Decremental margin measures how much profit disappears for every unit of revenue that disappears. It is a ratio of changes rather than a level: the fall in operating profit between two comparable periods divided by the fall in revenue across the same interval, expressed as a percentage. A business whose revenue drops by 100 million and whose operating profit drops by 40 million has a decremental margin of 40%.
The measure exists because an average margin says nothing about what happens at the edge of a business. Costs do not move in step with sales. Salaries, leases, data licences and platform engineering continue whether or not clients spend, so the first increment of revenue to leave takes almost pure profit with it. That asymmetry is why a media group can report a single-digit revenue decline alongside a collapse in earnings.
How the calculation works
The arithmetic is one subtraction and one division. Take the current profit figure, subtract the same figure from the prior period, then divide by the equivalent change in revenue. Both changes are negative, so the result reads as the percentage of lost revenue that failed to reach the bottom line.
Everything difficult sits in the choice of inputs. The numerator can be operating profit, earnings before interest and tax, adjusted EBITDA, or a figure stripped of restructuring items. Fortune Brands, one of the few companies to publish a formal definition, uses the change in operating income before charges and gains over the change in net sales, and warns that the result is not derived in accordance with generally accepted accounting principles and may be inconsistent with measures used elsewhere. Dover treats EBIT conversion and decremental margin as the same construction.
The denominator matters as much, and in advertising it is contested. A holding company reports gross billings, net revenue and revenue less pass-through costs. An ad tech vendor reports revenue that may include media bought as principal, alongside contribution excluding traffic acquisition costs, the net figure most of the sector now guides against. One profit decline divided by three revenue bases produces three answers, none of them wrong.
WPP supplies a clean worked example across consecutive halves. First-half 2025 revenue less pass-through costs was 5,026 million pounds against 5,599 million a year earlier, a fall of 573 million, while headline operating profit fell to 412 million pounds from 646 million. That 234 million decline gives a decremental margin near 41%, against an average headline margin of 11.5% in the prior period. PPC Land documented the workforce reduction announced alongside those figures, including 86 million pounds of severance, up from 36 million.
The following year inverted the pattern. First-half 2026 revenue less pass-through costs of 4,745 million pounds was 281 million lower, yet headline operating profit fell only 14 million, to 398 million, and the headline margin edged up to 8.4%. A reading near 5% on a still-shrinking top line is the signature of a cost base that has caught up with demand. Industrial manufacturers are typically modelled at 35% to 55%; businesses whose costs are overwhelmingly people run higher in the first year of a downturn and lower afterwards.
Origin and evolution
The underlying accounting is old. Henry Hess published a graphical analysis relating cost, volume and revenue in 1903, and Walter Rautenstrauch of Columbia University coined the term break-even point in his 1930 book The Successful Control of Profits. Contribution margin and cost-volume-profit analysis grew out of that work and still underpin the calculation.
The vocabulary arrived with a recession. The Wall Street Journal ran a piece headlined "Decremental? Fitting Word for Ugly Times" on March 2, 2009, noting that the adjective had begun appearing across sell-side research as earnings collapsed faster than sales. Deutsche Bank analyst Peter Reilly, quoted in that article, called it "an ugly word" and described it as a fancy way of measuring how far earnings fall with each dollar of sales lost. Investor relations practitioners advised against it.
Companies adopted it anyway, under constraint. Regulation G, published by the United States Securities and Exchange Commission on January 22, 2003 and effective that March, requires any non-GAAP measure in a public disclosure to carry the closest comparable GAAP figure and a quantitative reconciliation. Decremental margin falls inside that definition.
The pandemic normalised it. Omnicom revenue fell 11.9% in 2020, to 13,171.1 million dollars from 14,953.7 million, and the operating margin dropped to 12.1% from 14.2%. Operating profit fell by roughly 523 million dollars against a revenue decline of 1,783 million, near 29%. Excluding the 171.1 million dollars of net repositioning and impairment charges disclosed for the year, the figure falls to about 20%, a gap that shows how heavily one-off items load the ratio.
Why the number matters to marketers
Advertising is a cyclical industry with a high fixed cost base, attached to the most discretionary line in a client budget, which makes it a natural home for the measure even though the sector rarely uses the word.
Agency economics are the clearest case. Staff costs dominate, and a lost account removes revenue immediately while the team stays on payroll. WPP total headline operating costs fell 6.8% in the first half of 2025 against a 10.2% decline in revenue less pass-through costs. The restructuring announced in February 2026 targeted 676 million dollars of annual savings.
Broadcast and audio show the extreme. iHeartMedia's Multiplatform segment reported revenue down 1.6% to 535.7 million dollars while segment adjusted EBITDA fell 39.2%, to 58.6 million from 96.4 million, with operating expenses rising 6.4%. Roughly 9 million dollars of lost revenue against a 37.8 million dollar profit decline puts the ratio above 400%. Any reading over 100% means costs rose while revenue fell, which is a different diagnosis from operating leverage.
Ad tech sits between the two. Perion reported second-quarter 2026 revenue of 98.2 million dollars against 103 million a year earlier, with adjusted EBITDA of 2.8 million against 7.1 million: a 4.8 million dollar revenue decline took 4.3 million of profit with it, close to 90%. Measured on contribution excluding traffic acquisition costs, which fell 11% to 42.3 million, the same profit decline reads as roughly 83%. The comparison quarter had itself been a return to growthafter the Microsoft Bing distribution change of 2024.
Limitations and disputes
No standard definition exists, and the absence is not cosmetic. Two companies with identical results can publish figures 20 points apart by choosing different profit lines and revenue bases.
Denominator sensitivity is the sharpest problem in advertising. WPP first-half 2025 revenue less pass-through costs fell 10.2% as reported but 4.3% like-for-like, stripping currency, acquisitions and disposals. The same profit decline over the smaller base pushes the ratio past 100%. Neither calculation is incorrect, and published figures rarely say which basis was used.
The ratio also breaks when the revenue change approaches zero. Yelp adjusted EBITDA for the half fell 7.9%, to 170.78 million dollars from 185.43 million, with margin at 23% against 25%, implying a revenue line that barely moved. A large profit change over a near-zero revenue change is arithmetically enormous and analytically useless.
One-off items distort in both directions, and in agencies the distortion is structural: severance is booked precisely because revenue fell, so the measure counts the response to the decline as part of the decline. Selective disclosure draws criticism too. Incremental margins appear in growth-year presentations, decrementals mainly when management wants credit for controlling them.
Not the same as
Incremental margin is the identical calculation applied to a period of growth. The two are not symmetrical in practice, because costs added during expansion are harder to remove than they were to add.
Margin compression describes a falling margin level and requires no revenue decline at all. The Trade Desk adjusted EBITDA fell 11% to 241.3 million dollars with the margin narrowing to 34% from 39% in a quarter when revenue grew; the guided 12% third-quarter decline would be the first period where a decremental margin could be computed. AppLovin guided third-quarter adjusted EBITDA margin down to 83% while implying revenue growth above 46%, which is a margin story rather than a decremental one.
Contribution margin is a structural property of unit economics, the price of a unit less its variable cost. The decremental figure is an observation about two reported periods, contaminated by mix, currency and one-off charges.
Agency margin in media buying means something else entirely: the spread an intermediary keeps between what it pays and what it charges. That sense governs take rate, principal media and the contractual clauses determining whether a spread is disclosed, and sits closer to arbitrage than to operating leverage.
Recent developments
Two years of declining fee income have made the calculation routine. Omnicom reported first-half 2026 margins of 15.5%on revenue inflated by the Interpublic combination, while PubMatic guided second-quarter revenue to between 68 and 70 million dollars against 71.1 million, a fourth consecutive quarter of decline. Teads reported first-quarter 2026 revenue down 7% with connected television up 50%, a mix shift that moves profit independently of volume.
Rising costs complicate the picture. Inference and tokenisation expenses now sit inside the agency cost base, and the ISBA finding that publishers receive 36 cents of every media dollar points at an intermediary layer whose own fee compression feeds the same ratios. Sprinklr guided first-quarter fiscal 2027 non-GAAP operating margin to roughly 13% against 17% with professional services revenue expected to fall.
The measure will not stabilise while cost bases are being rebuilt around automation. What it captures in advertising is less a fixed property of each company than a record of how quickly each one cut.
Timeline
- 1903: Henry Hess publishes a graphical analysis relating cost, volume and revenue
- 1930: Walter Rautenstrauch coins the term break-even point in The Successful Control of Profits
- January 22, 2003: The SEC publishes Regulation G and amendments to Item 10 of Regulation S-K governing non-GAAP measures
- March 28, 2003: Regulation G takes effect
- March 2, 2009: The Wall Street Journal reports the spread of the adjective across sell-side research during the financial crisis
- 2020: Omnicom revenue falls 11.9% and operating margin drops to 12.1% from 14.2% during the pandemic
- July 22, 2020: Dover files investor materials defining EBIT conversion and decremental margin as change in adjusted segment earnings over change in revenue
- 2022: Fortune Brands publishes a formal non-GAAP definition based on operating income before charges and gains
- August 7, 2025: WPP reports first-half headline operating profit of 412 million pounds against 646 million, on revenue less pass-through costs down 573 million
- February 26, 2026: WPP reports its weakest annual performance since the pandemic and announces a four-division restructuring
- August 6, 2026: WPP reports first-half headline operating profit of 398 million pounds, a margin of 8.4%, on a further decline in revenue less pass-through costs
Related PPC Land coverage
- WPP cuts workforce - First-half 2025 results showing a 7.8% revenue decline, margin falling to 8.2% from 11.5%, and severance of 86 million pounds.
- Explaining WPP - Covers the first-half 2026 interim results in which headline margin rose to 8.4% despite a further revenue decline.
- WPP collapses, Trade Desk stumbles, and AI rewrites ad industry rules - The February 2026 annual results, the four-division Elevate28 structure and the 676 million dollar savings target.
- iHeartMedia radio profit drops 39% as programmatic revenue targets $200m - A segment where a 1.6% revenue decline accompanied a 39.2% fall in adjusted EBITDA and rising operating expenses.
- Perion EBITDA drops 61% to $2.8m as open web pulls revenue down 5% - Second-quarter 2026 figures used here as the ad tech worked example.
- Perion reports Q2 2025 revenue increases as advertising solutions business returns to growth - The comparison quarter, with revenue of 103 million dollars and adjusted EBITDA of 7.1 million.
- Trade Desk stock drops 24% as Q3 guidance points to 12% revenue decline - Margin compression without a revenue decline, and guidance pointing to one.
- AppLovin revenue growth slows to 53% as Q3 margin guidance drops to 83% - A guided margin step down alongside continued revenue growth.
- Yelp halts share buybacks to pay down $100 million credit facility - Half-year adjusted EBITDA down 7.9% on a revenue line that barely moved.
- Explaining ex-TAC gross profit - The net revenue base most ad tech companies guide against, and the reason denominator choice changes the answer.
- Explaining take rate - The intermediary spread sense of margin, distinct from operating leverage.
- Explaining master services agreement - Contractual disclosure of principal media and undisclosed spreads.
- Explaining arbitrage - Buying low and selling high as a business model rather than a cost structure.
- Agencies turn AI tokens into a margin business as agentic spend stalls - Omnicom first-half 2026 margins and PubMatic's fourth consecutive quarter of declining revenue.
- Teads Q1 2026: CTV surges 50% but revenue falls 7% as integration pain lingers - Mix shift moving profit independently of volume.
- Margin wars reshape programmatic as Amazon undercuts the Trade Desk - Fee compression across the supply chain and the ISBA finding on publisher share.
- Sprinklr closes fiscal 2026 with $857M revenue - but growth is slowing - Guidance implying a margin step down alongside a declining services line.
Summary
Who. Sell-side and buy-side analysts calculate it; chief financial officers at agency holding companies, broadcasters, publishers and ad tech vendors are asked about it on earnings calls. A small number of industrial companies, including Fortune Brands and Dover, publish formal definitions. Advertising companies generally do not, leaving the calculation to readers of their disclosures.
What. The change in operating profit divided by the change in revenue across two comparable periods when revenue has fallen, expressed as a percentage. It quantifies operating leverage in reverse.
When. The underlying cost-volume-profit accounting dates from the first decades of the twentieth century. The term entered mainstream financial vocabulary during the 2009 recession and became common in advertising coverage through the pandemic and the fee declines of 2025 and 2026.
Where. It appears in earnings releases, investor presentations, sell-side models and non-GAAP reconciliations governed by Regulation G in the United States. In advertising it is applied to revenue, net revenue, revenue less pass-through costs and contribution excluding traffic acquisition costs, with materially different results.
Why. Costs in media and marketing services are largely fixed in the short term and largely people in the long term, so revenue declines fall through to profit at multiples of the average margin. The ratio indicates whether a company is cutting fast enough, and a reading above 100% signals that costs are still rising into a falling market.
Discussion