The financial reports every IT services CEO should actually read

Most founders of IT services companies run their business on two numbers: the balance in the bank account and whatever their accountant files with the tax office. Both feel like a grip on the finances, and neither one really is.

The bank balance tells you what happened, not why, and not what is coming. The tax filings exist to satisfy the state; they are built for compliance, not for decisions. You can read both carefully every month and still have no idea whether your company is actually healthy, which clients are draining you, or whether you can make payroll in March.

This article is the reporting stack we set up for IT services companies when we take on the CFO function. It walks through which reports matter, what each one answers, the metrics that generic bookkeeping never gives you, and how to read all of it without a finance degree. To keep it concrete, we will run one example company through every step.

Who this is for

You own or co-own a Ukrainian IT services company — custom development, consulting, an AI or data shop — with somewhere between 15 and 200 people. Most of your revenue comes from billable engineering time. Most of your costs are payroll and contractors. You bill clients abroad in dollars or euros, pay your team partly in hryvnia, and you may run a Ukrainian entity with a foreign company on top.

You are not a finance professional, and you do not want to become one. You want to look at the right numbers, understand what they are telling you, and make better decisions. That is exactly what a good CFO gives a CEO.

You have two sets of numbers, and only one helps you run the company

Statutory accounting is what your accountant prepares for the tax authorities. It has a fixed format, hard deadlines, and one purpose: keeping you compliant. An LLC (TOV) on the simplified system files tax on gross revenue, and that filing contains nothing about your margins, your project profitability, or your cash runway. A Diia.City resident files a compliance report to keep its status, not a dashboard to run the business.

Management accounting is the other set. It is optional, it follows no prescribed format, and it exists for one audience: you. It is forward-looking, refreshed monthly, compared against a budget, and sliced the way your business actually works, by project, by client, by team. Put plainly: statutory accounts keep you legal, management accounts help you run the company.

	Statutory / tax accounts	Management accounts
Audience	The state, tax office, regulators	You and your leadership team
Mandatory	Yes — fixed format and deadlines	No — any format you want
Direction	Backward (the past year)	Forward (where you are heading)
Frequency	Annual, plus periodic tax filings	Monthly or quarterly
Compared against	Prior year only	Budget, forecast, trends
Purpose	Compliance	Decisions

This is not about keeping a “second set of books” in any shady sense. It is internal analytics built from the same underlying data. There is a simple test: if the management report your accountant hands you looks identical to your statutory accounts, it is not a management report. Real management reporting shows monthly figures, percentage trends, comparison against the plan, and written commentary on what changed.

Everything below is management reporting.

The three statements, in plain language

Three reports form the backbone. Each answers a different question, and watching only one of them leaves you guessing about the other two.

Statement	The question it answers
Profit & Loss (P&L)	Are we making money?
Cash Flow	Can we afford to keep operating?
Balance Sheet	What do we own and owe right now?

The P&L (or income statement) shows revenue, costs, and what is left over across a period. This is where you read margin and profit trends month over month.

The Cash Flow statement shows money actually moving in and out. This is the one that tells you whether payroll and taxes will clear.

The Balance Sheet is a snapshot of what you own (cash, receivables) and owe (payables, debt) at a point in time. Small business owners use it least and underrate it most: it ties the P&L and Cash Flow together and surfaces bookkeeping errors the other two can hide.

Read them in that order: performance first, then survival, then the sense-check that confirms both.

Profit is not cash, and this is what kills companies

The thing to get straight first: your P&L can show a healthy profit while your bank account heads toward zero.

It happens because profit is recorded when you earn revenue, but cash arrives when the client actually pays. Deliver $40,000 of work in May on 60-day terms and your P&L shows the profit in May, but the cash does not land until July, while salaries, rent, and taxes are all due now. You look profitable on paper and feel squeezed in reality. Nearly two in five startups that fail do so because they run out of cash, not because they were unprofitable.

The P&L tells you how well you are performing. The Cash Flow statement tells you whether you can afford to keep performing. You need both.

Meet the example company

Here is a composite of a company we see often: an AI consultancy with about 20 people and $60,000 in monthly revenue, or $720,000 a year. The founder is busy, the team is busy, clients are happy. By the two numbers most founders watch, money in the bank and taxes filed, everything looks fine.

Now run it through management reporting.

Step one: split the team into billable and non-billable. This is the move a standard accountant never makes, and everything depends on it.

Group	People	Role
Billable delivery	~13	Engineers, ML/data, billable PMs — time clients pay for
Non-billable	~7	Founders, sales, ops, admin, internal R&D

Billable people are the cost of delivering revenue (COGS). Non-billable people are the cost of running the company (operating expenses). Tax books lump all 20 into one “salaries” line, which makes margin impossible to see.

Step two: build the real P&L. Using illustrative fully-loaded costs — roughly $3,000 a month for a billable person, $2,500 for a non-billable one, including salary, taxes, and tools:

Line	Monthly	% of revenue
Revenue	$60,000	100%
− Cost of delivery (13 × $3,000)	$39,000	65%
= Gross profit	$21,000	35%
− Operating expenses (7 × $2,500 + ~$5,000 tools, office, marketing)	$22,500	37.5%
= EBITDA	−$1,500	−2.5%

A 20-person company doing $60,000 a month, everyone flat out, running at roughly break-even and slightly in the red. Gross margin is 35%, against a healthy services benchmark of 50% or more.

More sales will not fix this company; better margins will. And nothing in the bank balance gives any hint of where those margins have gone.

The metrics generic bookkeeping never gives you

Gross margin is the result. The levers that move it are operational metrics your accountant does not track. These are what make reporting for an IT services business rather than reporting in general.

Utilization

Utilization is the share of your team’s available time that goes to billable work. It is the single best predictor of services profitability.

Our example has 13 billable people, each with roughly 160 working hours a month — about 2,080 available hours. Work backward from the revenue to see how hard those hours are working:

Average billed rate	Hours needed for $60k	Utilization
$40 / hour	1,500	72%
$50 / hour	1,200	58%
$60 / hour	1,000	48%

The healthy target for engineers is 75–85%. Anything below that gap is bench: people you pay a full salary who are not currently earning. Bench drains margin directly, and it is invisible on a statutory P&L. Push past 90% and you trade margin for burnout and slipping quality, so treat this as a band to land in rather than a number to chase.

Realization

Utilization tells you whether your team was busy. Realization tells you whether you got paid for it. It is billed hours divided by worked hours: log 100 hours on a project, invoice only 80, and your realization is 80%. The missing 20% leaked away through scope creep, “we’ll just fix that for free,” or under-scoping the deal. Healthy realization is 85–95%. Underpriced projects hide here. The team was busy, the client was happy, the work shipped, and you ended up charging $10,000 for $15,000 of effort.

Effective rate and revenue per head

A few more that belong on the list:

• Effective rate is delivery revenue divided by delivery hours: your real price after discounts and overruns. Compare it to your loaded cost per hour. In the example, $3,000 a month over 160 hours works out to about $19/hour of cost, so an effective rate of $40 is healthy and overruns dragging it down to $25 are not.
• Revenue per employee is a workforce-efficiency signal that should trend up year over year. IT services companies cluster around $0.3M per head in global benchmarks. The Ukrainian cost base is lower, so watch the trend rather than the absolute, and compare against peers your size.

None of these come out of standard bookkeeping, and all of them depend on time tracking. If your team is not logging hours against projects, start this week. Without it, every number above is a guess.

Cash you can see coming

For a services business with milestone invoices and foreign clients on net-30 or net-60 terms, cash deserves its own reports, separate from profit.

The 13-week cash flow forecast is your early-warning system. It is a simple weekly grid covering the next quarter: opening cash, expected client receipts, then payroll, taxes, tools, and contractor payments going out, ending in a projected balance for each week. Build the inflows from your actual outstanding invoices and their due dates rather than from hope, and segment clients by how they really pay, so a prompt payer lands on the due date and a chronically late one gets pushed two or three weeks out. Done properly, this shows you a payroll squeeze 8 to 10 weeks before it arrives, while you still have time to act.

Days Sales Outstanding (DSO) measures how long cash sits trapped in receivables: accounts receivable divided by revenue, times days in the period. If you have $90,000 in unpaid invoices on $60,000 of monthly revenue, your DSO is about 45 days, meaning a month and a half of billings is sitting outside your account at any given time. An aging report, showing who owes you and how overdue they are, turns that number into an action list.

Runway, done correctly, is cash divided by your average monthly burn over the last three months, not the last single month, which can mislead in either direction. Never count money that is not yet in the bank. “The big client will pay soon” is not runway. The classic blow-up is a founder who believes they have eight months because they are counting expected receivables and an anticipated investment, when the real figure, cash on hand over actual burn, is one month.

One more thing for Ukrainian companies specifically: you earn in dollars or euros and spend partly in hryvnia, often across two legal entities. Your management reporting should consolidate into one reporting currency with consistent exchange rates, so you can see your true position and your currency exposure in one place. Statutory hryvnia books hide both.

Which clients actually make money

Back to the example and its 35% blended gross margin. That figure is an average, and averages hide things. In almost every services company we look at, two or three clients are running at a loss, subsidized by the profitable rest.

You find them with a per-client (or per-project) P&L: revenue minus the fully-loaded cost of the hours that went into it, giving a margin for each engagement. Sort every client by margin and the spread is usually stark. The marquee logo you are proud of turns up at a 5% margin or a loss because you over-serviced it, while a small, undemanding client sits at 60%.

The usual culprit is contract structure. Fixed-price deals hand the client budget certainty and hand you all the estimation risk, so every overrun comes straight out of your margin. That is especially dangerous on the open-ended, research-shaped work typical of AI and data projects. Time-and-materials shares that risk and tends to run slightly higher margins because the client carries the overrun. A reliable pattern is a fixed-price discovery phase to scope the work, then time-and-materials delivery to build it.

Looking forward, not just back

Everything so far reports the past. A CFO also helps you see what is coming.

Skip the static annual budget that is obsolete by the second quarter and use a rolling forecast instead, refreshed every month and always looking 12 months ahead. Tie it to the drivers that actually move a services business: headcount, billable rate, utilization, and how reliably your pipeline converts to signed revenue. Headcount is your single largest cost lever, so model hiring carefully.

Then run a few scenarios off that base: what happens to cash and profit if you lose your largest client, if you hire four engineers ahead of demand, or if the exchange rate swings against you. The goal is not precision. It is having your next move ready before you need it.

Your CEO dashboard

You do not need a 50-page financial pack. You need one page, reviewed on a regular rhythm. More pages do not mean more understanding; a busy CEO needs the handful of numbers that drive decisions.

Weekly, 15 minutes:

• Cash balance and the 13-week forecast
• Accounts receivable aging — who owes, how late
• Utilization for the week

Monthly, after the books close:

• P&L: revenue, gross margin %, EBITDA — against last month and against target
• Realization, effective rate, revenue per billable head
• The three lowest-margin clients
• Runway in months

Quarterly:

• Re-forecast, re-price, and fix or fire the loss-making clients

For each metric, set a target, because a figure with nothing to compare it against tells you little. And ask your finance partner for insight, not just reports: what changed this month, what risk they see, what you should stop doing. The numbers rarely speak for themselves, and the commentary is what makes them useful.

What this is worth

Return to the example one last time.

That company sat at a 35% gross margin and roughly break-even EBITDA, losing about $1,500 a month. Suppose it does nothing more than tighten the operational levers the reports just exposed: lift utilization from the high-50s toward 75%, stop the realization leakage, and nudge underpriced rates up. Moving gross margin from 35% to 50% takes monthly gross profit from $21,000 to $30,000, and EBITDA from −$1,500 to roughly +$8,000.

That is a swing from losing money to a double-digit margin with zero new clients. Same team, same revenue. The reporting did not create that money; it showed the founder where it had been leaking out all along.

The mistakes we see most often

• Running on the bank balance alone, treating cash on hand as proof of health without asking whether it came from operations or from a prepayment you still owe work against.
• Only having statutory books, which reveal nothing about margin, projects, or runway.
• Confusing profit with cash, and getting blindsided by the gap between them.
• Miscalculating runway by using a single month’s burn, or counting money not yet received.
• Trusting blended margins that hide loss-making clients.
• Ignoring the bench, where idle engineers draw full salary against no revenue.
• Generic outsourced bookkeeping with no services context, which never produces utilization, realization, or project margin.

What good looks like

A few benchmarks to calibrate against. Treat them as rough direction rather than hard targets, since they lean toward cost bases higher than Ukraine’s. Compare yourself to peers at your size.

Metric	Healthy benchmark
Gross / delivery margin	50%+
Net profit margin	15–25%
Billable utilization (engineers)	75–85%
Realization rate	85–95%
Revenue per employee (IT services)	~$0.3M (global; lower in UA)

What matters more than any single figure is measuring it every month and watching which way it moves.

What we help with

At Harha, we build and run this reporting stack for IT services companies — the monthly P&L, cash flow, and balance sheet, the utilization and project-margin analysis, the 13-week cash forecast, and the one-page dashboard — and we explain every number in plain language. For founders who want the full function without a full-time hire, we run it as CFO-as-a-Service.

If you recognized your own company in the example above, the first step is a 30-minute call. Bring whatever numbers you have — even just revenue, headcount, and your bank balance — and we will show you where your margin is actually going.