MER (Marketing Efficiency Ratio)

MER (Marketing Efficiency Ratio), also called Media Efficiency Ratio, blended ROAS, total ROAS or eROAS, measures total revenue ÷ total marketing spend over the same window — a single, attribution-independent read on the efficiency of the entire marketing engine (Triple Whale, via search). It is the cross-channel sibling the run-103→108 measurement cluster (Retail Media → Incrementality → Media Mix Modeling (MMM) → Multi-Touch Attribution (MTA) → ROAS → POAS (Profit on Ad Spend)) repeatedly names as the correction for per-channel ROAS's double-counting problem. Where ROAS divides one channel's attributed revenue by that channel's spend, MER drops the attribution layer entirely.

Firewall: every claim below is what a source reports. See ../../CONTEXT.md Rule 1. This was a web-only run — most fetched sources are tool vendors or CFO/agency firms with a commercial interest (Superscale, Triple Whale, Northbeam, Measured all sell measurement or bidding products; Eightx and Upcounting are ecommerce-finance firms). The Reddit (reddit-research MCP not connected) and YouTube (daily cap reached, 19/2) practitioner streams were both down. No independent/academic source on MER surfaced. Treat all benchmarks and target bands as self-reported.

Definition & formula

The canonical formula is MER = Total Revenue ÷ Total Marketing Spend over the same window; a MER of 4 means $4 of revenue for every $1 of marketing spend (Eightx, 2026-06-01, finance firm). Eightx's key framing is that MER is calculated off Shopify backend revenue (truth) and total platform spend (truth), so "it cannot be inflated by overcounting on one channel or deflated by undercounting on another" — the property that makes it survive a broken attribution layer (Eightx, 2026-06-01).

Eightx reports that "marketing spend" should include every marketing dollar — paid media (Meta, Google, TikTok, Amazon DSP), retention tools (Klaviyo, Postscript), affiliate/influencer payouts, and agency fees — and that excluding agency retainers and tools inflates the number and breaks benchmark comparisons (Eightx, 2026-06-01).

MER vs blended ROAS vs platform ROAS

Eightx (corroborated by EasyInsights, via search) reports MER is the same calculation as blended ROAS / eROAS, the difference being emphasis — "blended ROAS" frames it against Meta's/Google's reported ROAS, "MER" frames it as the efficiency of the entire marketing engine (Eightx, 2026-06-01).

Superscale draws a sharper line by denominator: Blended ROAS = total revenue ÷ total paid-media spend; MER = total revenue ÷ total marketing spend (incl. agency, tools, creators). It claims the two diverge by 10–30% depending on how creator-heavy the mix is (as-of 2026-06-16), calling blended ROAS "the daily dashboard number" and MER "the weekly P&L view" (Superscale, 2026-06-16, vendor).

Platform ROAS divides one channel's attributed revenue by that channel's spend and depends heavily on attribution, so it is distorted by iOS/pixel gaps and double-counts across channels; MER drops the attribution layer entirely, so it is unaffected by those gaps and cannot double-count (Superscale, 2026-06-16, vendor).

[!unverified] Eightx reports a frequent source of operator confusion: Triple Whale inverts the ratio in-product, surfacing "Blended Ad Spend ÷ Order Revenue" (a cost % of revenue, where lower is better) rather than the canonical Revenue÷Spend multiple, while Polar Analytics and Northbeam use the canonical formulation (Eightx, 2026-06-01).

Why brands adopted MER (privacy / iOS ATT / double-counting)

Eightx and Superscale both attribute MER's rise to the post-2021 attribution breakdown. Apple App Tracking Transparency (ATT) (April 2021) stripped Meta, TikTok and others of deterministic conversion data; Eightx states "platform ROAS became a number a CFO could not defend in a board pack" (Eightx, 2026-06-01).

[!unverified] Superscale reports that after iOS 14.5, Facebook under-counts conversions typically by 15–30%, reaching 50–70% for many advertisers, while summed platform ROAS simultaneously over-attributes — platforms collectively claim 30–40% more revenue than the store actually recorded (the "double-counting tax") (as-of 2026-06-16; citing Adligator) (Superscale, 2026-06-16, vendor).

MER vs POAS vs ROAS — where each fits

Practitioner positioning across sources: MER is the strategic, business-level number for multi-channel brands (ties to P&L, attribution-independent); ROAS is the tactical kill-and-scale signal inside one ad account; and POAS (Profit on Ad Spend) fixes ROAS's margin blind spot at campaign level. Superscale frames MER as the right primary metric for brands above ~$5M ARR; JudeLuxe's taxonomy is "use all three, each for a different job" (Superscale, 2026-06-16; JudeLuxe, via search). The critical distinction from POAS: MER measures revenue efficiency (gross, pre-returns), not profit — see POAS (Profit on Ad Spend) and Contribution Margin.

Break-even MER

Eightx reports break-even MER = 1 ÷ contribution-margin % — at 30% contribution margin break-even MER is 3.3; at 40% it is 2.5. Below break-even, every incremental ad dollar loses money on a first-order basis (Eightx, 2026-06-01). This mirrors break-even ROAS = 1 ÷ gross margin, but uses the fuller contribution-margin denominator.

Benchmarks / target bands (all volatile — as-of 2026-06-27)

[!unverified] All MER target bands below are vendor/finance-firm figures, several self-described as "synthesis, not a published benchmark." Directional only.

General DTC: 2026 MER benchmarks sit around 3×–5× for most DTC verticals, with mature subscription brands pushing above 6× (Eightx, 2026-06-01).
Stage-dependent (Eightx): $1M–$5M brands run 1.5–2.5 (often lose money on first order); $5M–$10M run 2.5–3.5; $10M–$25M run 3.0–4.5; $25M–$100M run 3.5–6.0+; subscription/high-LTV brands intentionally run lower (1.5–2.5) and defend it via LTV (Eightx, 2026-06-01).
The "MER > 4" rule: Superscale reports Common Thread Collective's Taylor Holiday popularised the framing that MER above 4 keeps CAC under 25% of revenue (the line most DTC P&Ls need to clear), pushing for 5–8 at scale (Superscale, 2026-06-16, citing commonthreadco.com).

Vertical MER bands (Superscale — self-described synthesis, low confidence, as-of 2026-06-16)

Vertical (gross margin)	Early-stage MER	Mature MER
Apparel / accessories (55–70% GM)	3.5–4.5	4.5–6.0
Beauty / skincare	3.5–4.5	4.5–6.5
Supplements	4.0–5.0	5.0–7.0
Food & beverage	5.0–6.5	6.5–8.5
Home / furniture	4.0–5.0	5.0–7.0
Consumer electronics	6.0–8.0	7.5–10.0

[!unverified] Triple Whale's in-product framing: the 2025 median MER for Triple Whale customers was 41% expressed as its inverted cost-ratio (≈2.4× in canonical terms), where lower is better (Triple Whale, via search).

New-customer MER — nMER / aMER

Eightx reports the most common MER mistake is making acquisition decisions off blended MER, which includes retention revenue (email/SMS) and can mask a broken acquisition engine. The diagnostic fix:

nMER = new-customer revenue ÷ acquisition-only ad spend — "the honest read on whether the brand is buying customers profitably today" (Eightx, 2026-06-01). AdSights maintains a glossary entry defining nMER as a distinct metric (AdSights, via search).
aMER (acquisition MER) = new-customer revenue ÷ acquisition spend; set as a target via target aMER = new-customer AOV ÷ target CAC (e.g. $112 AOV ÷ $50 target CAC = 2.24) (Upcounting, 2026-era, finance firm).

Upcounting stresses "economics first, aMER second": aMER is not a universal benchmark — it shifts with AOV and CAC target, and the same 2.24 can be healthy for a high-margin, high-repeat brand and cash-burning for a thin-margin, low-repeat one. It frames Customer Acquisition Cost (CAC) and aMER as the same constraint in two formats (CAC in dollars, aMER as the revenue multiple), and warns a higher aMER target is not automatically better, since chasing efficiency can cut total spend and total contribution margin (Upcounting, 2026-era).

Criticism / limitations

Measured (which sells incrementality/MMM and is openly anti-MER) argues MER is "simply the marketing/ad spend to total sales percentage by another name," a premise rather than a conclusion, because it assumes every ad dollar was effective and correlated to sales. It states a high MER is "more likely attributed to a tremendous baseline of organic or earned sales that paid is gobbling up credit for," and concludes MER "is neither an actionable data point nor should it be used as a conclusion to guide behavior" (Measured, 2023-09-05, mod 2024-08-14, vendor).

Other reported limits:

MER does not isolate Incrementality — it credits all spend with all sales and cannot distinguish demand creation from demand capture; the proposed fix is to divide spend by only the sales that would not have occurred without the ads (Measured; corroborated Prescient AI, via search).
No channel/campaign granularity — a strong blended MER can hide inefficiency in specific performance efforts propped up by effective brand-building (The Good, via search).
Revenue, not profit — typically calculated on gross revenue before returns (Diversification.com, via search).
Lagging / window-sensitive — measuring over too short a window ignores natural payback lags; if halo/incrementality/attribution-inflation effects exist, ROAS can stay stable while MER weakens (Keends, via search).

Even MER's advocates concede the hard limit: Superscale states "MER is a correlation metric, not a causal one" — above ~$10M in spend you need incrementality experiments (geo holdouts, MMM, conversion lift) to know what MER would do if a channel were switched off: "MER tells you the system is efficient. It does not tell you which part earned it" (Superscale, 2026-06-16, vendor).

Contradictions

Key terms

Term	Meaning
MER	Total revenue ÷ total marketing spend (attribution-independent; aka Media Efficiency Ratio)
Blended ROAS / eROAS	Same idea; Superscale narrows the denominator to paid-media spend (vs MER's all-marketing spend)
nMER	New-customer revenue ÷ acquisition-only spend (acquisition diagnostic)
aMER	Acquisition MER; target aMER = new-customer AOV ÷ target CAC
Break-even MER	1 ÷ contribution-margin %
Inverted MER	Triple Whale's in-product cost-ratio (spend ÷ revenue; lower is better)

What practitioners report

Not gathered this run — the Reddit (reddit-research MCP not connected; agent returned tool_uses 0, honest empty report) and YouTube (daily cap reached 19/2) streams were both down, so there is no practitioner counter-narrative on real-world target MER values, how operators actually split blended vs new-customer MER, or whether MER earns its keep vs incrementality testing.