Operating Expense Modeling: Fixed vs Variable Costs

Key Takeaways

• Operating expense modeling separates COGS, SG&A, and R&D into variable and fixed components so the model captures how margins actually behave as revenue changes.
• A software company growing revenue 20 percent can see operating margins expand from 16 to 22 percent when fixed R&D and G&A stay flat while variable sales and marketing scales at a lower rate.
• Treating every cost line as a percentage of revenue is the most common mistake, it hides operating leverage and produces a falsely smooth margin forecast in both up and down cycles.
• OpEx quality directly affects every portfolio metric that depends on margins: free cash flow, EBITDA, and ultimately the enterprise value a DCF or comparable-company analysis produces.

What It Is

OpEx in a financial model typically splits into COGS (the direct costs of producing the good or service), and operating expenses below the gross profit line: selling, general and administrative (SG&A); research and development (R&D); and sometimes marketing or stock-based compensation as their own lines. Each category gets projected with a method appropriate to its behavior.

COGS is mostly variable, scaling with unit volume. SG&A contains a heavy fixed core plus a variable component tied to revenue. R&D sits closer to fixed in the short run because headcount does not rise and fall with quarterly sales.

The Intuition

Costs that move one-for-one with revenue (like raw materials for a manufacturer) scale linearly. Costs that stay flat as revenue changes (like rent or the CFO's salary) create operating leverage. When revenue rises and fixed costs stay put, profit margins expand. When revenue falls, those same fixed costs now consume a bigger share of a smaller top line and margins compress fast.

Modeling every cost as a percentage of revenue hides this dynamic. You get a smooth margin line that ignores the fact that real businesses see big margin swings in recessions and expansions. A thoughtful OpEx build captures the asymmetry.

How It Works

Three approaches cover most forecasts. Pick one per line item.

Percentage of revenue. Appropriate for variable costs. Project the cost as a stable ratio to sales, often smoothed from two or three historical years.

Forecast cost = Revenue x Historical cost margin

Fixed plus variable. Split the line into a fixed baseline that grows with inflation or headcount, plus a variable component tied to revenue or units.

Forecast cost = Fixed base x (1 + inflation) + Variable rate x Revenue

Bottom-up headcount build. Appropriate for SG&A and R&D in people-heavy businesses. Build headcount by function, multiply by fully loaded cost per employee, add non-headcount items (software licenses, rent, travel).

SG&A = (Headcount x Loaded cost per head) + Other SG&A

Stress-testing a forecast always benefits from back-checking. A bottom-up OpEx projection should reconcile to historical margins. If the new build suddenly shows margins 500 basis points higher than last year with no operational change, something is wrong in the driver assumptions.

Worked Example

A hypothetical software company reported the following year 0 financials.

Revenue           500
COGS              (100)    20 percent of revenue
Gross profit      400      80 percent margin
S&M               (150)    30 percent of revenue
R&D               (120)    fixed at 120
G&A                (50)    fixed at 50
Operating income   80      16 percent margin

Management forecasts 20 percent revenue growth to 600, COGS holding at 20 percent of sales, S&M scaling at 28 percent of sales (marginal efficiency), R&D growing 10 percent with headcount, and G&A flat.

Revenue           600
COGS              (120)    20 percent
S&M               (168)    28 percent
R&D               (132)    +10 percent
G&A                (50)    flat
Operating income  130      21.7 percent margin

Margin expands from 16 percent to 21.7 percent, and the expansion comes from two sources the reviewer can see: S&M efficiency (2 points) and operating leverage on fixed costs (3.7 points). A model that just grew total OpEx at 10 percent would produce a similar bottom line with none of this visibility.

Common Mistakes

1. Treating everything as a percentage of revenue. It is fast to model but wrong for fixed costs. A company whose rent is 8 percent of revenue today will not have rent equal to 8 percent of revenue after five years of growth. The rent stayed flat; the ratio only falls as sales rise.

2. Missing operating leverage on the downside. In a stress case, revenue falls but the fixed cost base does not vanish overnight. A model that scales all OpEx with revenue understates how painful a recession is for margins.

3. Double-counting stock-based compensation. SBC is a non-cash expense that typically sits inside COGS, R&D, and SG&A. Modelers who forecast SBC as a separate line on top of those categories accidentally expense it twice.

4. Forgetting inflation on the fixed base. Salaries, rent, and insurance rise over time even if headcount does not change. A multi-year forecast that holds fixed costs nominally flat for ten years is implicitly assuming zero inflation.

5. Not reconciling bottom-up builds to historical margins. A headcount-driven SG&A build is only as good as the cost-per-employee assumption. Sense-check the output against prior-year margins before locking it in.

Frequently Asked Questions

Q: What is operating expense modeling in simple terms? Operating expense modeling is the process of projecting a company's cost of goods sold and operating expenses by identifying which costs move with revenue and which are fixed, rather than assuming every line grows at the same rate.

Q: How does operating expense modeling affect investment decisions? It reveals operating leverage, how much of each incremental revenue dollar falls to profit. Companies with high fixed-cost bases deliver strong margin expansion in good times and severe compression in downturns, which directly shapes valuation and risk assessment.

Q: What is a real-world example of operating expense modeling? A software company holds R&D headcount flat and caps G&A at a fixed dollar amount while revenue grows 20 percent. The result is a 5-point margin expansion that a model treating all costs as a percentage of revenue would never show.

Q: How can investors use or avoid operating expense modeling errors? Investors should check whether a model breaks costs into fixed and variable components and whether the forecast reflects realistic operating leverage. A margin forecast that stays perfectly constant as revenue grows is almost always wrong.

Q: How is operating expense modeling different from simply projecting an EBITDA margin? Projecting a flat EBITDA margin assumes every cost grows in lock-step with revenue. A proper OpEx build lets fixed costs stay fixed, variable costs scale, and headcount grow by function, producing a margin path that reflects real business economics rather than a circular assumption.

Sources

1. Corporate Finance Institute. "Projecting Income Statement Line Items." https://corporatefinanceinstitute.com/resources/financial-modeling/projecting-income-statement-line-items/
2. Corporate Finance Institute. "Cost of Goods Sold (COGS)." https://corporatefinanceinstitute.com/resources/accounting/cost-of-goods-sold-cogs/
3. AnalystPrep. "Approaches to Forecasting a Company's Operating Expenses and Working Capital." https://analystprep.com/cfa-level-1-exam/equity/approaches-to-forecasting-a-companys-operating-expenses-and-working-capital/
4. Corporate Finance Institute. "Financial Forecasting Guide." https://corporatefinanceinstitute.com/resources/financial-modeling/financial-forecasting-guide/

Disclaimer

This article is educational content only and is not financial advice. Nothing here is a recommendation to buy, sell, or hold any security. Consult a licensed advisor before making investment decisions.