In-House vs Outsourced Bookkeeping: What One Company's Real Numbers Showed

In-house vs outsourced bookkeeping usually gets argued in the abstract. This one got settled by a two-sentence resignation letter, left on a desk in a Round Rock warehouse office on a Tuesday morning in February — nine days before 1099s were due. The company was a commercial cleaning contractor doing about $1.9 million a year, thirty-four W-2 cleaners on night crews, eleven subcontracted floor-care crews, and one bookkeeper who had been there six years and had never once taken a full week off.

What follows is a composite. I've combined details from several engagements so no single client is identifiable, and every number below is either a published figure or ordinary arithmetic you can redo yourself. But the sequence is real, and it's the sequence I see most often. I'm walking through it month by month, with a note after each beat explaining what was actually happening underneath.

The setup: one seat, no backup, and a P&L nobody trusted

Call the owner Dana. Dana had built the company from two vans and a Shop-Vac. The bookkeeper — I'll call her the controller, because that's what the org chart said, even though the actual work was data entry — handled payables, ran payroll through a third-party service, reconciled the operating account, and produced a monthly P&L that Dana glanced at and filed.

Dana's stated reason for keeping the role in-house was the one I hear most: she knows our business. That was true. It was also the entire problem.

What was actually happening: the company had a single point of failure with no documented process, no second set of eyes on cash, and no report that answered the only question that matters in janitorial — which contracts make money. Dana's P&L showed one revenue line and one payroll line. In commercial cleaning, direct labor typically eats half or more of contract revenue, so a company-wide gross margin tells you almost nothing. Two accounts at 38% margin and two at 6% average out to a number that looks survivable right up until the good ones churn. If you've never seen your own statement broken out that way, reading a profit and loss statement line by line is the first place to start.

Week one: the cost of the seat, added up honestly for the first time

Dana's mental math had always been "she costs me fifty-eight thousand." That was the base wage. When we built the actual loaded cost, it looked like this:

• Base wage: $58,000. For reference, the Bureau of Labor Statistics tracks bookkeeping, accounting, and auditing clerks under SOC 43-3031 with a national median near $50,000; Austin runs above that.
• Employer FICA at 7.65%: about $4,437.
• FUTA: 6.0% on the first $7,000 of wages, reduced to an effective 0.6% with the state credit — roughly $42.
• Texas unemployment insurance: the Texas Workforce Commission assesses UI on only the first $9,000 of wages; at a 2.7% rate that's about $243.
• Employer health contribution: about $6,600 across the year.
• Software seat, continuing education, professional dues: about $1,500.
• Recruiting and onboarding, amortized over an expected three-year tenure: about $2,000 a year.

Loaded cost: roughly $72,800, or about 26% above the wage Dana had in her head. That tracks with the Bureau of Labor Statistics' Employer Costs for Employee Compensation series, which has long put benefits near 30% of total private-industry compensation.

What was actually happening: Dana wasn't being careless. She was doing what nearly every owner does, which is compare a wage to a fee instead of comparing a total cost to a total cost. Two things also worked in her favor and are worth naming, because most national articles get them wrong for Texas readers: Texas has no state income tax to administer, and Texas is the only state where workers' compensation coverage is elective for most private employers. A clerical role is a cheap workers' comp classification anyway. The in-house load in Texas is genuinely lighter than it would be in California — which is exactly why this decision shouldn't be made from a generic calculator.

The line that never appeared anywhere: Dana's own time. She spent roughly six hours a week chasing coding questions, approving payables, and re-explaining how job numbers worked. At an owner's real opportunity cost — she was the one closing new contracts — that was the single most expensive item on the list. Six hours a week is roughly 300 hours a year, or about seven and a half full work weeks spent supervising data entry instead of selling.

Week two: what the resignation exposed

With the controller gone, we opened the books. Nothing criminal. What we found was worse in a slower way:

• The undeposited funds account had a balance of $31,000, some of it three years old.
• Eleven floor-care crews had been paid as subcontractors. Four of them worked exclusively for the company, on the company's schedule, with the company's equipment.
• The 1099-NEC file for the prior year had been paper-filed. There were fourteen recipients.
• Nobody but the controller had ever reconciled the operating account. She both entered the bills and had check-signing authority under $2,500.

What was actually happening: three separate exposures, none of them visible on a P&L.

The undeposited funds balance meant the revenue number was fiction and the balance sheet had been quietly accumulating garbage. That's the classic symptom of books that are being maintained rather than closed — and it's the reason a real catch-up and cleanup pass almost always starts on the balance sheet, not the income statement.

The four exclusive floor-care crews were a worker classification problem. Whether someone is a contractor or an employee turns on control and independence, not on what the agreement says, and the exposure compounds across payroll taxes, unemployment, and penalties. If any of that sounds familiar in your own shop, work through how to classify workers correctly before your next quarter closes, and get the specific answer for your facts from your CPA or an employment attorney — that call is not a bookkeeping call.

The paper-filed 1099s were a smaller miss but a telling one: the IRS threshold for mandatory electronic filing of information returns dropped to 10 or more in aggregate, effective with returns filed in 2024. A firm that files those returns every January for dozens of clients knows that in its bones. A solo in-house bookkeeper who files fourteen forms once a year has no reason to.

And the check authority plus reconciliation duty in one person is the textbook control failure. The Association of Certified Fraud Examiners' Report to the Nations has consistently found that organizations with under 100 employees carry a median fraud loss around $150,000, and that the most commonly cited contributing weakness is simply a lack of internal controls. Not malice. Structure.

Is outsourced bookkeeping cheaper than hiring in-house?

Not automatically, and anyone who tells you otherwise is selling. For most businesses under roughly $5 million in revenue, one bookkeeper is more capacity than the transaction volume needs but less expertise than the compliance calendar demands — you end up overpaying for hours and underbuying for judgment. Outsourcing usually wins on total cost below that line because you buy a fraction of several specialists instead of all of one generalist. In-house wins once transaction volume, inventory, or job-costing complexity is heavy enough to keep a full-time person genuinely busy, and once you can afford a second finance person so the duties can be split.

There's a third comparison owners skip entirely, and it belongs on the page: doing it yourself in software with no human attached. That's a real option at low volume and a bad one past it, for reasons worked through in QuickBooks versus hiring a bookkeeper. Dana had passed that threshold four years earlier and never revisited the decision.

Month two: the four questions that actually decided it

Dana didn't decide on price. She decided on these four, and I'd put them to any owner sitting where she was.

1. Can this role be covered when the person is out?

A one-person department has no vacation, no sick week, no notice period that doesn't hurt. Dana's close had already slipped twice in two years for exactly this reason. An outside team has bench depth by construction. If you hire in-house, the honest version of the plan includes who runs payroll the week your bookkeeper has surgery.

2. Can the duties be separated?

Entering bills, cutting checks, and reconciling the account should not live in one pair of hands. In-house at one seat, they always do. Outsourced, they separate naturally — the firm reconciles, you approve payments. That separation is the strongest single argument for outsourcing at this revenue band, and it costs nothing extra.

3. What does the role actually need to know?

Dana needed someone fluent in per-account job costing, night-shift payroll with differentials, subcontractor compliance, and monthly billing in arrears with 30-to-45-day collection cycles. That's a specific skill set. She would have been hiring for it in a candidate pool that mostly comes from retail and professional services. This is the same trap I described in how to choose a bookkeeper: owners screen for software fluency and skip industry fluency, then wonder why the reports don't answer their questions.

4. What is the cost of the reports being late or wrong?

Dana lost a 12-month contract renewal the prior spring because she'd been pricing it off a margin she believed and couldn't prove. That's the cost of inaction — not a fee, a lost account. It dwarfed every line in the loaded-cost table.

Month three: what changed, and what didn't

Dana went outsourced. Six months later, three things were different.

The chart of accounts was rebuilt around service lines and contracts, so she could finally see gross margin per account. Two of her eleven biggest contracts were underwater on labor — one because a building had added 8,000 square feet without a rate change. She renegotiated one and let the other go.

The close landed on the twelfth business day, every month, with the same three reports attached. If you've never had that rhythm, the practical guide to which financial reports to read and when is a good place to see what a real monthly package should contain.

And the four misclassified crews were converted to W-2 with counsel involved. That raised her direct labor cost. It also removed the exposure that would have made the company un-sellable in a diligence review.

What was actually happening: the win wasn't the arithmetic. It was that the bookkeeping function stopped being a person and started being a process. Documented, covered, and separated. That distinction survives whichever side of the in-house vs outsourced bookkeeping question you land on.

Where in-house still wins — and I mean it

I run an outsourced firm, so read this section skeptically and then check it against your own operation. There are real cases where hiring is the better answer:

• High-touch daily volume. Retail or distribution with inventory counts, daily deposits, and constant vendor contact. Someone in the building beats someone on a schedule.
• Deep operational entanglement. If the finance role also handles dispatch, purchasing, or customer service, splitting it out creates more friction than it removes.
• You can staff two. Once the finance function supports a manager plus a clerk, you get separation of duties in-house and the strongest argument for outsourcing goes away.
• You genuinely have the management bandwidth. An unmanaged in-house bookkeeper is the most expensive configuration there is, because nobody catches the drift until it's a cleanup project.

A hybrid is also legitimate and underused: keep an in-house clerk for daily entry and AR follow-up, and outsource the reconciliation, close, and reporting layer. You get responsiveness and separation at the same time. For a longer look at the trade-offs, my earlier piece on outsourced vs. in-house bookkeeping works through seven of them side by side.

The comparison worksheet — run this before you post the job

Compare like for like. Most owners compare a wage to a fee, which is not a comparison at all. On one page, write down for each option:

1. Total annual cost. For in-house, that's wage plus FICA plus FUTA plus state UI plus benefits plus software plus training plus amortized recruiting. For outsourced, it's whatever the engagement actually costs you all-in. Put both numbers in the same units before you look at either.
2. Coverage. How many business days per year is the function unstaffed? In-house at one seat, that's typically fifteen to twenty-five.
3. Separation of duties. Who enters, who pays, who reconciles? If the answer is one name three times, note the exposure explicitly.
4. Cycle time. Which business day do you get closed financials? Write the commitment down and hold both options to it.
5. Scope boundary. Sales tax filings, 1099s, payroll tax notices, year-end package to the CPA — list every deliverable and assign an owner. Anything unassigned lands on you.
6. Escalation. Who tells you when something looks wrong, and how fast? A bookkeeper who only answers questions is not the same as one who raises them.

Whichever way that page comes out, the compliance calendar is unchanged, and it's the part owners underestimate most — payroll deposits, quarterly filings, and worker classification all keep running regardless of who does the entry. The payroll and contractor compliance rules are worth reading before you commit either way.

Dana's decision wasn't really in-house versus outsourced. It was whether the finance function would be a single person's habits or a written process with someone accountable to it. We do this work for trades and service businesses — including bookkeeping built around cleaning companies — and the pattern holds across nearly all of them. Get the process right, then decide who runs it.

Frequently asked questions

Is outsourced bookkeeping cheaper than hiring in-house?

Not automatically. Compare the loaded cost of a W-2 hire — wage, employer FICA, FUTA, state unemployment, benefits, software, training, and amortized recruiting — against the all-in cost of an outside engagement. Below roughly $5 million in revenue, one full-time seat is usually more capacity than you need and less expertise than the compliance calendar demands.

What does an in-house bookkeeper really cost beyond salary?

Employer payroll taxes, unemployment insurance, health contributions, software seats, continuing education, and recruiting typically add 25 to 30 percent on top of base wage. In Texas the load is lighter than in most states because there is no state income tax to administer and workers' comp is elective for most private employers. The uncounted line is owner time spent supervising the role.

What is the biggest risk of a one-person bookkeeping department?

No separation of duties. When the same person enters bills, signs checks, and reconciles the bank account, there is no second set of eyes on cash. Fraud studies consistently identify missing internal controls as the most common contributing weakness at companies under 100 employees, and a single seat also means no coverage during vacation, illness, or a resignation.

Can you combine in-house and outsourced bookkeeping?

Yes, and it is underused. Keep an in-house clerk for daily entry and receivables follow-up, then outsource reconciliation, the monthly close, and reporting. You get same-day responsiveness plus separation of duties, which a single in-house seat cannot provide.

When does hiring in-house become the better decision?

When transaction volume, inventory counts, or job-costing complexity keep a full-time person genuinely busy; when the finance role is entangled with dispatch, purchasing, or customer service; or when you can staff two finance people so duties separate internally. All three require real management bandwidth to supervise the function.