes (months of history to restate), and how many distinct defects are producing the mess.

Step 1: Diagnose before you quote — always

Never price a cleanup from a phone call. Get read access to the QuickBooks file and the store, and run a structured diagnostic — the Shopify QuickBooks file diagnostic walks through the exact checks, and it ends in a verdict, not a vibe. You're counting two things: which of the recurring e-commerce defects are present (double-counted income, net-deposit revenue, sales tax booked as income, mishandled refunds, a clearing account that never zeroes), and how many months back the damage runs.

Charge a small fixed fee for the diagnostic and credit it against the project if they proceed. It takes about an hour with file access, and it converts the engagement's biggest unknown into a written finding — which doubles as the most convincing sales document the client will ever see about their own books.

Step 2: Band the severity, price per month of history

Restatement effort scales with months of bad history, so the e-commerce bookkeeping cleanup fee should too: a per-month rate, set by how severe the defect list is. Every number below is fictional — round figures to show the mechanics, not a benchmark:

Band	What the diagnostic found	Per month of restated history (illustrative)
Light	1–2 defects, cosmetic — miscategorization, no structural damage	$150
Standard	3–4 defects, structural — fees invisible, tax in income, clearing never zeroed	$250
Heavy	5+ defects, or multi-channel / multi-currency complications	$400

The bands do the honesty work. A three-defect file and a five-defect file are different engagements, and a per-month rate that ignores that difference is just a flat fee wearing a costume.

Step 3: Multiply, fence, quote

The worked example, clearly fictional: the diagnostic on that March client finds three defects — deposits double-counted against synced sales, fees buried in net deposits, sales tax sitting in income — running back nine months. Standard band. Nine months × $250 is $2,250, plus the $350 diagnostic already paid and credited. Fixed quote: $2,600, with the rebuild of the go-forward sync configuration included as the final deliverable.

Then fence it in the engagement letter: the quote covers the defects named in the diagnostic, across the window it measured. A new defect discovered mid-engagement — an undisclosed sales channel, a second payment processor — is a named change order at the same per-month logic, not silent scope creep you absorb. The client can see the unit math ("you're buying nine restated months"), which makes both the quote and the change order explainable in one sentence.

One question decides whether these bands are brave or safe: what's your worst-case labor per month of history? Done manually — rebuilding each month's journals by hand from payout reports — the worst case is ugly, and the bands above would be underpriced. The reason a fixed band can be safe is that restating history doesn't have to be data entry anymore. LedgerPort's Time Machine pushes up to 24 months of historical order data into QuickBooks with your account mapping applied — meaning the corrected history is generated by the sync layer and your billable work is reviewing it, not typing it. Tooling is what caps the per-month labor, and the cap is what makes the per-month price honest.

The Delivery Playbook, Condensed

The full sequence is covered in the CPA's guide to cleaning up a Shopify client's QuickBooks file — what follows is the shape of it, because the delivery order is what protects your fixed fee.

1. Stop the inflow. Pause whatever is still posting defects — the misconfigured sync, the bank-feed rules — before touching anything. A file that's still accumulating errors mid-cleanup is billable hours evaporating.

2. Fix the structure before the history. Stand up the chart of accounts the file should have had: income, refunds as contra-revenue, fees, sales tax liability, one clearing account. Corrections posted into a broken structure are a second mess with your name on it.

3. Delete the bad originals, re-push the correct history. For the restatement window, remove the defective entries and let the sync layer regenerate the months with correct structure — that's the Time Machine pass from the section above. Then verify instead of hoping: filter the audit log to Error status, and an empty list means the restatement actually landed. The clearing account zeroing at each payout date is your proof of completion — show the client.

Notice what this sequence does to the quote: every step is review-and-verify work with a visible endpoint. Nothing in it is open-ended archaeology.

When to Walk Away: The Unquotable Red Flags

The method bounds most cleanups. It does not bound all of them, and quoting an unquotable file is how the March story restarts. Decline — or re-scope — when the diagnostic hits any of these:

• No source of truth. The store is closed, admin access is lost, or the platform history is gone. If correct history can't be re-pulled, restatement is manual reconstruction and the per-month math no longer holds.
• Damage beyond the recoverable window. History older than what the tooling can re-push (24 months, in LedgerPort's case) is manual again. Quote the recoverable window at your band rate, and handle older years as summary adjusting entries coordinated with the tax preparer — or not at all.
• A client who wants the mess fixed but refuses the rebuild. Cleanup without the structural fix means the same defects grow back and you're re-quoting this file next year. That's not an engagement, it's a subscription to someone else's problem.
• Signs of commingling or intentional misstatement. That's forensic work — a different engagement, different letter, different pricing, and possibly not your engagement at all.

Walking away is usually re-scoping: fixed quote on what's boundable, a written pass on what isn't. Clients respect the line more than you'd expect — it's the first evidence you know exactly what you're doing.

The Real Prize: The Cleanup Is the Audition

A cleanup client is the cheapest retainer client you will ever acquire. They've watched your diagnostic name their problems before you touched the file, watched the quote hold, and watched the clearing account hit zero for the first time. No cold prospect trusts you like that.

So never deliver a cleanup without the retainer proposal attached to the final walkthrough — the go-forward sync you configured in the rebuild is the retainer's infrastructure, already running. Price that recurring engagement properly, though: the models and the margin math are their own topic, covered in how to price e-commerce accounting services. That post treats cleanup as a one-line onboarding fee; this one is that line item, expanded. Together they're the whole engagement: fixed-fee cleanup in the front, banded retainer behind it.

Go back to the March file one last time. Under the method, that engagement was a $350 diagnostic, a three-defect verdict, a $2,600 fixed quote the client understood, and a re-push you reviewed instead of typed. Fictional numbers — but the shape is real, and the shape is the point: the cleanup had edges all along. You just have to measure them before you price them.

The fastest way to test the method is on your next inquiry: run the diagnostic, band the verdict, and let the tooling cap the restatement. Start with the CPA onboarding → — we'll walk the first mapping with you, so the rebuild layer is ready before the quote goes out.

---

Related: The CPA's Guide to Cleaning Up a Shopify Client's QuickBooks File · How to Price E-Commerce Accounting Services