A three-statement model links your income statement, balance sheet and cash flow statement into one dynamic set of numbers. Change an assumption once and all three move together. Done well, it tells you whether the business can fund its plan, when cash runs tight, and how much you might need to raise.
Most founder models do not do that. They are a revenue forecast with some costs underneath, and a cash line that is a rough guess. The problems are common and fixable.
Building the P&L and bolting on the rest
The most frequent mistake is to build a profit and loss forecast first. The balance sheet and cash flow statement are then added afterwards, often by hand, with numbers adjusted simply to make the totals agree.
The result looks complete but does not behave like a real business. A change in sales does not flow through to receivables. A change in hiring does not flow through to cash. The statements are three separate documents sharing a file.
The fix is to design the links before you enter any numbers:
- Net profit from the income statement feeds retained earnings on the balance sheet.
- Balance sheet movements (receivables, payables, inventory, fixed assets, debt) feed the cash flow statement.
- Closing cash from the cash flow statement feeds back to the balance sheet.
If you build in that order, the model balances because the logic is right, not because someone forced it.
Assumptions buried in formulas
A model is only as credible as its assumptions, and in many founder models you cannot find them. A growth rate is typed directly into a formula. A hiring date is hard-coded in another cell. Soon, nobody remembers why.
This causes two problems. You cannot test the model, because changing an assumption means hunting through the file. And an investor or board member cannot see what the numbers rest on.
Keep every assumption in one clearly labelled inputs area. For each one, note:
- what it is and its unit (per month, per cent, days);
- where it came from (a contract, historic data, a market quote, your own judgement);
- who owns it.
This is not bureaucracy. It is the difference between a model people can challenge and one they have to take on trust.
Treating profit as cash
A profitable business can run out of cash. This is the single most important idea a three-statement model teaches, and the one founder models most often miss.
Profit is recorded when you earn revenue and incur cost. Cash moves when customers pay and you pay suppliers, staff, lenders and tax. The gap between the two sits in working capital and in items that never appear on the income statement.
Common sources of the gap:
- Receivables. If customers pay in 60 days, you fund two months of sales before you see the money.
- Inventory. Stock bought ahead of sales ties up cash.
- Payables. Suppliers may demand faster payment than you expect, especially early on.
- Deferred revenue. Annual upfront billing can help cash, but only if the model recognises it properly.
- Capital expenditure. Equipment and capitalised development costs leave the bank but only reach the P&L slowly through depreciation or amortisation.
- Debt repayments and tax. Principal repayments are not expenses, and tax is often paid on a different timetable from the one it is earned.
If your cash flow line is simply net profit plus depreciation, the model is missing most of what matters.
Ignoring working capital timing
Even founders who include working capital often use a single flat assumption, such as a fixed payment period applied to every customer for the whole forecast.
Reality is lumpier. Large customers pay late. Seasonal businesses build stock before peak months. Growth itself consumes cash, because each additional sale needs funding before it is collected.
Test the model against the awkward cases. What happens to cash if your biggest customer pays 30 days late? What if growth doubles but payment terms stay the same? A model that cannot answer these questions will not warn you in time.
Forcing the balance sheet to balance
A balance sheet that does not balance means there is an error in the logic. Some models hide this with a plug: a line labelled 'other' or 'adjustment' that absorbs the difference.
Plugs make the check pass while the model stays wrong. Avoid them entirely. Instead:
- add a visible check row that compares total assets with total liabilities plus equity, every period;
- add a second check that closing cash on the cash flow statement equals cash on the balance sheet;
- treat any non-zero result as a fault to trace, not a number to tidy away.
Revenue built only from the top down
Many models start from a market size and a target share: 'if we win a small percentage of this market, revenue is X'. That is a hope, not a forecast.
Build revenue from the drivers you control and can observe: leads, conversion, price, volume, churn, sales capacity. Use the top-down view as a sense check on the result, not as the source. If your bottom-up numbers need an implausible share of the market, that is useful information.
Funding that does not respond to the plan
Founders often type in a raise amount and date, then never check whether it is enough. A better model works out the funding need from the plan.
Set a minimum cash balance you are willing to hold, usually some months of operating costs. Then look at when cash falls below it, and by how much. That gives you the size and timing of the raise, with some margin, rather than a number chosen because it sounds right.
Watch the interest on debt too. If interest depends on average cash or debt balances, the model can become circular. Keep it simple until you need otherwise, for example by calculating interest on opening balances.
One scenario and no version control
A single set of numbers implies certainty that does not exist. At minimum, build a base case, a downside and an upside, switchable from the inputs area. The downside matters most. It shows how much room you have before the plan breaks.
Also keep a record of versions. When a model changes, note what changed and why. Otherwise board papers, investor decks and the working file drift apart.
A simple, robust structure
You do not need complexity to avoid these errors. A clean layout is usually enough:
- Inputs and assumptions. One place, clearly labelled.
- Calculations. Revenue, costs, headcount, working capital, fixed assets, debt.
- The three statements. Income statement, balance sheet, cash flow statement.
- Checks. Balance and cash reconciliations, flagged clearly.
- Outputs. A short summary of cash runway, funding need and key ratios.
Use monthly periods for the first two or three years, where cash timing matters, and annual periods after that. If you plan to use the model to raise capital or make major decisions, ask someone who did not build it to review it.
What to do next
Open your current model and run three quick tests. Can you find every assumption in one place? Does the balance sheet balance without a plug? If sales fall or customers pay late, does cash respond as you would expect?
If the answer to any of these is no, it is worth restructuring before the model is used for a raise, a budget or a major commitment. You can read more about our approach on the business planning and financial modelling page. If you would like to talk through your model or a planned rebuild, contact GainWells.