DCF model shell

How to use this structure

A discounted cash flow model is arithmetic wrapped around three judgements: the cash flows, the discount rate and what happens after the forecast ends. Everything below exists to keep those three visible instead of letting them dissolve into the spreadsheet.

The structure is written out tab by tab so you can build it from scratch or audit an existing model against it. No file is required, and none is offered. The value is in knowing what belongs where and which relationships have to hold.

Tab 1. Assumptions

Every input the model uses, in one place, each with a source or a stated rationale.

• Revenue drivers. Units, price, churn or renewal, and segment mix. A single growth percentage is not a driver, it is a conclusion.
• Margin build. Gross margin path with the reason it moves, operating expense by function, and the point at which operating leverage is assumed to appear.
• Capital. Maintenance and growth capital expenditure separately, depreciation policy, and working capital days for receivables, payables and inventory.
• Discount rate inputs. Risk free rate with its source and date, equity risk premium with its source, beta with the method used to derive it, cost of debt, tax rate and target capital structure.
• Terminal inputs. Perpetuity growth rate and, separately, an exit multiple with the basis it applies to.

Tab 2. Operating projection

A three statement build. The cash flow statement is the point of the exercise, so do not shortcut to an earnings proxy.

• Income statement. Revenue from drivers, gross profit, operating expenses, earnings before interest and tax, depreciation and amortisation shown separately, tax on unlevered earnings for the valuation view.
• Balance sheet. Working capital from days assumptions, fixed assets rolled forward, and enough of the capital structure to derive net debt at the valuation date.
• Cash flow. Operating cash flow, working capital movement, capital expenditure, and the resulting free cash flow before financing.
• Forecast horizon. Long enough that the business reaches a steady state before the terminal period begins. If margins or capital intensity are still moving in the final year, the terminal value is being asked to do work it cannot do.

Tab 3. Unlevered free cash flow bridge

State the bridge explicitly rather than assembling it inside one formula.

• Earnings before interest and tax.
• Less tax on those earnings at the unlevered rate.
• Plus depreciation and amortisation.
• Less capital expenditure, maintenance and growth shown separately.
• Less the increase in net working capital.
• Equals unlevered free cash flow, which is the only cash flow the enterprise discount rate should be applied to.

The single most common structural error in a DCF is discounting a levered cash flow at the weighted average cost of capital. Keep the levered and unlevered views on separate rows so the mismatch cannot happen quietly.

Tab 4. Discount rate build

Build the weighted average cost of capital from its components, each sourced and dated.

• Cost of equity. Risk free rate plus beta multiplied by the equity risk premium, with any size or country adjustment stated separately and justified rather than folded into the premium.
• Beta. Show the comparable set, the raw betas, the unlevering and relevering at the target capital structure. Do not present a single number without the derivation.
• Cost of debt. The rate actually achievable for this borrower, after tax, rather than the current coupon on legacy debt.
• Weights. Target capital structure at market values, not book values, with the target stated as a policy rather than backed into.
• Sourcing. Every market input needs a source and an as-at date. A discount rate assembled from remembered figures cannot be reviewed.

Tab 5. Terminal value, both ways

Calculate it under both approaches and show them side by side. Disagreement between them is information.

• Perpetuity growth. Final year unlevered free cash flow grown one period, divided by the discount rate less the growth rate. The growth rate must be defensible as a long run rate for the economy the business operates in, and should be tested for sensitivity rather than asserted.
• Exit multiple. The final year metric multiplied by a multiple that is justified by structurally comparable transactions, cited where terms are public.
• Cross-check. Derive the implied growth rate from the exit multiple, and the implied multiple from the growth rate. If the two approaches disagree materially, one of the assumptions is wrong and the memo should say which.
• Proportion test. Show terminal value as a share of total enterprise value. When the great majority of the value sits in the terminal period, the forecast is decoration and the reader should be told.

Tab 6. Valuation output

From enterprise value to a value per share, with every bridge item visible.

• Present value of the explicit forecast period, with the discounting convention stated as mid-year or year-end.
• Plus the present value of the terminal value, on the same convention.
• Equals enterprise value.
• Less net debt, less minority interests and other claims, plus non-operating assets, each itemised.
• Equals equity value, then divided by diluted shares using a stated dilution method.

Tab 7. Sensitivities

Two dimensional tables on the variables the answer is actually sensitive to.

• Discount rate against perpetuity growth rate.
• Discount rate against exit multiple.
• Revenue growth against steady state operating margin.
• A tornado view ranking the inputs by their effect on equity value, which is usually more informative than any single grid.

Checklist against classic modelling errors

• Discounting levered cash flow at the weighted average cost of capital.
• Double counting: financing effects appearing in both the cash flow and the discount rate, or synergies counted in the projection and again in the exit multiple.
• A terminal growth rate above the long run growth rate of the economy the business operates in.
• A forecast that ends before the business reaches steady state.
• Working capital assumptions that ignore seasonality.
• Mid-year and year-end discounting mixed between the forecast period and the terminal value.
• Market inputs with no source or no date attached.
• A terminal value that dominates enterprise value without the memo acknowledging it.

How this maps to Reuben AI

In Reuben AI, discount rate inputs, drivers and terminal assumptions are structured fields with a source and an as-at date attached, so a reviewer can see where the risk free rate came from and when it was last refreshed rather than trusting a hard-coded cell.

Because the valuation, the IC memo and the reporting pack read the same structured values, a change to an assumption flows everywhere it is used and is recorded with who changed it and when. Scenarios are named driver sets rather than duplicated files, which is what allows two valuations to be compared meaningfully.

Related templates and tools

• LBO model shell
• IC memo template: venture and growth
• Private-market glossary
• All investment templates