Bookkeeper for IT Support Companies

Where does the money come from?

An IT support company makes money three ways: a flat monthly contract per seat, a one-time project, and hourly work that is often sold in prepaid blocks.

The flat contract is the main line. You charge a set fee per user or per computer every month, no matter if the month was a quiet or frantic. The vendor pricing guides are in the same band: roughly $95 to $295 per seat month, with the common middle range at $125 to $220. A typical small client with ten seats comes in around $1,000 to $2,500 a month. From a bookkeeping point of view, that is the comfortable line, because you know the same invoice comes on the same day.

Projects are the second line and they work differently: one defined job, like a migration, a refresh, or a phone system swap, billed at a fixed fee or an hourly rate. The published hourly rates for this kind of work include about $100 to $350 an hour, with the larger migration projects running from $5,000 to $25,000. We are not going to pretend one of those numbers is the right one. The bookkeeping point is that a project carries its own costs and deserves its own line, not a spot in the monthly service income.

The third line is break-fix hours, usually sold in advance as a block of ten or twenty hours. That line is the one that most often gets booked wrong, and it is the subject of its own section below.

What does running this business really cost?

The dominant cost is labor, and whether that labor is payroll or a subcontractor changes how the margin of every job shows up.

Techs on payroll carry wages, benefits, workers comp, software licenses, and the payroll tax an employer pays. Contractors are different: when you bring in an outside tech for one client job, that invoice is a direct cost of that job. If the migration is quoted at $8,000 and the contractor bills you $2,400, what is left on the job is the margin. That is a number that never appears on a glance at your bank balance, and it is the reason the project costs need the same books as the project income.

Then there is the software stack: the remote monitoring tool, the ticketing and PSA, backups, email security. We looked for a reliable published number for what this stack costs per seat and did not find one, so we are not going to give you a round number. What we do is track your own subscriptions that report the real spend, and put the stack in the books as a direct cost of delivering the service, not as a loose overhead. That way the seat margin shows the true picture.

Hardware is the pass-through piece. If a seat includes a laptop, you buy it from the supplier and the client pays you for the same machine. The books show both sides: it is a purchase and a sale. The margin on that product is a different kind of margin from the margin on the support and the licenses, and it cannot be followed unless the charge is on its own line.

How do prepaid blocks land in the books?

Prepaid money is not earned until the service actually happens, no matter what the bank says the day the payment lands.

A client buys ten hours of support blocks for $1,000 and pays for them in January. The books should show that $1,000 as deferred money in the liability column, not as income. As each support hour is actually used, $100 moves over to income, and the rest still sits as a job you owe. That is not a subtle distinction. It is the difference between knowing how much of your cash is already spent and believing an full prepaid package is profit.

The same mechanic matters even more for the big annual contract accounts. A client that pays the whole year up front has covered twelve months of work, so the revenue belongs across those same twelve months. Thirty dollars or twelve slices depending on the calendar, with the unconsumed part sitting in deferred. The whole check does not belong in the month it was written.

If a shop uses a strict cash accounting method, the month a prepaid check arrives looks like a record. The month the actual work happens looks like nothing. Both are false. A set of books that defer the income until the hours are used will tell a different story, and it is the story that will show the owner what the business actually earns. That is the first fix we bring up with a new client in this trade.

What usually goes wrong in these books?

The three biggest mistakes in this trade are the one big client that floats the whole month, a product margin hidden in the service margin, and the counting of money that has not been earned yet.

The prepaid piece we already covered, so we will not repeat it. The product margin is the one we see the most: when the hardware and the service sit in the same income line, you cannot tell how much of what looks like profit is really a pass-through you have marked up slightly. The service side can be sinking while the hardware side makes up for it, and the books would not reveal it. Service income and product income belong on separate lines.

The third is the concentration risk: the client that carries the month. In this trade, that usually means the shop with the one enterprise client on forty seats, or a client that signs a big project every year and pumps the rest. The community forums for small business owners are full of the same story: the largest client pays thirty days out against a fifteen-day invoice and never notices the late fee. Payment terms on the invoice are not the same as cash in the account. A good set of books makes the forecast show the gap before it appears, not after.

Which numbers should you watch?

Four numbers matter every month: the margin on the seat revenue, the size of the deferred pool, the unbilled backlog, and the actual cash window in front of you.

The seat margin is the recurring revenue from the monthly contracts, less the direct cost of keeping those clients running: the labor, the stack, the equipment piece. That is the number that tells you whether the flat per-seat price is actually enough. The published benchmarks for IT services firms put a healthy net margin in the ten to twenty percent range, and a margin report that does not get close to that is the signal that something is off.

The deferred balance is the block hours and the unconsumed annual payments still waiting to be earned. It tells you how much of today's cash is already spoken for, and it keeps you from spending profit that is not there.

The unbilled backlog is the work you have done in the month but have not sent an invoice for yet: the job hours on the job site, plus the migration work-in-progress. That is income that exists but is not on any statement, and it has no way to be something you track.

Last is the cash forecast. In a business with one or two large accounts that pay slow, the forecast is the only way to see the month when incoming invoices are looking good but the account is empty. The P&L will never express it. The cash projection exists for that month.

What should the books help you decide?

The books should make four decisions easier than they were before: when to hire, what a seat actually costs you, whether to take a project, and which client is the one that floats your month.

The hiring question comes out of the seat margin. You are comparing the fixed weekly cost of a new tech on payroll to the hours that come back from that employee against the variable hours you are buying from a subcontractor on the jobs. If the lines show that a share of every prevented was leaving as a subcontractor invoice, then a hire that covers the same hours is a decision the books can justify.

The seat price is the other decision. The margin report for a client that uses ten hours of support on twenty seats has a different direct cost than the client sitting quiet on the same twenty seats, but the flat fee was the same. One of them is paying for the other. The books show which is which, and that is what you take into the rate discussion at renewal.

Project acceptance is a simpler version of the same. The price of the job minus the subcontractor and the materials leaves the actual project margin. If the margin is there, the project floats normally; if the fixed fee is consumed by the work of another, the job is only keeping the calendar full. The numbers make that call without argument.

Related reading: Our Monthly Bookkeeping page for the ongoing work, Catch-Up bookkeeping if the last months are still behind, Clean-Up Bookkeeping if the current books have mistakes in them, and Custom Financial Reporting for the margin reports the standard P&L will not give you.