Building the income statement

The income statement is the first of the three statements and the one every beginner is sure they can build — revenue at the top, costs in the middle, profit at the bottom; everyone has seen one. The difference between that and a modeller's income statement is not the shape. It is that every single line is built from a driver you can change, and that two of the lines are not really the income statement's at all — they arrive from schedules you build later. Get those two ideas right and the statement becomes a live engine. Get them wrong and you have typed a table that looks like a forecast and behaves like a photograph.

Build from a driver, never a typed number

A driver is the rate or quantity that generates a line: a growth rate that produces revenue, a margin that produces gross profit, a percentage that produces operating expense. A typed number is the alternative — a value keyed straight into the cell — and it is dead the moment you write it, because it has no relationship to anything and will not move when the world you are modelling changes. For Sokoni you drive revenue with a single 10% growth rate, which is the right simplification for learning the mechanics. In a real forecast you would usually go one level deeper and split revenue into volume times price — units sold times price per unit — because that is where the business actually lives and where a top-line story can be tested for arithmetic possibility (the bottom-up discipline the DCF course builds out in full). Either way the principle holds: revenue is generated by a driver, never asserted as a figure. The same is true of every line beneath it.

Sokoni's top line, gross profit, and EBITDA

Start at the top. Sokoni's base-year revenue is KES 1,000,000 and the single revenue driver is a 10% growth rate, so Year 1 revenue is 1,000,000 x 1.10 = 1,100,000, Year 2 is 1,210,000, and the line compounds to 1,610,510 by Year 5 — each cell the previous year times one-plus-the-growth-input, one formula copied across, exactly Module 2's cardinal rule. Gross margin is 40%, so gross profit is 40% of revenue — 440,000 in Year 1, rising to 644,204 in Year 5 — and cost of goods sold is the other 60%, from 660,000 to 966,306. Operating expenses run at 20% of revenue, 220,000 in Year 1. Subtract them from gross profit and you have EBITDA: 220,000 in Year 1, which is exactly 20% of revenue, because a 40% gross margin less 20% of revenue in opex leaves 20%. Notice what did not happen: not one of those numbers was typed. Each is a driver — a rate — applied to a line above it.

EBIT, interest, and the two forward references

EBITDA is earnings before interest, tax, depreciation, and amortisation — operating profit while you still pretend the assets were free. EBIT is what remains after you charge for using them up: EBIT = EBITDA minus depreciation. But where does depreciation come from? Not a typed number. It is 10% of Sokoni's opening net PP&E, and that balance rolls forward with capex and prior depreciation on a PP&E schedule you build in a later module. So in the income statement you leave a forward reference — a clean link to a schedule cell you have not populated yet — and when you wire the schedule up, Year 1 depreciation resolves to 50,000 and EBIT to 170,000. Interest works the same way: it is 12% of opening debt, calculated on the debt schedule, and it too enters the income statement as a link, resolving to 36,000 in Year 1. Building forward references deliberately — links to cells you will fill in, never hardcoded placeholders you mean to fix later — is how the three statements end up genuinely integrated rather than three tables that happen to sit near each other.

From EBIT the rest is arithmetic — but arithmetic that still flows from drivers. Subtract interest to reach profit before tax: 170,000 minus 36,000 = 134,000 in Year 1. Apply the 30% marginal tax rate — a charge of 40,200 — and net income is 93,800. Do the same across the row and net income rises from 93,800 in Year 1 to 156,778 in Year 5. Two things are worth watching. Interest falls every year — 36,000, 30,000, 24,000, 18,000, 12,000 — not because you assumed it would, but because Sokoni repays 50,000 of debt a year and interest is charged on the declining opening balance; the debt schedule drives it. And net income does not stop here: 20% of it is paid out as dividends and the remaining 80% is retained, a number the balance sheet will collect as retained earnings — another link, running the other way, that you connect when you build the balance sheet.

• Revenue — 1,100,000 / 1,210,000 / 1,331,000 / 1,464,100 / 1,610,510
• Cost of goods sold (60% of revenue) — (660,000) / (726,000) / (798,600) / (878,460) / (966,306)
• Gross profit (40% margin) — 440,000 / 484,000 / 532,400 / 585,640 / 644,204
• Operating expenses (20% of revenue) — (220,000) / (242,000) / (266,200) / (292,820) / (322,102)
• EBITDA — 220,000 / 242,000 / 266,200 / 292,820 / 322,102
• Depreciation (from the PP&E schedule) — (50,000) / (58,200) / (66,900) / (76,182) / (86,133)
• EBIT — 170,000 / 183,800 / 199,300 / 216,638 / 235,969
• Interest (from the debt schedule) — (36,000) / (30,000) / (24,000) / (18,000) / (12,000)
• Profit before tax — 134,000 / 153,800 / 175,300 / 198,638 / 223,969
• Tax at 30% — (40,200) / (46,140) / (52,590) / (59,591.4) / (67,190.7)
• Net income — 93,800 / 107,660 / 122,710 / 139,046.6 / 156,778.3

Why the margin must be defended, not assumed

Sokoni's model holds gross margin flat at 40% and opex flat at 20% of revenue across all five years. That looks neutral. It is not — it is a claim, and you should be able to defend it. Holding gross margin flat asserts that input costs, competition, and pricing power all stay in balance as the business grows. Holding opex at a constant percentage of revenue asserts that costs scale exactly in line with sales — no economies of scale, no cost creep. Reality usually bends both. Some operating costs are fixed — head-office salaries, core systems, rent — so as revenue grows they shrink as a share of it, and the operating margin drifts up: this is operating leverage, and it is real. But it is almost always smaller than a naive model implies, because growth also demands fresh spending on people, marketing, and capacity, and competition tends to compete the gains away. The point is not that flat is wrong; it is that flat, up, and down are all assertions about the world, and each needs a sentence of justification you could say out loud to the person relying on the number.