← Quantery Blog

How to read operating margin on a stock screen

October 12, 2026 · 8 min readfundamentalsstock-screening

Operating margin is operating income divided by revenue. It tells you how much of each sales dollar remains after the costs of running the business, but before interest and income taxes. A widening margin can mean better pricing, a cheaper cost base, or a shift toward richer products. The ratio alone can't tell you which one happened.

That makes operating margin useful in a stock thesis when you test both its level and its direction, then open the filing to explain the move. Treat it as a compact reading of the core business. It can't deliver a verdict on the whole company.

What does operating margin measure?

Start with the middle of the income statement:

operating margin = operating income / revenue

Revenue is the top line. Operating income is what remains after the operating expenses recognized in that period. Those expenses usually include the direct cost of delivering the product, sales and marketing, research and development, and administrative overhead. Companies may also put restructuring or asset-impairment charges above the operating-income line.

Interest sits below that line because it follows from how the company is financed. Income tax sits below it too. Gains or losses outside the regular operation can appear below it as well. By stopping before those items, operating margin asks a narrower question than net margin: how much profit did the business report from its operations for each dollar of revenue?

The boundary isn't perfectly uniform. A bank's revenue and interest expense describe the operating business itself. A property company has its own statement logic. Even among ordinary industrial and technology companies, managers make choices about which costs belong in each line. Compare a business with its own history first. Cross-company comparisons work best inside a similar business model.

Operating margin also differs from gross margin. Gross margin stops after the direct cost of the product. Operating margin goes farther and charges the business for the organization around that product. The distinction matters. A software company can keep a rich gross margin while hiring enough engineers and salespeople to drive its operating margin down.

For the full walk from revenue to the bottom line, read revenue, net income, and margins on the income statement. Operating margin lives between the ratios covered there.

Meta shows why the middle of the statement matters

Meta's fiscal 2022 and 2023 filings give a clean historical example because the operating margin moved hard in both directions.

In its 2022 Form 10-K, Meta reported revenue of $116.609 billion, costs and expenses of $87.665 billion, and operating income of $28.944 billion. Its operating margin was 25 percent, down from 40 percent in 2021. Revenue had slipped 1 percent while costs and expenses rose 23 percent. The result came from the middle of the income statement as well as the top line.

The same filing gives the pieces behind that cost increase. Meta recorded $4.610 billion of restructuring charges during 2022, including facilities consolidation, data-center asset charges, and severance. Those charges were included in costs and expenses, so they reduced operating income. They weren't the entire explanation. Employee costs, infrastructure spending, and continued investment in its businesses also rose.

Then the ratio snapped back. Meta's 2023 Form 10-K reports revenue of $134.902 billion and operating income of $46.751 billion. Operating margin reached 35 percent. Revenue rose 16 percent, while costs and expenses rose 1 percent. The filing attributes the improvement mainly to higher revenue and lower restructuring charges, partly offset by higher operating expenses.

A margin rule would have caught the fall and the recovery. It wouldn't have told you that restructuring charges reduced the numerator, operating income, or that the company's two segments were pulling in opposite directions. In 2023, Family of Apps produced $62.871 billion of operating income while Reality Labs recorded a $16.120 billion operating loss. Those two figures reconcile exactly to consolidated operating income of $46.751 billion.

The consolidated ratio was real. It was also an average of one profitable segment and one loss-making segment. Read the segment note before calling that average a business model.

A wider margin needs an explanation

There are several ordinary ways for operating margin to widen.

Revenue can outrun fixed costs. A business doesn't need another headquarters every time sales rise. When the same organization supports more revenue, operating expenses grow slower than the top line. This is operating leverage, and it works backward when sales fall.

Prices or product mix can improve. A company may charge more without a matching increase in delivery cost, or customers may buy more of its richer-margin products. Gross margin often moves too, though the filing is needed to separate price from mix.

Management can cut the cost base. Fewer employees, less advertising, closed facilities, and smaller research programs can lift today's operating margin. Some cuts remove waste. Others take capacity out of the business and hand tomorrow a bill. The ratio doesn't know the difference.

Accounting classifications can move. A restructuring charge can depress one period and disappear from the next. A company can also exclude selected costs from a non-GAAP operating measure. For a market-wide rule, stick to the filed operating-income line unless you've defined and sourced every adjustment yourself.

The Meta example contains more than one mechanism. Revenue recovered. Expenses barely grew, and restructuring charges fell. Calling the resulting margin expansion "pricing power" would outrun the evidence. The filing supports a broader claim: more revenue passed through a cost base that grew much slower.

One period can also flatter the trend. Compare the same fiscal quarter a year earlier when seasonality is strong, or use trailing figures to smooth a lumpy charge. Trailing versus forward numbers explains why the filing dates and comparison windows need to match.

How should a thesis test operating margin?

A useful rule separates three questions.

First, is the company profitable at the operating line? Second, is its margin holding up against its own prior period? Third, did revenue hold up too? That last check keeps a cost-cutting program from looking like healthy expansion while the customer base shrinks underneath it.

The shape below is illustrative, and the bundled templates are the reference for exact fields. The Quantery DSL documentation covers the available functions.

# "Operating margin with sales support" (illustrative)
params:
  min_history: 5
  margin_strong: 0.15
  margin_ok: 0.08
  growth_floor: 0.00

quality:
  flags:
    - { when: "history_periods < $min_history", flag: "short_history:{history_periods}" }

features:
  revenue_ttm: ttm(revenue)
  operating_income_ttm: ttm(operating_income)
  operating_margin: if(revenue_ttm > 0,
    operating_income_ttm / revenue_ttm, null)
  revenue_now: newest(revenue)
  revenue_year_ago: lag(revenue, 4)
  revenue_growth: if(history_periods >= $min_history and revenue_year_ago > 0,
    revenue_now / revenue_year_ago - 1, null)
  margin_now: if(newest(revenue) > 0,
    newest(operating_income) / newest(revenue), null)
  margin_year_ago: if(history_periods >= $min_history and lag(revenue, 4) > 0,
    lag(operating_income, 4) / lag(revenue, 4), null)

criteria:
  margin_level:
    rules:
      - { when: "is_null(operating_margin)", score: 0, flag: no_margin }
      - { when: "operating_margin >= $margin_strong", score: 2 }
      - { when: "operating_margin >= $margin_ok", score: 1 }
      - { else: 0 }
  margin_direction:
    rules:
      - { when: "is_null(margin_year_ago)", score: 0, flag: no_comparison }
      - { when: "margin_now > margin_year_ago", score: 2 }
      - { when: "margin_now == margin_year_ago", score: 1 }
      - { else: 0 }
  sales_support:
    rules:
      - { when: "is_null(revenue_growth)", score: 0, flag: no_sales_comparison }
      - { when: "revenue_growth > $growth_floor", score: 2 }
      - { else: 0 }
  operating_profit:
    rules:
      - { when: "operating_income_ttm > 0", score: 1 }
      - { else: 0 }

gate:
  mode: strict

The history guard keeps fewer than five observed filings from earning a comparison score. It can't prove that those filings are contiguous, because lag() is positional. A filing gap can still make four observed periods land on the wrong fiscal quarter, so check the sequence before treating either comparison as year over year.

The round thresholds are proposed research parameters. They don't establish what a good business must earn. An 8 percent floor will exclude normal low-margin industries. A 15 percent bar makes that exclusion stronger. If you compare companies across the whole market, you may end up ranking industry economics more than management quality.

Run the thesis within comparable industries, or repeat the backtest with several sensible floors. If the result vanishes when margin_strong moves a little, the threshold did the work. How to build a thesis worth testing covers that proxy and parameter discipline in more detail.

What can an operating-margin backtest establish?

The testable claim isn't "wide margins are good." Make it specific: companies with positive revenue growth and operating margins that exceed both a floor and their year-ago level subsequently behaved differently from an appropriate benchmark.

A walk-forward backtest should use only filings available at each decision date. Point-in-time SEC fundamentals support Quantery backtests from 2013, while price depth depends on the user's market-data plan. The results are hypothetical and exclude costs. More important here, a backtest grades the numerical rule. It doesn't prove that pricing power caused the outcome.

Break the result apart. Compare a margin-level rule with a margin-improvement rule. Remove the revenue-growth condition and see what changes. Split by industry. Inspect the companies with the largest one-period jumps, because asset sales and restructuring reversals like to live at the extremes. What makes a backtest honest sets out the data and timing checks behind that exercise.

Operating margin earns its place because it compresses the operating cost structure into one ratio you can compare through time. The compression is also its limit. Let the rule find the changes. Let the filing tell you what changed.

Want to try this on your own rules? Quantery is free for 14 days: the full app, no card required.

← All articles