Valuation & Analysis Glossary

A DCF answers a single question: "If I bought this entire business today, what is a rational price to pay?" It does that in three steps. (1) Forecast the cash the business will generate every year for a defined explicit period (in this model, five years, FY2026–FY2030). (2) Estimate a "Terminal Value" that captures every year of cash after the explicit period, all the way out to infinity. (3) Discount every one of those future cash amounts back to today using a rate that reflects the time value of money and the risk of not receiving those cash flows. The DCF is a "fundamental" or "intrinsic" valuation — it does not depend on what other companies are trading for or what a buyer might pay in a hot market. It depends only on the cash the business itself is expected to produce and the risk of producing it.

If you can earn 10% per year risk-free, then $100 today grows to $110 in one year — meaning $110 next year is only worth $100 today. Reversing that logic: $110 received one year from now, discounted at 10%, has a "present value" of $100. Every future cash flow in a DCF is dragged back to today using this same logic, with the "10%" replaced by WACC (the risk-adjusted discount rate for this specific business).

Enterprise Value (EV) is what the "operating enterprise" is worth to all its capital providers — both debt and equity holders — before considering how the business is financed. Equity Value is EV minus net debt (debt minus cash) — it is what shareholders actually own. If EV = $75M and the company has $10M of debt and $2M of cash, equity value = $75M − $10M + $2M = $67M. The DCF produces EV first (because FCFF is a pre-financing cash flow — see below), then applies a "bridge" to arrive at equity value.

"Unlevered" means "before the effect of leverage (debt)." An unlevered DCF discounts Free Cash Flow to the Firm (FCFF, which is pre-interest) at WACC (which blends debt and equity cost). The output is Enterprise Value — the value of the operating business to all capital providers. We then subtract net debt to get to equity. The alternative — a "levered" DCF — discounts Free Cash Flow to Equity (FCFE, after interest) at the cost of equity alone. This model uses the unlevered approach because it isolates operating value from capital-structure choices.

Revenue is the top of the income statement — the total dollar amount the business collects from customers for delivered placements. "Net" means after taking out the amounts you never actually keep: canceled campaigns, make-goods (free placements given when something went wrong), volume discounts, refunds, and bad-debt provisions. Net revenue is what the DCF actually cares about because it is the true billable pool.

EBITDA strips out four things that are not part of core operations: (1) interest (a financing decision), (2) taxes (a jurisdictional decision), (3) depreciation (a non-cash allocation of past capex), and (4) amortization (a non-cash allocation of past intangibles). What's left is a rough measure of the cash the operating business generates before capital-structure and tax choices. Investors use EBITDA to compare businesses with different debt loads and tax rates on an apples-to-apples basis.

EBIT is EBITDA after subtracting D&A. It matters in a DCF because the tax authority does not tax EBITDA — it taxes EBIT (essentially). We compute a "tax on EBIT" figure to arrive at NOPAT, then add D&A back in the FCFF walk since D&A is a non-cash charge.

NOPAT is the starting point for FCFF. It answers: "If this business had no debt (and therefore no interest expense), what would its after-tax operating profit look like?" We use NOPAT rather than actual net income because we are computing an unlevered cash flow — we want to isolate the operating business from financing decisions.

When the company buys a $500k asset with a 5-year useful life, accounting rules require it to expense $100k per year for five years rather than $500k in year one. That $100k/year charge is depreciation (for tangible assets) or amortization (for intangibles). It reduces reported EBIT and reduces taxes — but no cash actually leaves the building in the year of the charge. In the FCFF walk we subtract D&A to compute EBIT (for tax purposes), then add it back because it is not a real cash outflow. Assets depreciated in this model: (1) computer hardware (3 yrs), (2) software / SaaS build-out (3 yrs), (3) production equipment (5 yrs), (4) leasehold improvements (5–7 yrs), (5) capitalized internal-use software (3–5 yrs).

CapEx is money out the door to buy productive assets — as opposed to expenses, which are consumed within a year. Because CapEx is capitalized (recorded as an asset, then depreciated), it does not hit EBIT in the year spent. That's why we subtract CapEx separately in the FCFF walk — it is a real cash outflow that never appeared as an expense.

As a business grows, it typically extends credit to customers (accounts receivable rises) and stocks up on prepaids — that ties up cash. It also gets credit from vendors (AP rises), receives prepayments from sponsors (deferred revenue rises), and accrues wages — that releases cash. The net effect is Δ NWC. In the FCFF walk we subtract an increase in NWC (cash consumed) and add a decrease (cash released).

FCFF is the "cash for everyone who financed the business" — before deciding how much to pay in interest or dividends. It is computed by taking NOPAT (after-tax operating profit as if there were no debt), adding back non-cash D&A, subtracting CapEx (real cash spent on assets), and subtracting Δ NWC (cash tied up in working capital). Every year of the 5-year forecast produces an FCFF value that gets discounted back to today.

If the business has losses in early years, those losses accumulate as an NOL balance that can be used to offset future profits — reducing (or delaying) cash taxes owed. The model tracks an NOL balance and applies it against pretax income before computing cash taxes. Book taxes (used to compute NOPAT) treat the effective rate as constant; the NOL only affects the cash-tax line and the balance sheet's deferred-tax posture.

WACC captures two ideas: (1) different investors demand different returns — bondholders accept less because they get paid first; equity holders demand more because they bear residual risk; and (2) debt is tax-advantaged because interest is tax-deductible. WACC weights each source of capital by its share of the capital structure and applies the after-tax cost. WACC is the "hurdle rate" — any project earning less than WACC destroys value; any project earning more creates value. In a DCF, WACC is applied to every year's FCFF to strip away the time-value premium.

Equity investors are last in line if things go wrong — they demand a higher return than debt holders. The standard framework (CAPM — Capital Asset Pricing Model) says the cost of equity equals a risk-free rate (what you'd earn on a Treasury bond) plus a premium for stock-market risk (the "equity risk premium"), scaled by how much this company's returns move with the market (beta), plus a "size premium" for smaller, less liquid businesses.

Usually the 10-year or 20-year Treasury yield. It anchors every other required return: no rational investor would put money into a risky business for less than they could get from a risk-free Treasury.

Historically the U.S. equity market has returned roughly 5–7 percentage points more per year than long-dated Treasuries. That gap — the ERP — represents the compensation investors require for the volatility and drawdown risk of owning equities as an asset class.

Beta scales the equity risk premium. A defensive utility might have a beta of 0.6 (less risky than the market); a high-growth software company might have a beta of 1.4 (more risky). For a private, early-stage company we typically use the "unlevered beta" of comparable public companies, re-levered to reflect the target capital structure.

Empirically, smaller-cap companies have delivered higher returns than large caps — reflecting greater business risk, less liquid stock, and less analyst coverage. Small-cap size premia are typically 2–5 percentage points added to the cost of equity for private companies of comparable scale.

For public companies, this is the yield on their outstanding bonds. For private companies, we estimate it from the rates of comparable borrowers or from actual quotes on credit facilities. We use the AFTER-tax cost of debt in WACC because interest is tax-deductible, effectively reducing the cost by the tax rate.

If a company has $20M of debt and $80M of equity market value, its capital structure is 20% debt / 80% equity. WACC uses market-value weights (not book) because we are pricing today's cost of capital, not historical cost.

If the valuation date is 6/30/2026, FCFF for FY2026 (year 1) has a discount period of 0.5 years under mid-year convention, or 1.0 year under year-end convention. FCFF for FY2027 has 1.5 or 2.0 respectively, and so on.

Under year-end convention, every dollar of FY2027 cash is treated as if it arrives on 12/31/2027 — clearly wrong, since the business actually earns cash all year. Mid-year convention assumes cash arrives on 6/30 (the midpoint) — a better proxy for reality. Mid-year discounting produces slightly higher present values (roughly half a year of discounting less on every cash flow).

A discount factor of 0.60 means "a dollar received in that future year is worth 60 cents today." Discount factors always decline as you move further into the future because more time = more compounding of the discount rate.

This is the value today of everything the business is expected to generate during the explicit 5-year forecast. It is typically 20–40% of Enterprise Value; the rest sits in the Terminal Value (see next section), because most of a growing business's value lies beyond year 5.

Because we cannot forecast every year forever, we condense the "long tail" into one lump sum called Terminal Value, valued as of the end of year N (year 5 in this model), and then discount that lump sum back to today at WACC. Terminal Value typically represents 50–80% of a growth company's Enterprise Value — meaning the DCF is extremely sensitive to terminal assumptions.

The Gordon Growth formula values a cash flow that grows at a constant rate forever. The perpetuity growth rate g must be lower than WACC (otherwise the formula produces infinity or nonsense). Sensible values: 2–3% (long-run inflation) for a mature market; 3–4% for a growing market. The model defaults to a moderate perpetuity rate; higher = higher TV = higher EV.

Instead of assuming cash grows forever, this method says: "At the end of year 5, we sell the business to a strategic or financial acquirer for X × EBITDA." The multiple is usually anchored to comparable public-company or precedent-transaction multiples. Exit-multiple TV is more market-based; perpetuity-growth TV is more theoretical. Best practice is to compute both, use the average, and cross-check the implied multiple of the perpetuity method (and vice versa) for sanity.

Terminal Value is stated as of the end of the explicit period (end of year 5 in this model). To make it comparable to today's dollars, we discount it back N years at WACC.

If Terminal Value is >85% of EV, the DCF is essentially valuing a business you can't yet forecast — a red flag suggesting either the growth period is too short or terminal assumptions are too rich. Healthy DCFs typically show terminal % between 50% and 75%.

The final EV is the sum of the two discounted components. Any change to WACC, terminal method, growth, exit multiple, or the underlying FCFF forecast flows through to this number.

Because FCFF is pre-financing, EV includes value belonging to debt holders. To get to equity we subtract the debt that would be repaid, and add back any excess cash on the balance sheet.

The Pivot convertible is a $10M loan whose interest accrues (Pay-In-Kind) rather than being paid in cash — the balance compounds at 10% annually. Principal plus all accrued PIK is due at the FY2028 maturity, at which point $10M × 1.10^3 = $13.31M is repaid in cash. Until then, the accreting balance sits on the balance sheet as debt and the accrued interest flows through the income statement as a non-cash interest expense. In the equity bridge, the outstanding balance of the convertible is subtracted from EV (like any other debt) in the years it is outstanding. If the note converts to equity at some agreed price before maturity, the debt disappears and new shares are added to the cap table — this diluted-share treatment is what gives the note its optionality.

Equity Value is the residual claim after paying off all debt-like obligations. Divided by fully-diluted shares outstanding, it gives implied share value. In this model, we present equity value at the valuation date; the cap-table waterfall on the Investor Returns tab shows how that value is distributed across preferred, common, options, and the Pivot convert (in its equity-converted state).

The model runs four scenarios: Conservative, Base, Growth, and Aggressive. Each is a full set of inputs (utilization, WACC, pricing, escalation, ramp) — not just a single slider. Switching scenarios re-computes the entire model: revenue, costs, statements, FCFF, DCF, and the returns waterfall. The scenario name shown on the Executive Dashboard is the "selected" scenario.

Utilization is the single most powerful lever in the model. At 30% utilization the business is selling roughly a third of its available slots; at 40% it is selling 40%. Because acquisition cost scales with capacity (not with slots sold), incremental utilization drops nearly straight to gross profit — making it the primary driver of EV differences across scenarios.

Music and brand placement prices escalate each year to reflect contract renewals, market growth, and inflation. To make utilization the primary driver of scenario differences, price escalation is held constant across all four scenarios — otherwise scenarios would be confounded and hard to compare.

Conservative uses 25% utilization and a higher WACC (reflecting greater perceived risk). Its EV is the lowest of the four scenarios by design. Use this scenario to test downside sensitivity and to demonstrate to investors what value looks like if the ramp underperforms.

Base is the "central estimate" and the scenario the model defaults to. Its 30% utilization ties directly to the 6.13 source proforma. All reconciliation tabs on the Income Statement use Base as the app-side comparison.

Growth reflects the outcome if the business achieves stronger adoption and demonstrates enough proof to warrant a modest re-rating (lower WACC). This scenario should produce an EV above Base — if it does not, an input assumption has broken the monotonicity of the scenario set (a common bug we watch for on the DCF check page).

Aggressive is the bull case: utilization at 40%, WACC compressed further, and (uniquely) data revenue turned on to reflect a fully monetized data product. This scenario produces the highest EV and highest MOIC/IRR. Use it to frame upside — but recognize that terminal % of EV rises with growth, so scrutinize the terminal assumptions carefully.

Example: Hold all inputs at Base except utilization; flex utilization from 20% to 50% in 5-point steps; report EV at each step. This isolates the "elasticity" of EV to utilization and helps identify which inputs are worth arguing about.

Two-way sensitivity is standard practice in valuation. The most common grids in this model: (1) WACC on the columns × terminal growth on the rows → EV in each cell; (2) WACC × exit multiple → EV; (3) utilization × price escalation → EV. Each cell shows what EV would be if the two named inputs took those values, with everything else held constant.

A tornado shows the effect of flexing each input across a defined range (e.g., ±20%) and ranks them by the magnitude of EV impact. The longest bars are the inputs to focus on in diligence. Typically utilization, WACC, and terminal growth top the tornado; below-the-line items like tax rate rarely make a material difference.

If a 1% absolute change in utilization moves EV by $2M and Base EV is $80M, then EV elasticity to utilization is 2.5% per 1% util. Elasticity metrics are useful for comparing sensitivity across inputs measured in different units.

MOIC is a pure "how much did I multiply my money" measure. It ignores time — a 3.0x MOIC in 2 years is dramatically better than a 3.0x MOIC in 10 years. That's why MOIC is always paired with IRR.

IRR is the discount rate at which the net present value of an investor's cash flows equals zero. A 25% IRR means "the investment grew at 25% per year." IRR is time-sensitive: earlier distributions produce higher IRRs even at the same MOIC.

In a typical VC-backed structure, proceeds flow: (1) debt (repaid first), (2) preferred equity (return of capital, sometimes with a preferred return), (3) common equity (residual), split pro-rata across the fully-diluted cap table. Convertible notes typically sit between debt and preferred, depending on their conversion status at the exit event. The Investor Returns tab shows this waterfall explicitly.

The cap table drives per-investor allocations. It includes founders, employee option pool, preferred investors, and the Pivot convert (in its as-converted state). Dilution occurs whenever new shares are issued — the pie doesn't grow, but each existing slice becomes proportionally smaller.

A 1x non-participating liquidation preference means preferred gets 1x their invested capital back before common sees a dollar; a 1x participating preference means preferred gets their 1x back and then also participates pro-rata in the remaining proceeds alongside common. Preference structures materially affect returns at low exits but converge to pro-rata at high exits.

If you own 20% of a company that then issues 25% new shares to a new investor, your stake dilutes to 20% ÷ 1.25 = 16%. Dilution is not inherently bad — if the new capital increases enterprise value enough, your smaller slice of a bigger pie is worth more than your larger slice of the smaller pie. But dilution modeling matters for accurate per-investor return math.