01 — Forecast structure
Lay out the forecast before valuing anything
A DCF is a forecast of operating cash flows, a discount rate and a terminal value. Build the three in that order, on one timeline.
Put assumptions, the operating forecast, free cash flow, discounting and the valuation bridge in separate labelled blocks on one consistent timeline, so a reviewer can trace every output back to an input.
Quick rules
- One timeline across the model: the same period columns for the income statement, cash flow build and discounting.
- Historical periods to the left, forecast periods to the right, with the last actual year marked so growth rates have a visible base.
- Assumptions (growth, margins, tax rate, capex, working capital, WACC, terminal growth or multiple) live in one input block and are referenced, never retyped.
- Forecast until the business reaches a steady state: five years is common, longer for a business still growing quickly or cycling through heavy investment.
- Enterprise value first, equity value second, per-share value last. Each step is its own labelled line.
How to fix it
- Draw the flow on paper: revenue → EBIT → NOPAT → unlevered free cash flow → present value → terminal value → enterprise value → equity value. Give each arrow a row.
- Move every typed number out of calculation rows into the assumption block and replace it with a reference.
- Label the period basis (fiscal year end, months in the stub year, actual or forecast) once at the top of the timeline.
02 — Free cash flow
EBIT, NOPAT and unlevered free cash flow
Unlevered free cash flow is the cash the operations generate for all capital providers before any financing.
Start from EBIT, tax it at the operating tax rate to get NOPAT, add back non-cash charges, then subtract capital expenditure and the increase in net working capital. Do not deduct interest, and do not use net income as the starting point.
Quick rules
- NOPAT = EBIT × (1 − tax rate). Tax the operating profit, not profit after interest.
- UFCF = NOPAT + depreciation and amortisation − capital expenditure − increase in net working capital.
- An increase in net working capital is a use of cash and reduces UFCF; a decrease adds to it. Get the sign from the balance-sheet movement, not from memory.
- Stock-based compensation, impairments and other non-cash items need a deliberate decision, stated in the assumptions, on whether they are added back.
- Keep the cash flow unlevered: interest, debt repayment and dividends belong in the bridge or the financing section, not in UFCF.
How to fix it
- Worked example (invented): EBIT 150, tax rate 25%, D&A 40, capex 55, net working capital rises by 12. NOPAT = 150 × 0.75 = 112.5. UFCF = 112.5 + 40 − 55 − 12 = 85.5.
- Check that the tax line in the cash flow build is EBIT × tax rate, not the income statement's tax expense, which is after interest.
- Reconcile D&A and capex to the fixed-asset schedule so the same figures appear in the income statement, the cash flow and the DCF.
- Test the sign of the working capital line with a deliberate change: raise the receivables days assumption and confirm UFCF falls.
03 — Discount rate
WACC: cost of equity, cost of debt and weights
The discount rate must match the cash flow. Unlevered cash flows are discounted at the weighted average cost of capital, not at the cost of equity.
Compute the cost of equity with CAPM, the after-tax cost of debt from the borrowing rate and the tax rate, and weight the two by a target capital structure at market values. Use the same currency, inflation basis and tax rate as the forecast.
Quick rules
- Cost of equity = risk-free rate + β × equity risk premium. Add a size or country premium only when the assumption block states one.
- After-tax cost of debt = pre-tax cost of debt × (1 − tax rate). Interest is deductible, so the shield sits in the discount rate, not in UFCF.
- WACC = E ÷ (D + E) × cost of equity + D ÷ (D + E) × after-tax cost of debt, with D and E at market values or a stated target structure.
- Unlever peer betas at their capital structures and relever at the target structure before using them: βL = βU × (1 + (1 − t) × D ÷ E).
- A nominal forecast takes a nominal risk-free rate; a real forecast takes a real one. Do not mix.
How to fix it
- Worked example (invented): risk-free rate 4.0%, β 1.2, equity risk premium 5.5% → cost of equity = 4.0% + 1.2 × 5.5% = 10.6%. Pre-tax cost of debt 6.0% at a 25% tax rate → 4.5% after tax. With 70% equity and 30% debt, WACC = 0.7 × 10.6% + 0.3 × 4.5% = 8.77%.
- Store every WACC input as a separate labelled assumption and build the calculation in visible steps rather than in one long formula.
- If the case gives peer betas, keep the unlever and relever steps on separate rows so a reviewer can see which capital structure was used at each step.
- Read the resulting WACC back against the cost of equity and the after-tax cost of debt: it must sit between them.
04 — Terminal value
Terminal value: perpetuity growth and exit multiple
Terminal value usually carries most of the enterprise value, so its method, timing and inputs need to be explicit and cross-checked.
Calculate terminal value with the perpetuity growth method on the year after the final forecast year, or with an exit multiple applied to a final-year metric, and show the implied multiple or implied growth rate from the other method as a cross-check.
Quick rules
- Perpetuity growth: TV = UFCF in the final forecast year × (1 + g) ÷ (WACC − g). The growth rate must be below WACC and no higher than long-run nominal economic growth.
- Exit multiple: TV = final-year EBITDA (or EBIT) × the chosen multiple, with the multiple justified by comparable companies or transactions.
- Discount the terminal value from the end of the final forecast year, using the same discount factor as the final year's cash flow under end-of-year discounting.
- Always show the implied exit multiple when using perpetuity growth, and the implied perpetuity growth when using an exit multiple.
- The final forecast year should be a steady state: capex close to depreciation, stable margins and a working capital movement consistent with the growth rate.
How to fix it
- Worked example (invented): final-year UFCF 100, WACC 9%, g 2% → TV = 100 × 1.02 ÷ (0.09 − 0.02) = 1,457. If final-year EBITDA is 180, the implied exit multiple is 1,457 ÷ 180 = 8.1x.
- Check WACC − g in its own cell; if it is small or negative, the terminal value is unreliable and the assumptions need revisiting.
- State which cash flow the growth is applied to. Growing the final-year cash flow by (1 + g) is the standard form; using the final-year figure without growth understates the value.
- Show terminal value as a share of enterprise value. A figure above roughly 75% is not wrong, but it tells the reviewer where the sensitivity lives.
05 — Valuation bridge
From enterprise value to equity value and per share
Enterprise value belongs to all capital providers. Subtract the claims that rank ahead of ordinary shareholders and add non-operating assets.
Equity value = enterprise value − debt and debt-like items − preferred equity − non-controlling interests + cash and non-operating assets. Divide by diluted shares to reach value per share.
Quick rules
- Use balance-sheet figures at the valuation date, not forecast figures, for debt, cash and other bridge items.
- Debt-like items include finance leases, pension deficits and other fixed claims; include them consistently with how the comparable multiples were built.
- Add cash only to the extent it is not needed for operations, if the case distinguishes operating cash; otherwise add all cash and say so.
- Non-controlling interests are subtracted because the enterprise value covers 100% of consolidated operations while shareholders own less.
- Use diluted shares: options and warrants that are in the money at the implied price via the treasury stock method, and convertibles that are in the money via the if-converted method, where the case provides the data.
How to fix it
- Worked example (invented): enterprise value 1,200, debt 300, cash 80, non-controlling interests 20, diluted shares 100. Equity value = 1,200 − 300 + 80 − 20 = 960. Value per share = 9.60.
- Lay the bridge out vertically with one item per row and an explicit sign convention; the reviewer should be able to add the column.
- Keep the bridge items referenced to the balance sheet or the assumption block rather than typed in.
- If value per share depends on diluted shares that depend on value per share, use a small iterative block or a stated simplifying assumption; do not hide the loop.
06 — Sensitivities
Sensitivity tables that are live and readable
A DCF output without a sensitivity table hides how much of the value rests on two assumptions.
Build a two-way data table of value per share (or enterprise value) against WACC and the terminal growth rate or exit multiple, centred on the base case, driven by the model's real inputs, and formatted so the base case is visible.
Quick rules
- Use Excel's Data Table feature (What-If Analysis) or an equivalent formula grid so the table recalculates from the live model rather than pasted values.
- Row and column inputs must be the actual assumption cells the model uses; a table pointed at a copy of the input does nothing.
- Centre the table on the base case and step the axes in sensible increments: 0.5 percentage points for WACC, 0.25 to 0.5 points for growth, 0.5x to 1.0x for a multiple.
- Mark the base case cell and keep the table on the same sheet as the inputs it references, which Excel's data table requires.
- Set calculation to automatic including data tables, or state in the model that tables must be refreshed with F9.
How to fix it
- Worked example (invented): with a base case of WACC 9% and growth 2%, step WACC from 8% to 10% across the columns and growth from 1% to 3% down the rows. The corners will sit well above and well below the base case; a wide range is normal and is the point of the table.
- Verify the centre of the table equals the model's headline output; if it does not, the row or column input cell is wrong.
- After any structural change to the model, check the table still recalculates by nudging an input and watching the corner cells move.
- Format the table with the same number format as the headline output and shade the base case.
07 — Review
Common DCF errors and how to catch them
Most DCF mistakes are mismatches: the wrong cash flow for the discount rate, the wrong year for the terminal value, the wrong figures in the bridge.
Before handing over, reconcile each pairing in the model: cash flow to discount rate, terminal value to its base year and discount factor, bridge items to the balance sheet, and the sensitivity table to the headline output.
Quick rules
- Discounting UFCF at the cost of equity, or levered cash flows at WACC, mixes the cash flow with the wrong claim holders.
- Terminal growth at or above WACC, or above long-run economic growth, produces a value that cannot be defended.
- Applying the perpetuity formula to the final-year cash flow without growing it, or discounting the terminal value one year too many or too few.
- Using forecast debt and cash in the bridge, or omitting non-controlling interests, leases or preferred equity that the comparable multiples include.
- Deducting interest in UFCF, using the income statement tax charge instead of tax on EBIT, or getting the working capital sign backwards.
- Sensitivity tables that point at a copy of the input, or that were pasted as values and no longer move.
How to fix it
- Add a checks block: WACC between cost of equity and after-tax cost of debt; WACC − g positive; each period's discount factor lower than the one before; terminal value share of enterprise value shown; sensitivity centre equal to the headline output.
- Trace one full path with the formula bar from an assumption to value per share and confirm no typed values interrupt it.
- Read the implied exit multiple and implied growth rate against the comparable set before accepting the answer.
Final pass
Checks before handover
- 01Assumptions sit in one block and every calculation references them.
- 02UFCF starts from EBIT, taxes operating profit and excludes all financing flows.
- 03WACC inputs are labelled, the result sits between the cost of equity and the after-tax cost of debt, and the basis matches the forecast.
- 04Terminal value uses a stated method, a growth rate below WACC and the correct base year, with the implied cross-check shown.
- 05The bridge uses valuation-date balance-sheet figures with an explicit sign per row and diluted shares.
- 06The sensitivity table is live, centred on the base case and matches the headline output.
- 07A checks block flags the mismatches above before delivery.
StreetCred practice cases assess submitted workbooks and return cell-level feedback.
Start the free diagnostic →