dcf-review-checklist

DCF Review Checklist

This skill is for FP&A analysts, controllers, deal teams, and valuation reviewers who need to challenge a discounted-cash-flow valuation they did not build. It does not rebuild the model. Instead it interrogates the assumptions and mechanics that most often drive a wrong answer: the discount rate, the terminal value, the discounting convention, internal consistency (nominal vs real, currency), and whether the implied exit multiple is defensible against market comparables. The output is a structured reviewer memo that ranks findings by their effect on enterprise value so the preparer fixes what matters first.

When to use this

• A colleague or counterparty sends you a DCF and asks "does this look right?"
• You are signing off on a valuation used for budgeting, planning, impairment testing (IAS 36), purchase price allocation, or a board paper.
• You suspect the equity value is being driven almost entirely by terminal value and want to test that.
• You need a second-opinion memo before a deal committee or audit committee meeting.
• You want a repeatable review checklist rather than ad hoc spot-checks.

Do not use this to produce a definitive valuation opinion or to replace a qualified valuation specialist.

Inputs

Provide whatever you have; the more of the model the better:

• The DCF workbook (.xlsx) or an export of the key tabs, OR a summary of assumptions in CSV form. The bundled sample samples/dcf-assumptions-sample.csv shows the minimum field set: one row per assumption with section, assumption, value, unit, and source.
• The explicit forecast cash flows (free cash flow to firm or to equity) by year, ideally with the build-up (EBIT, tax, D&A, capex, change in net working capital).
• WACC build-up: risk-free rate, equity risk premium (ERP), beta (levered/unlevered), cost of debt (pre/post tax), tax rate, and target capital structure weights.
• Terminal value method (Gordon growth or exit multiple), the terminal growth rate g, or the exit multiple used.
• The discount period convention (year-end or mid-year) and valuation date.
• The currency of the cash flows and whether figures are nominal or real.
• Any comparable-company or precedent-transaction multiples already gathered (EV/EBITDA, EV/EBIT, EV/Sales).

If a field is missing, flag it as an open item rather than guessing.

How to do it

Work the model from the discount rate outward, then test the terminal value, then mechanics, then consistency, then the market cross-check. For each finding, estimate the directional and (where possible) quantified impact on enterprise value (EV).

1. WACC components vs source

WACC = E/(D+E) * Ke + D/(D+E) * Kd * (1 - t).

• Cost of equity (CAPM): Ke = Rf + beta * ERP (+ size/specific premium if used). Check each input against a citable source:
  • Risk-free rate: should match a long-dated government bond yield in the SAME currency and roughly the SAME duration as the cash flows (commonly the 10Y or 20Y). A USD model using a EUR Bund yield is an error.
  • ERP: confirm it is a defensible market figure (broadly 4.5%-6.5% for developed markets in normal conditions) and consistent with the chosen Rf. Do not mix a "normalized" Rf with a "spot" ERP without disclosure.
  • Beta: confirm whether it is levered or unlevered, the comparable set, and that relevering uses the target (not current) capital structure: beta_L = beta_U * (1 + (1 - t) * D/E).
• Cost of debt: should reflect current marginal borrowing cost (risk-free + credit spread), not the historical coupon, and be tax-affected at the same marginal rate used elsewhere.
• Weights: target/long-run capital structure at market values, not book values, and not a single snapshot. Confirm weights sum to 100%.
• Tax rate: the marginal statutory (or long-run effective) rate, applied consistently in both the unlevered cash flow tax and the after-tax cost of debt.
• Recompute WACC from the stated components and compare to the rate actually used to discount. A 50-100bp unexplained gap is a material finding.

2. Terminal growth vs long-run GDP

For a Gordon-growth terminal value: TV = FCF_(n+1) / (WACC - g), where FCF_(n+1) = FCF_n * (1 + g).

• g is a perpetual nominal growth rate. It MUST NOT exceed long-run nominal GDP growth of the economy the firm operates in (a common guardrail is roughly 1%-3% real, plus expected long-run inflation; total nominal often 2%-4%). A perpetual g above nominal GDP implies the firm eventually becomes larger than the economy, which is impossible.
• Confirm g is consistent with the cash-flow basis: nominal g for nominal cash flows, real g for real cash flows (see step 4).
• Check the terminal-year cash flow is "normalized": capex should approximate D&A (a steady-state firm is not perpetually shrinking or hyper-investing), and net working capital change should scale with g, not sit at a forecast-year spike.
• Confirm WACC > g; if WACC - g is small (say < 2%), note the high sensitivity: TV swings sharply with tiny changes in either input.
• Compute the terminal value as a share of total EV. If TV is >70%-75% of EV, flag that the valuation rests largely on perpetuity assumptions and recommend an exit-multiple cross-check.

3. Discounting convention (mid-year vs year-end)

• Identify the convention. Year-end discount factor for year t = 1 / (1 + WACC)^t. Mid-year = 1 / (1 + WACC)^(t - 0.5), reflecting cash received roughly evenly through the year.
• Mid-year convention raises PV by approximately (1 + WACC)^0.5, i.e. roughly +2%-3% on EV at a 5%-6% WACC. It is a legitimate choice but must be applied CONSISTENTLY to the explicit period AND the terminal value.
• A frequent error: mid-year on the explicit cash flows but year-end on the terminal value (or vice versa). Check the terminal-value discount period matches the convention.
• Confirm the first discount period aligns with the valuation date (a stub period if the valuation date is mid-year) and that periods are not off by one.

4. Nominal/real and currency consistency

• Inflation basis: nominal cash flows must be discounted at a nominal WACC; real cash flows at a real WACC. Mixing them double-counts or omits inflation. The Fisher relation: (1 + nominal) = (1 + real) * (1 + inflation).
• Confirm revenue growth, cost inflation, capex, and terminal g are all on the same basis.
• Currency: the cash flows, the risk-free rate, the ERP/country risk, and g must all be in the SAME currency. If cash flows are in a foreign currency, either (a) discount at a WACC built in that currency and convert the result at spot, or (b) convert cash flows to the home currency using forward rates (interest-rate-parity-consistent) and discount at the home WACC. Do not convert at spot AND use a foreign-currency discount rate.
• Check that any country-risk premium is added once and not double-counted in both the ERP and a separate adjustment.

5. Implied exit multiple cross-check

• Even for a Gordon-growth model, back out the implied exit multiple: implied EV/EBITDA = TV (undiscounted, at end of year n) / terminal-year EBITDA. Compare to current trading comparables and precedent transactions.
• A Gordon-growth g of 2.5% at a 9% WACC implies a forward EV/EBITDA roughly in the high single digits to low teens depending on the EBITDA-to-FCF conversion; if the implied multiple is wildly above the comparable set, the g (or margin/capex normalization) is too aggressive.
• Conversely, sense-check an exit-multiple TV by deriving the g it implies and confirming it is below nominal GDP.
• Reconcile the two methods: a credible model lands within a reasonable band under both.

6. Rank and quantify

For each finding, classify impact as High / Medium / Low using a rough EV sensitivity:

• WACC and g are usually highest-leverage (a 50bp move in either often shifts EV by mid-single to double-digit percentages, especially when WACC - g is small).
• Discounting convention and stub-period errors are typically low-to-medium (low single-digit percent) but easy to fix.
• Document the estimated EV impact (percent or range) so the preparer prioritizes correctly.

Output

Produce a Markdown reviewer memo with these sections:

1. Header - valuation reviewed, valuation date, currency, basis (nominal/real), preparer, reviewer, review date, headline EV and equity value.

2. Summary of findings - a table ranked by impact:

   #	Area	Finding	Estimated EV impact	Severity (H/M/L)	Recommended action

3. Detailed observations - one short paragraph per finding, citing the standard/guardrail and the source the assumption should match.

4. Open items - missing inputs needed to complete the review.

5. Reviewer's overall conclusion - whether the valuation is fit for its stated purpose subject to the actions above.

If the user prefers, also emit a CSV of the findings table for tracking. Do NOT alter the source workbook; the memo is a separate deliverable.

Quality checks

• Recompute WACC from components yourself; do not trust the labelled cell.
• Confirm WACC and g are on the same currency and inflation basis before commenting on g.
• Verify the discounting convention is identical for explicit cash flows and terminal value.
• Check TV as a percentage of EV and always run the implied-exit-multiple cross-check.
• Avoid these common errors: relevering beta at current rather than target gearing; tax-affecting cost of debt at a different rate than the cash-flow tax; off-by-one discount periods; perpetual g above nominal GDP; spot-rate currency conversion combined with a foreign-currency discount rate; double-counting country risk.
• Quantify impact wherever a number is available; never report a finding as "wrong" without stating the direction and rough magnitude on EV.

Worked example

Using samples/dcf-assumptions-sample.csv for Aurora Components Ltd (a fictional EUR-reporting manufacturer):

• The model uses Rf = 2.8% (EUR 10Y govvie) and ERP = 5.5%, beta_L = 1.10, giving Ke = 2.8% + 1.10 * 5.5% = 8.85%. Cost of debt 4.5% pre-tax, tax 25%, so Kd post-tax = 3.38%. With weights E 70% / D 30%: WACC = 0.70 * 8.85% + 0.30 * 3.38% = 7.21%. The workbook discounts at 7.0% - a ~20bp unexplained gap (Low-Medium; ask the preparer to reconcile).
• Terminal growth g is 3.0%. Long-run EUR-area nominal GDP growth is closer to 2.5%-3.0% (roughly 1.0%-1.5% real plus ~2% inflation), so g sits at the very top of the defensible range - flag as Medium and recommend 2.0%-2.5% as a base case, noting EV sensitivity since WACC - g is only ~4%.
• TV is 78% of EV - High concentration; mandate the implied-multiple cross-check. Backing out TV / terminal EBITDA gives an implied EV/EBITDA of ~11.5x against a comparable set trading at 8x-10x, confirming g (and a possibly under-normalized capex) is too generous.
• Convention: mid-year applied to explicit years but year-end on the terminal value - inconsistent; correct the TV discount period (Medium).

The memo would rank these as: (1) terminal growth / implied multiple [High], (2) mid-year inconsistency [Medium], (3) WACC reconciliation [Low-Medium], with estimated EV impact ranges for each.

Disclaimer

Disclaimer: This skill is a drafting and analysis aid, not professional advice. It does not provide accounting, audit, tax, investment, or legal advice. All output must be reviewed and approved by a qualified professional before use or reliance.

---

Part of LedgerSkills - security-vetted Claude Code skills for finance teams. New skills scanned and tested weekly; subscribe on the site.