Introduction
Stock-based compensation is the most debated EBITDA adjustment in investment banking. Unlike restructuring charges or litigation costs (which are unambiguously non-recurring), SBC is a recurring, intentional cost that companies choose to incur as part of their compensation strategy. It reduces reported earnings through a non-cash accounting charge, but the economic impact is real: it dilutes existing shareholders by creating new shares.
The Basic Mechanics
When a company grants stock options, restricted stock units (RSUs), or other equity-based compensation to employees, accounting standards (ASC 718) require the company to record an expense on the income statement equal to the fair value of the grant, spread over the vesting period. This expense reduces reported EBITDA and net income.
The expense is non-cash: no cash leaves the company when it grants RSUs or options. The "cost" is borne by existing shareholders through dilution. When those RSUs vest and convert to common stock, the total share count increases, reducing each existing share's proportional claim on earnings and value.
The magnitude of SBC varies dramatically by sector and company maturity. Alphabet recorded $27.1 billion of SBC expense in 2025, according to its 2025 annual report on Form 10-K, yet that was under 7% of its $402.8 billion of revenue. At the other end, Snowflake's SBC of roughly $1.6 billion in fiscal 2026 equalled 34% of revenue and was larger than its entire $1.4 billion GAAP operating loss, per Snowflake's fiscal 2026 10-K, so the add-back alone turns a deep operating loss into a profit. Younger, high-growth software companies typically sit at the upper end because employees who join early expect equity participation.
- Stock-Based Compensation (SBC)
A non-cash expense recognized on the income statement for equity awards (stock options, RSUs, performance shares) granted to employees as part of their compensation. SBC is measured at grant-date fair value and expensed over the vesting period. While the expense is non-cash, it creates real economic dilution for existing shareholders when the awards vest and new shares are issued. The treatment of SBC in adjusted EBITDA calculations is one of the most debated topics in technology and growth company valuation.
The Case for Adding SBC Back (Non-Cash Treatment)
SBC is non-cash. Like depreciation and amortization, SBC is an accounting charge that does not involve a cash outflow. Adding it back to EBITDA is consistent with the treatment of other non-cash expenses.
The dilution is captured elsewhere. SBC's economic impact (creating new shares) is captured in the diluted share count used to calculate equity value per share. Adding SBC back in the numerator (EBITDA) while using diluted shares in the denominator (per-share calculation) captures the cost through dilution rather than through the income statement.
Comparability across compensation structures. Companies that compensate employees with cash bonuses have lower SBC but higher cash expenses. Companies that use more equity have higher SBC but lower cash compensation. Adding SBC back puts both on a comparable basis by focusing on cash operating performance.
The Case for Keeping SBC as an Expense
SBC is a real cost of running the business. Employees receive equity compensation as a substitute for higher cash salaries. If the company eliminated its equity program, it would need to increase cash compensation to retain talent. Treating SBC as "free" ignores this substitution effect.
Adding back SBC significantly inflates EBITDA. Even at Alphabet, 2025 SBC equalled roughly 18% of EBITDA ($129.0 billion of operating income plus $21.1 billion of depreciation). At high-SBC software companies like Snowflake, SBC exceeds GAAP EBITDA entirely. Adding it back creates an "adjusted EBITDA" figure that bears little resemblance to actual cash generation once share repurchases (which many companies undertake to offset dilution) are considered.
Share repurchases offset the "non-cash" argument. Many companies that add back SBC spend significant cash on share buybacks to offset the dilution from equity issuance. If the company spends $200 million annually on buybacks to counteract $200 million in SBC, the net cash cost is real, not non-cash. At scale, the numbers are large: Alphabet spent $45.4 billion repurchasing its shares in 2025, against $27.1 billion of SBC expense.
The diluted share count only captures past grants. The treasury stock method counts awards that already exist at the valuation date. A company that keeps granting equity will issue more shares every year, and none of that future dilution sits in today's share count. Adding SBC back on the assumption that "dilution is captured elsewhere" therefore misses the cost of every grant still to come.
How much this debate matters for a given company comes down to one ratio.
- SBC Intensity (SBC as % of Revenue)
The ratio of stock-based compensation expense to total revenue, used to compare the relative magnitude of equity-based compensation across companies. SBC intensity varies dramatically: mature technology companies like Microsoft and Apple typically run at 3-5% of revenue, while high-growth SaaS companies can exceed 15-20% of revenue. Companies with SBC intensity above 10% of revenue warrant particularly careful analysis because the difference between SBC-inclusive and SBC-exclusive EBITDA becomes large enough to materially affect valuation conclusions. The metric is also useful for assessing whether a company's SBC levels are sustainable; very high SBC intensity eventually creates unsustainable dilution that the market will penalize.
For a company like Microsoft, whose SBC of $12.0 billion in fiscal 2025 was about 4% of its $281.7 billion of revenue (per its fiscal 2025 10-K), either treatment produces broadly similar multiples. At Snowflake's 34%, the choice decides whether the company looks profitable at all, so the add-back distorts valuation most for equity-heavy growth software.
How Investment Banks Handle SBC in Practice
Most banks present both metrics in their comps tables and models:
- EBITDA (including SBC as an expense): The more conservative measure
- Adjusted EBITDA (excluding SBC): The more commonly used measure in technology coverage
In most technology-focused pitchbooks, the comps table shows both columns side by side: "EV/EBITDA" and "EV/EBITDA (ex-SBC)." This allows the reader to see the impact of the SBC treatment and make their own judgment. The implied valuation range on the football field chart typically uses the metric that is standard for the sector, but the backup materials include both versions.
The choice of which to emphasize depends on the sector and the audience:
| Context | Typical Treatment |
|---|---|
| Technology / SaaS comps | Add back SBC (exclude from EBITDA) |
| Industrial / consumer comps | Include SBC (rarely significant) |
| DCF cash flow projections | Often deduct SBC as a cash expense; if added back, account for future dilution |
| LBO models | Treat SBC (or the cash pay that replaces it) as a real cost |
| Fairness opinions | DCFs frequently treat SBC as a cash expense |
The critical requirement is consistency across the peer group. If you add back SBC for the target, you must add it back for every peer in the comps table. If you keep it in for some and add it back for others, the multiples are not comparable.
Treating SBC as a Cash Expense in DCFs and Fairness Opinions
The DCF is where the add-back has the most direct effect on value, and it is where banker practice has moved furthest toward treating SBC as a cost. Merger proxies regularly state that the financial advisor's discounted cash flow analysis treated SBC as a cash expense. In Permira's 2024 take-private of Squarespace, Centerview's DCF for the special committee used unlevered free cash flow after SBC: the Squarespace definitive merger proxy shows projected 2024 UFCF falling from $331 million to $198 million once SBC was deducted, a cut of about 40%, with SBC forecast at 10-11% of revenue.
The same proxy shows why practice still looks mixed. Centerview's trading comps applied EV to next-twelve-months unlevered free cash flow multiples (peer figures drawn from analyst consensus) to Squarespace's forecast of $353 million, a figure before SBC, while the DCF burdened every projected year with SBC. That difference in SBC treatment between methods is one reason comps and DCF ranges can sit far apart on the same football field.
The practical rule follows from everything above. In comps, match the peer group's convention and apply it to every company; when SBC is material, show both EBITDA figures so the reader can see the gap. In a DCF, never add SBC back without a plan for future dilution, and expect a sophisticated buyer or special committee advisor to treat it as a cash cost. The higher a company's SBC intensity, the more these choices move the answer.
