A DCF is where you'll lean on Data Tables hardest. A discounted-cash-flow valuation is exquisitely sensitive to two assumptions nobody can pin down, the discount rate (WACC) and the terminal growth rate (g). That makes a single point-estimate enterprise value almost dishonest, because it hides how much the answer swings on two judgement calls. So every credible DCF is shown as a sensitivity grid, with enterprise value laid across a matrix of WACC (down the side) and g (across the top). A committee then sees the range and can find its own assumptions inside it.
For an explicit forecast of unlevered free cash flows FCF₁…FCF₅ and a discount rate r, the present value of the explicit period is Σ FCFₜ / (1+r)ᵗ. Beyond year 5 you assume a perpetual growth g and apply the Gordon terminal value, TV₅ = FCF₅ × (1+g) / (r − g), which you then discount back five years. Enterprise value is the sum:
EV = Σ_{t=1..5} FCFₜ/(1+r)^t + [ FCF₅ × (1+g)/(r − g) ] / (1+r)^5Advertisement
How do you build the grid with a Data Table?
Both r and g are single input cells, and the EV cell is a formula chain from them. That's the setup a two-variable Data Table needs. Put WACC values down the left column, g values across the top and a reference to the EV cell in the top-left corner, and Excel fills the grid with one EV per (WACC, g) pair. WACC is the Column input cell, because its values run down the column. Growth is the Row input cell, because its values run across the row. Get those two the wrong way round and the grid fills with plausible nonsense. It's the signature Data Table error, and nothing on screen will warn you. Worked example 1 builds this grid end to end and recomputes several of its cells by hand.
How do you read a DCF sensitivity grid?
Reading the grid is a skill in itself. EV falls as you move down, because a higher WACC discounts harder. It rises as you move right, because higher terminal growth inflates the tail. The bottom-left corner (high WACC, low g) is the conservative one, and the top-right (low WACC, high g) is the aggressive one. Analysts shade the cell holding the base-case pair. Then they quote a range, something like "EV of roughly ₹1,800 to 2,600 cr across reasonable assumptions", and leave the false-precision point out.