2026-10-11 · AI-drafted · Finance · Modeling · Accounting · 6 min read
Build a Three-Statement Model That Ties Out Every Time
Most financial models look convincing until someone changes an assumption — then the balance sheet stops balancing, the cash flow contradicts the income statement, and the DCF spits out a number nobody can defend. The problem is almost never the formula. It is the architecture. Build the model correctly from the first cell and the tie-out is not a check you run at the end; it is a property the model has by construction.
Why Models Break Before Anyone Touches a Scenario
A three-statement model links three financial statements — the income statement, the balance sheet, and the cash flow statement — so that a single change in one assumption flows through all three consistently. When that linkage breaks, the model produces numbers that are internally contradictory. A practitioner reviewing the output cannot trust any of it, even the parts that look correct.
The failure mode is almost always structural, not mathematical. Someone hardcodes a number that should be a formula. Someone builds the cash flow statement by copying line items instead of referencing the balance sheet movement. Someone adds a new debt tranche in one sheet and forgets to update the interest schedule in another. Each shortcut feels harmless in isolation. Together they produce a model that balances only under the original assumptions and collapses the moment a scenario changes.
The discipline that prevents this is not complicated, but it must be applied before the first formula is written. Decide where every input lives. Decide how every schedule feeds the next. Decide what a successful tie-out looks like. Then build in that order — inputs first, schedules second, statements third, outputs last.
The Single Source of Truth: Why Every Input Lives in One Place
Consider a manufacturer reporting in EUR under IFRS. Its model might reference revenue growth in the income statement build, in the working capital schedule, and in the DCF terminal value assumption. If the analyst types that growth rate in three separate cells, changing the assumption requires three edits — and missing one produces a model that is inconsistent with itself. The fix is a dedicated inputs sheet where every assumption lives exactly once, and every other sheet references that cell.
This principle — one cell, one truth — extends beyond growth rates. Tax rates, depreciation useful lives, days-sales-outstanding assumptions for receivables, days-payable-outstanding for payables, capital expenditure as a percentage of revenue: every driver belongs on the inputs sheet with a label, a unit, and a base-case value. Scenario analysis then becomes a matter of changing inputs, not hunting through formulas.
An Indian IT services firm modeling in INR might carry a dozen currency and cost assumptions that shift under different macro scenarios. A Brazilian retailer in BRL might model three inflation scenarios that affect both revenue pricing and cost of goods. In each case, the inputs sheet is the only place those assumptions appear. Every downstream formula is a reference, never a hardcoded number.
Schedules Are the Engine: How the Statements Get Their Numbers
The income statement does not generate its own depreciation figure. The depreciation figure comes from the asset schedule — a separate calculation that tracks the opening gross asset balance, adds capital expenditure, subtracts disposals, applies the depreciation method, and produces a closing net book value. The income statement references that output. The balance sheet references the same closing balance. When both reference the same schedule cell, they will always agree.
The same logic applies to debt. A company that carries a term loan and a revolving credit facility needs a debt schedule that tracks opening balances, drawdowns, repayments, and closing balances for each instrument. Interest expense on the income statement is a function of that schedule — average balance multiplied by the applicable rate. The balance sheet picks up the closing balance from the same schedule. No manual entry, no risk of disagreement.
Working capital follows the same pattern. Days-sales-outstanding drives the receivables balance: if a retailer with EUR 40m of annual revenue carries 45 days of receivables, the balance sheet shows EUR 40m × (45 ÷ 365) = EUR 4.93m. Change the DSO assumption on the inputs sheet and the receivables balance, the cash flow movement, and the balance sheet all update simultaneously. That is the model working correctly.
The Tie-Out: What It Is and How to Build It In
A three-statement model ties out when the cash and cash equivalents balance on the balance sheet equals the ending cash balance on the cash flow statement, and when total assets equal total liabilities plus equity — in every period, under every scenario. These are not aspirational targets. They are accounting identities. If they fail, something is wrong with the model's structure, not with the business being modeled.
The practical way to enforce this is a dedicated check row — often labeled 'Balance Check' — that subtracts assets from liabilities plus equity and displays the result. If the result is zero, the model is balanced. If it is not zero, the model flags the error before anyone reads the output. Some practitioners color this cell green for zero and red for anything else. The point is that the error is visible immediately, not buried in a printout.
Building the cash flow statement by reference — not by retyping — is what makes the tie-out automatic. Net income comes from the income statement. Depreciation and amortization come from the asset schedule. Working capital movements come from the difference between opening and closing balance sheet balances. Capital expenditure comes from the capex schedule. Debt movements come from the debt schedule. When every line is a reference, the cash flow statement cannot disagree with the other two statements.
Scenarios and the DCF: Where the Architecture Pays Off
A sensitivity table in Excel recalculates a single output — say, enterprise value from a DCF — across a grid of two assumptions, such as revenue growth and EBIT margin. That table is only meaningful if the model's architecture ensures that changing those two inputs flows correctly through every schedule and every statement. A model with hardcoded numbers produces a sensitivity table that is decorative, not analytical.
The DCF itself is the final output of a well-built model, not a separate spreadsheet. Free cash flow to the firm — earnings before interest and tax, adjusted for taxes, plus depreciation and amortization, less capital expenditure, less the change in working capital — comes directly from the schedules already built. The discount rate, typically the weighted average cost of capital, is a function of inputs already on the inputs sheet. The terminal value uses the same long-run growth assumption already defined. Nothing new is introduced at the DCF stage.
A German manufacturer with EUR 200m of revenue, a 12% EBIT margin, EUR 8m of annual depreciation, EUR 10m of capex, and a EUR 3m working capital build in a growth year produces approximately EUR 200m × 12% × (1 − 0.30 tax rate) + EUR 8m − EUR 10m − EUR 3m = EUR 11.8m of free cash flow in that year. Discount that stream at a 9% WACC over a 5-year explicit forecast period, add a terminal value, and you have a defensible valuation — one that updates automatically when any assumption changes.
What Course 2 Builds
Course 2 — Excel Modeling for Financial Decisions — walks through this architecture from the first sheet to the final DCF output. Each module adds one layer: workbook structure, the inputs sheet, the income statement build, the working capital schedule, the asset schedule, the debt schedule, the balance sheet and cash flow tie-out, scenario and sensitivity outputs, and the DCF. By the end, the model is complete and internally consistent.
The discipline taught in Course 2 applies regardless of the reporting currency, the accounting standard — IFRS or US GAAP — or the industry. A retailer, a manufacturer, a technology firm, and a services business all require the same underlying architecture. The numbers differ; the structure does not. That structure is what this course teaches.
The courses are pure education, not credentialing. We use AI heavily and we're transparent about it.
A Note on What This Course Is — and Isn’t
We don’t pursue CE accreditation. The courses are pure education, not credentialing.
Nothing in this course constitutes personalized financial, legal, or investment advice. You’ll learn frameworks and analytical tools — what you do with them is your decision.
We use AI heavily and we’re transparent about it.
Drafted with AI from the outline of Dr. Kareem Tannous’s Course 2. Published automatically under his editorial standards.
$189 per course. $504 for the bundle of three.
100% refund within 3 days of enrollment AND zero module access. Accessing any module — even briefly — waives the right to a refund permanently. Decisions are final; no appeals.
Course 2: Excel Modeling for Financial Decisions →
New to financial statements? Start with Course 1 →
Instructor: Kareem — DBA International Business · MS Applied Economics & Predictive Analytics · MBA Finance & Accounting · Series 65 · university-level instructor since 2014.
— Dr. Kareem Tannous