Valuation and Forecasting
Forecasting discipline
A credible forecast is driver-based and auditable. Break revenue into the smallest reliable unit - units × price, subscribers × ARPU, square footage × sales per square foot - and state each assumption in one sentence so a reviewer can challenge it. Separate fixed from variable costs so operating leverage is modeled rather than asserted; never assume margin expansion without a mechanism (mix, pricing, scale, or a cost program). Forecast the balance sheet and cash flow too: working capital, capex, dividends and buybacks (which change share count), and debt maturities all feed equity value.
Horizon and terminal value
Forecast until the business reaches steady state - growth near the terminal rate and margins near mid-cycle. Two terminal methods: perpetuity growth, FCF(n+1) ÷ (WACC − g), and an exit multiple. Cross-check both: the implied exit multiple should resemble today's peer multiples, and implied perpetual growth should not exceed nominal GDP growth.
Scenarios and sanity checks
Probability-weight bull, base, and bear cases rather than presenting one point. Then run reverse checks: what growth, margin, or multiple does the current price imply? What must be true for the upside case? A price target is a 12-month expected price, not intrinsic value - name the catalysts and the horizon explicitly.
Traps
- Extrapolating peak cyclical margins or a one-off growth spurt.
- Setting g at or above WACC, or above long-run nominal GDP.
- Forgetting share-count changes from buybacks, option exercises, or converts.
- Terminal value quietly driving most of the answer.
- Confusing a target price with a valuation range, or omitting dividends from total-return math.
Rule of thumb
Perpetuity multiple = 1 ÷ (WACC − g). At WACC 9% and g 3%, the terminal multiple is 16.7x FCF; if that sits far above peer multiples, revisit the assumptions.