The questions your bookkeeper keeps asking, and why your job software can't answer them

It’s the 27th, three days before close, and my inbox has five questions from a bookkeeper. Every one of them is the same question wearing different clothes.

“What was the $3,400 deposit on the 14th?” “Which job are these $600 in parts for?” “Why is there a $150 refund with no invoice?” “Is the $2,000 to the owner a draw or an expense?” “This December job shows up in January income, is that right?” Five questions, and under all five is one plain sentence: what job does this money belong to? Your bookkeeper can see every dollar that moved. They can’t see the work behind it. And that gap is where the monthly close goes to die.

The books speak transactions. Your job software speaks jobs. Nobody translates.

Here is the structural problem, and it’s nobody’s fault. QuickBooks records transactions: a deposit, a charge, a transfer, each one a dollar amount with a date and a bank line. Jobber and Housecall Pro record jobs: a customer, an address, a crew, a scope of work. Those are two different languages describing the same day of business, and there is no built-in dictionary between them.

So every month, someone rebuilds the dictionary by hand. Your bookkeeper opens the bank feed, sees a $3,400 deposit, and has to go find which invoices it paid. You get the email, open Jobber, and hunt. Multiply that by a few dozen transactions and you get the thing every trades owner knows in their gut: the close takes a Saturday, and most of the Saturday is detective work. That labor is real, it’s unpaid, and it repeats every single month because the tools forget the answer as soon as you’ve found it.

The five questions, decoded

Once you see the pattern, the recurring questions sort into five buckets. Each one is a specific seam between jobs and books.

  • Deposits that aren’t invoices. A customer pays two jobs with one check, or a card batch settles six tickets as a single $3,400 line. The bank shows one number; the work was several jobs. Someone has to split it back apart.
  • Materials without a job. $600 at the supply house hits the books as an expense. But which job eats that cost? If it lands in a bare “materials” account, your job-level profit is fiction, and the bookkeeper has to ask you, one line at a time.
  • Refunds and adjustments. A $150 goodwill credit to a customer, a partial refund on a callback. These rarely carry the original invoice with them, so they float free in the books until someone reconnects them to the job.
  • Owner draws vs. expenses. The $2,000 transfer to your personal account is a draw, not a cost of doing business, and coding it wrong quietly overstates your expenses and understates your profit. Only you know which it was.
  • Timing across the close. The December job invoiced in January. On a cash basis it’s January income; on accrual it belonged to December. Every job that straddles month-end is a small argument between when the work happened and when the money did.

None of these is an accounting mistake. Every one is a reconciliation question, and reconciliation is exactly the work no single tool in your stack is built to do.

What good hygiene fixes, and the one thing it can’t

Some of this you can take off your bookkeeper’s plate this month, with habits, not software. Put the invoice number or the customer name in the memo field on every deposit. Tag materials to the job in Jobber or Housecall Pro before they ever hit the books. Close out your invoicing before you close the books, so no finished job is sitting unbilled. Those three conventions kill most of the back-and-forth, and I’d adopt them whether or not you ever talk to us.

But hygiene has a ceiling, and it’s worth naming honestly. Memos and naming conventions are you re-typing the connection by hand, every time, hoping you don’t fat-finger it. What they can’t create is a real shared identity for a job across both tools: a fact that says this deposit line is these two Jobber jobs, carried automatically, provable, and the same every time. That isn’t a discipline problem you can drill away. It’s a missing seam between two systems that were never designed to talk. You can paper over it with good habits. You can’t close it with them.

To the bookkeeper reading this

If an owner forwarded you this, you already live everything above. So let me say the part that’s for you directly. The reason serving trades clients is slower than serving, say, a law firm isn’t the volume; it’s that the source of truth is split. Every close, you’re doing forensic work that a retail client’s clean POS-to-books pipeline never demands of you. What changes that math is a feed where each transaction arrives already carrying the job it belongs to: the deposit split into its invoices, the materials tagged to their jobs, provenance on every line so you can click a number and see the two jobs behind it. That doesn’t replace you. It hands you back the Saturday. You go from re-deriving what happened to reviewing it, and you can take on more trades clients without the close eating your month.

One messy deposit, traced all the way back

Let me make it concrete on numbers you can check. Northside Comfort is our labeled sample shop, a fictional HVAC and plumbing business, so pull up the sample digest and follow along. In June, Northside’s operating account shows a single deposit: $3,400 on the 14th. The books record that and nothing more. Here’s the whole thing taken apart.

That $3,400 was one customer paying two jobs with one check. $2,150 was a furnace install closed on the 6th, invoice to Tina Alvarez. $1,250 was a drain-line job at the same property, closed on the 11th, a separate invoice. Two jobs, two crews, one payment. In QuickBooks it’s an undifferentiated lump; in Jobber it’s two closed jobs; and the fact that ties them together, that this deposit is those two invoices, exists in neither place until a human puts it there. Trace it back and the deposit stops being a question. It becomes two jobs you can name, cost, and check. That trace is the entire job of reconciliation, and it’s the seam our connected feed is built to hold.

Bring your bookkeeper to the demo. Seriously.

Most demos are pitched at the owner alone. This one shouldn’t be. If your bookkeeper is the person who feels the jobs-versus-books gap most sharply every month, they’re exactly who should be in the room.

Book a demo and bring whoever closes your books. I run the walkthrough myself, on the sample shop or your own numbers, and we’ll take one real messy deposit apart together, live, back to the jobs behind it. If your close already runs clean and you don’t need us, your bookkeeper will know inside ten minutes, and I’ll say so.

Want the owner’s-eye companion to this first? Read why Jobber and QuickBooks report different revenue, then the overdue-invoice routine and how a weekly business review keeps the gap from growing between closes.

Questions owners ask

Why does my bookkeeper keep asking what a deposit was for?

Because the deposit landed in QuickBooks as a number and a date, and nothing else. Your books record that $3,400 arrived on the 14th. They don't record which jobs it paid, because that fact lives in Jobber or Housecall Pro, not in the bank feed. A lump deposit that covers two invoices, or a customer who pays three jobs with one check, has to be split by hand before the books mean anything. Your bookkeeper isn't confused. The tools genuinely don't share the one fact that would answer the question: which job does this money belong to.

What can I do to make month-end close faster for a trades business?

Three habits fix most of the friction, and none of them need new software. Put the invoice number or customer name in the memo field on every deposit and card batch. Never book materials to a bare expense account; tag them to the job in your field-service tool first. And close the loop on invoicing before you close the books, so no finished job is sitting uninvoiced. These conventions remove the guesswork. They can't remove the one thing that isn't a habit problem: that your job software and your books have no shared ID for a job.

Should trades businesses hire a bookkeeper who knows the trades?

It helps, because a bookkeeper who has closed the books for a plumbing or HVAC shop already expects the deposits-don't-match-invoices problem and the December-job-invoiced-in-January problem. But even a specialist is doing unpaid detective work every month, mapping transactions back to jobs the books never saw. The right answer is a bookkeeper who knows the trades and a feed that carries the job identity into the books, so the mapping is already done when the deposit lands.

Written by Guidepost

Guidepost reads the numbers a home-service shop already has across its tools, then sends the few that need attention, each traced back to its source. The whole job is telling a real signal from noise: the line between a number worth acting on and one that’s only worth a closer look. More about Guidepost →

See it watch your numbers

Guidepost reads your Jobber, Housecall Pro, and QuickBooks numbers and tells you what needs attention, in plain English. Want to see the output first? Look at a sample digest.