What a Truck Really Costs Per Day, and Other Numbers Your Books Should Tell You: A Q&A With Xero's Bjorn Monsson

Most HVAC and appliance repair owners didn't get into the business to do bookkeeping, and it shows. The books get done at the kitchen table after the last call, a month or two behind, and the first time anyone looks at them closely is when a lender, an insurer or a private equity buyer asks.

Bjorn Monsson has seen that problem from both sides of the desk. He's a CPA who started in assurance, auditing companies across a range of industries, then moved in-house as a general ledger accountant running consolidations and month-end close. Today he's Head of Partner Growth & Solutions at Xero, the cloud accounting platform, where he leads sales strategy across the company's partner consulting and services, partner success and Melio Payments organizations. His job is working with the accountants and bookkeepers who set up and run the books for small businesses, including a lot of shops like yours.

We asked him the questions shop owners actually have: when to stop doing your own books, what a good bookkeeper costs, how to figure out what a truck really costs to put on the road, and what a buyer finds when they open the books of a typical 10-truck shop. His first answer was not what you'd expect from someone who sells accounting software.

---

ServiceMag: You're a CPA who ended up running partner services at a software company instead of practicing. What was the turn, and what do you still carry from the accounting side of the desk?

I started in assurance, auditing companies across a lot of industries, then went in-house as a GL accountant doing consolidations and month-end close. The turn happened in the second job. I was hired to do accounting and spent most of my time fixing the process around the accounting: rebuilding schedules, chasing information that should have been sitting somewhere, shortening a close that took too long. The accounting itself was the easy part. The systems and the process were the problem, and nobody was working on them.

What I carry is knowing what bad books look like from the inside, not in the abstract. I know what it feels like to be three weeks past month-end still trying to figure out what a number is, with someone waiting on it. When I'm talking to an accountant about software now, I'm aware there's always a deadline on the other end and a person who's going to be up late if the software or the process doesn't work.

ServiceMag: You work with accountants and bookkeepers, not directly with the businesses running the software. Explain that to a guy with five trucks. Is he supposed to pick Xero, or pick an accountant and let the software come with?

Pick the accountant. Let the software come with.

I say that as someone who sells the software. The reason is that how your books get set up matters more than what they get set up in. Your chart of accounts, how you treat parts inventory, whether job costs land against jobs, how warranty receivables get tracked: a bookkeeper who knows the trades gets that right in the first month, and you get useful numbers for the next 10 years. A bookkeeper who doesn't gets it wrong, and no software can fix it.

Most good firms already run a defined stack. If you show up and insist on something else, you'll pay more and get worse service, because you've just made yourself their only client who works differently.

Two things to watch for. If your accountant has no opinion at all about software, that tells you something. And whatever they pick, insist on your own login so you can see your own books. Don't accept a black box.

ServiceMag: Many smaller shop owners are doing their own books at night, badly, until something forces the issue. What are the actual signals it's time to stop, what does a competent bookkeeper cost a $1.5 million shop for a year, and what should an owner ask to weed out someone who's only ever done retail and restaurants?

The signals, roughly mildest to most serious:

• It's 9 p.m. and this is the thing you've been dreading all day.
• You're reconciling more than a month behind.
• You can't tell me your gross margin on a replacement versus a maintenance call.
• A lender or insurer asks for a P&L and it takes you a week to produce one.
• You're finding out about cash problems from your bank balance rather than your books.
• Sales tax or payroll tax filings have been late.

The first three mean start looking. The last three mean stop doing this yourself now, and the tax filing one especially, because unpaid payroll tax is the one liability that can follow you personally even through an entity.

On cost, these are market ranges for a shop your size:

Transaction volume and region move that a lot. And the cheap end isn't a bargain if the person doesn't know your industry.

Questions that separate a trades bookkeeper from a retail one:

• How many field service clients do you have right now?
• Which field service software have you worked in: ServiceTitan, Housecall Pro, Jobber?
• How do you handle job costing and work in progress?
• How do you treat inventory on the truck versus in the warehouse?
• How do you handle warranty receivables and progress billing?
• What's your monthly close timeline, and what specifically do I get each month?

Someone who's only done retail and restaurants will answer the first two correctly and go vague on job costing, WIP and warranty. That's the tell. Those are the three things that make trades accounting different.

ServiceMag: Xero turned on credit card bill pay in March. Walk an owner through when paying a distributor by card is worth the processing fee and when he's better off on ACH or 30-day terms.

First, the math nobody does. The card fee is roughly 2.9%. If you're using the card to buy yourself 25 days of float, that fee works out to something north of 40% annualized. As a financing strategy, that's worse than almost any line of credit you'd qualify for. So the card is a cash flow tool for specific situations, not a way to fund the business.

When the card is genuinely worth it: your card earns 2% back, which nets the cost down to under a point, and at that spread the convenience and the float are usually worth it. The distributor won't extend terms, or charges its own finance or late fee that's higher than the card fee. You need the parts today to keep a job, and the alternative is losing the job. Or you're chasing a spend threshold that's worth real money.

When ACH wins: recurring, predictable payments to a distributor who accepts it. It costs you nothing or close to it. This should be your default.

When terms win: always. If a distributor offers net-30 and you can reliably pay on day 30, take it. That's free money, and most shops leave it on the table because nobody asked.

The pattern to avoid is putting the same parts order on the card every single month. That's not managing a timing gap. That's a 35% annual drag on a structural cash problem, and the card is hiding it from you.

ServiceMag: Job costing is where we see shops get hurt. They know revenue and they know parts cost, and they have no idea what a truck really costs to put on the road for a day. What does that look like set up right, and how much of it depends on the field service software talking to the accounting system?

Set up correctly, three things carry to every job: labor at a fully loaded rate, parts at actual cost, and a truck-day charge.

Fully loaded labor means the tech's wage plus payroll taxes, benefits and workers' comp. That's usually 25% to 40% on top of the hourly number in your head. Most owners cost jobs at the wage and wonder where the margin went.

The truck-day is the one nobody builds. Take the payment or depreciation, insurance, fuel, maintenance and tires, and the tooling that lives in that truck. Annualize it, then divide by the days that truck is actually productive. Not 365, and not 250 either. Be honest about it. That gives you a daily number. Charge it to jobs.

Do that and you can see gross margin by job type, by tech and by customer. That's the whole point. Without it, you're looking at a company-level P&L and guessing.

On the systems point, the connection matters more than either tool does on its own. The field service software owns the job. It has the hours, the dispatch, the parts used and the job number. The accounting system owns the money. If your field service software pushes one summary invoice a day into your books with no job detail, you have revenue and no costing, and no amount of accounting software fixes that. What you want is job-level detail flowing across so parts and labor land against the specific job.

One honest limitation, since this comes up. Accounting platforms, including ours, use tracking categories for this, and they're good for a handful of dimensions: division, location, job type. They aren't designed to carry hundreds of individual job numbers, and there's a practical ceiling on them. So for a per-job P&L at 10 trucks, the job lives in the field service software and the accounting system carries the roll-up. Anyone telling you the accounting platform alone will do full job costing at that volume is selling you something.

ServiceMag: In plain terms, what do JAX and Melio Expense Management each actually do for a shop that isn't an accounting firm?

Take a typical situation without expense management. A few weeks ago, your tech pulled up to the supply house and put $340 of parts on the company card. Now somebody collects the receipts, matches each one to its charge and codes it, except about 15% of the receipts are missing.

Expense Management pulls the card spend in as it happens, captures the receipt, categorizes it and pushes it to your books. Two things worth knowing: you keep your existing cards and whatever rewards you're earning on them, so there's no switching, and it works whether or not your books are in Xero. It's built to be accounting-software-agnostic.

JAX is the AI layer inside Xero. Practically, it handles the work that happens before accounting can start: reading bills and documents, matching payments to bills, reconciling bank transactions and flagging what doesn't match. It also connects to Microsoft 365, ChatGPT and Claude, so you can pull live numbers out of your books in the place you're already working rather than logging in to go find them.

Neither of these is exciting. What they are is the difference between a bookkeeper spending 10 hours a month keying things in and spending those 10 hours telling you which job types are losing money.

ServiceMag: The harder version. What in the back office is genuinely automatable today, what's still oversold, and what should a human be touching every single month no matter how good the software gets?

Genuinely automatable now: bank and card reconciliation on recurring transactions. Data entry off bills and receipts. Categorization for vendors you use repeatedly. Payment scheduling and remittance. Sales tax calculation. Chasing people for missing documents.

Still oversold: anything implying your books close without a human. Also "AI advisory." A model can tell you gross margin dropped four points. It cannot tell you whether to hire a fifth tech, because that answer depends on things that were never in your accounting system. And accuracy claims generally. These tools are very good on the thousandth transaction from a vendor you use weekly and noticeably worse on a first-time vendor or an unusual transaction, which is exactly the opposite of how the demos are built.

What a human touches every month, no exceptions:

• The exceptions on your reconciliations. Not the matched items, the handful that didn't match, and why. Sometimes the data coming in is generic and you know best what you spent it on.
• Gross margin by job type against last month. That's the earliest place a problem shows up.
• AR over 30 days, name by name, out loud.
• Payroll. Never stop reviewing payroll.
• A cash forecast for the next 60 days.

I spent my early career in audit, and the errors that mattered were almost never in the transactions the system processed a thousand times. They were in the small number nobody looked at. That hasn't changed, and I don't think it's going to.

ServiceMag: Private equity is buying home services hard right now. What does a buyer find in a typical shop's books that either kills the deal or knocks the multiple down, and how far ahead does an owner need to start cleaning it up?

To be clear on what I'm speaking to: I'm telling you what shows up in the books, not giving you deal advice.

What kills deals is revenue that was never reported. You cannot get a multiple on income you didn't declare, and every owner who's done it thinks they'll be able to explain it in diligence. You won't. Next is no separation between you and the business: personal vehicles, a family member on payroll who doesn't work there, the truck that's actually a boat. Then unfiled or amended returns, payroll tax liabilities, and worker classification problems if you've been running techs as 1099s.

What knocks the multiple rather than killing the deal: cash-basis books with no accrual view, so a buyer can't see what any given month actually earned. No revenue split by service line, so they can't tell recurring maintenance from one-time replacement, and recurring is worth substantially more. No job-level margin, which forces them to assume the worst. Warranty and deferred obligations not accrued. Customer or referral concentration that's real but unquantified, because the data isn't there to quantify it.

On timing, a buyer wants three years of clean statements, so the honest answer is three years out. Two years is workable. If you start 12 months before you go to market, you're accepting a discount, and it will be larger than what the cleanup would have cost you.

If you do one thing this month, stop running personal expenses through the business. That one takes the longest to age out of the numbers.

ServiceMag: Two years out. What does the back office of a well-run 10-truck shop look like that it doesn't look like today?

The close stops being an event. Right now, most shops find out how March went in the third week of April. In two years, for the shops that get this right, the books are close to current continuously, and the owner is looking at margin by job type weekly instead of a P&L quarterly.

Bills arrive and get coded without anyone typing. Counter purchases show up the same day the tech makes them. Payroll, payments and the ledger sit in one place rather than three logins that don't talk. And the bookkeeper's job changes shape: less entering, more reviewing exceptions and answering questions like whether you can afford another truck.

What doesn't change is that someone owns the numbers. A person still looks at them every month.

I'd temper the timeline, too. Two years isn't long enough for this to be normal across the industry. The shops that already have a documented process will get most of the benefit, because the software moves faster over whatever process you hand it. My honest expectation is that the gap between the best-run shops and everyone else gets wider, not narrower.

---

Bjorn Monsson is Head of Partner Growth & Solutions at Xero, where he leads sales strategy and execution across the company's Partner Consultants and Services, Partner Success and Melio Payments organizations. His remit spans enterprise, pre- and post-sales, Gusto, Syft and Xero Practice Manager, along with partner cross-selling, upselling and retention, and full-cycle sales for Melio Payments. He is a Certified Public Accountant and holds a Bachelor of Business Administration in Accounting from San Francisco State University. Connect with him on LinkedIn.

Related reading: What Your Shop Is Worth: How Private Equity Values Home-Service Businesses | Who's Stealing From Your Shop? Fraud Controls for Owners Who Can't Afford a CFO | Chris Buttenham on Phantom Equity and Why Contractors Lose Their Best People