When does each channel break even on ad spend?

MetricPlatform-reported return on ad spend by acquisition channel

Unitx revenue per $1 of channel ad spend

Cumulative revenue divided by cumulative spend per channel over time. Different channels have different shapes (search ramps slow, social ramps fast then plateaus). Build by tracking spend and attributed revenue per channel per day. Example: Meta 1.4x at day 30, 3.2x at day 365; Google search 0.7x at day 30, 3.4x at day 365.

ROAS

Return on ad spend. Revenue attributed to a campaign divided by the spend on that campaign.

Ecommerce and DTC advertisers, full-year 2025 platform-reported ROAS, Triple Whale panel of roughly 35,000 brands for Meta and TikTok and roughly 18,000 for Google. This is platform-attributed revenue divided by channel spend, not incremental revenue and not contribution. · This is channel ROAS for a DTC brand and must not be read against the rep or partner ROAS families elsewhere in the atlas. Three definitional warnings. First, platform-reported ROAS over-attributes relative to incrementality testing, with reported over-attribution of roughly 20% to 50%, and Meta removed its 7-day and 28-day view-through windows in January 2026, which moves reported numbers without moving performance. Second, cross-channel comparison is not a quality ranking: Google captures existing demand while Meta and TikTok create it, which structurally explains most of the Google-to-Meta gap. Third, the only benchmark that decides profitability is break-even ROAS, which is 1 divided by contribution margin, so a 1.86x median is profitable above roughly 54% margin and loss-making below it. Varos data reports a higher Meta all-industry median of 2.19x, so treat the Meta median as a 1.86x to 2.19x band.

Every channel curve flattens below the break-even line drawn on the chart, or one channel's curve sits high only because the retargeting line has been blended into prospecting, so the curve reflects harvesting demand the brand already had.

Blended Meta ROAS reads 2.4x against a stated 1.86x median and looks fine, but the split shows prospecting at 1.3x and retargeting at 4.1x, while break-even at a 42% contribution margin is 2.38x. The prospecting engine has been losing money on every incremental dollar for two quarters.

• Draw your break-even ROAS line on the chart before reading any benchmark. Break-even is 1 divided by contribution margin, so 1.67x at 60% margin and 4.0x at 25% margin. The same 1.86x median is a pass or a fail depending only on your own CM.
• Split prospecting from retargeting on every channel curve. Retargeting runs about 71% higher ROAS than prospecting, so a blended account average hides which engine is actually acquiring customers.
• Split Google into Search and Performance Max before reallocating. Search averages 5.17:1 and PMax 2.57:1 in the same 18,000-brand dataset, so a channel-level Google number is an average of two different businesses.
• For TikTok in long-consideration or offline-heavy categories, judge on blended MER rather than platform ROAS. Reported TikTok ROAS of 0.08 in pets and 0.49 in food and beverage largely reflects pixel undercounting, and cutting budget on the platform-reported figure alone will cut a working channel.

1. Triple WhaleFacebook (Meta) Ad Benchmarks by Industry · 2026 · n=~35,000 ecommerce brands, 1 January to 31 December 2025 Meta brands invested 68.31% of total ad budget on the platform in 2025; efficiency varies significantly by vertical while costs rose across the board triplewhale.com ↗
2. Triple WhaleTikTok Ads Benchmarks by Industry · 2026 · n=~35,000 ecommerce brands using Triple Whale, full-year 2025 TikTok CPA rose 8.64% to $32.74, CVR fell 6.20% to 2.01% and ROAS declined 5.70% to 2.21, with CTR up 13.74% triplewhale.com ↗
3. Triple Whale2025 Ecommerce Benchmarks: DTC Data Trends Report · 2026 · Triple Whale brand panel, full-year 2025 Meta grew to 68.3% of ad spend; Amazon had 11.02% CVR and the lowest CPA of any major platform; TikTok ROAS varied by 234% between industries triplewhale.com ↗
4. Influee (citing Triple Whale)Facebook Ads Benchmarks 2026 · 2026 · Triple Whale analysis of roughly 35,000 ad accounts, January to December 2025 Median Meta ROAS 1.86 with CPM up 20.03%, CTR up 13.5%, CVR up 8.29% and CPA up only 1.04%; for most DTC categories 1.5 to 2.5 is normal influee.co ↗
5. Influee (citing Varos and Triple Whale)What Is a Good ROAS for Facebook Ads? 2026 Benchmarks by Industry · 2026 · Varos benchmarks through April 2025 cross-referenced with Triple Whale All-industry Meta median 2.19 with a spread from automotive parts 6.76 to healthcare 1.19; DTC verticals cluster below 2.0 (beauty 1.57, personal care 1.48, wellness 1.45, food and beverage 1.69) influee.co ↗
6. Superscale (citing Triple Whale, Varos, Focus Digital)ROAS Benchmarks by Industry 2026: Google and Meta Ads Data · 2026 · Triple Whale analysis of 18,000+ ecommerce brands for Google; Focus Digital 5,000+ Google Ads accounts March 2024 to April 2025 Google Ads median 3.68x (down 10.03% year on year) against Meta 1.86x to 2.19x; realistic Google planning range 3.0x to 3.7x superscale.ai ↗
7. Jetfuel Agency (citing Triple Whale)Google Ads for Ecommerce: 2026 Scaling Guide · 2026 · Triple Whale 2025 benchmark report across 18,000 ecommerce brands Within Google, Search campaigns average 5.17:1 ROAS while Performance Max averages 2.57:1, a spread wider than most cross-platform gaps jetfuel.agency ↗
8. Orange MonkE (citing Triple Whale)Facebook Ads Benchmarks 2026 · 2026 · Triple Whale, 35,000 brands, 2025 Platform-reported ROAS over-attributes by roughly 20% to 50% versus incrementality testing, and Meta removed 7-day and 28-day view-through windows on 12 January 2026; retargeting runs about 71% higher ROAS than prospecting orangemonke.com ↗
9. DTCROAS2026 eCommerce Ad Platform Benchmarks · 2026 · Synthesis of Triple Whale ecommerce benchmarks by platform Amazon Ads achieved 3.14x ROAS in 2025 with CPA of $13.35, against Meta 1.86x median dtcroas.com ↗