--- title: "Marketing & Margin Glossary: SMB Terms Defined | MarginFix" description: "Plain English definitions of blended ROAS, MER, incrementality, attribution inflation and every other audit term, each with a formula and a worked example." url: https://marginfix.ai/glossary/ lastmod: 2026-10-13 --- Your number See this page in your own figures: industry, spend and the estimated leak. [Run the estimator →](/estimator/) [Home](/) › Glossary Reference · 15 terms # Marketing and margin glossary The terms behind every MarginFix audit, in plain English. The ones that decide whether your reporting is telling you the truth about your margin. - Reviewed Oct 2026 - 6 min read - 15 terms, each with a formula and a worked example - Median finding $58k a year across 74 audits Terms 1. [Blended ROAS](#blended-roas) 2. [MER](#mer) 3. [CAC](#cac) 4. [Contribution margin](#contribution-margin) 5. [Break even ROAS](#break-even-roas) 6. [Incrementality](#incrementality) 7. [Non incremental spend](#non-incremental-spend) 8. [Attribution inflation](#attribution-inflation) 9. [Margin leak](#margin-leak) 10. [Dunning](#dunning) 11. [Take rate](#take-rate) 12. [Payback period](#payback-period) 13. [LTV:CAC](#ltv-cac) 14. [Wasted ad spend](#wasted-ad-spend) 15. [Marketing and margin audit](#marketing-and-margin-audit) 16. [Against the corpus](#corpus) 01 ## Blended ROAS Return on ad spend measured across every channel at once. It reads healthy while hiding channel level waste, and it over counts, because Meta, Google and GA4 each claim the same sale. Blended ROAS = **total revenue** ÷ **total ad spend** Worked example *Illustration, not a client figure* $420,000 of revenue on $140,000 of ad spend reads 3.0×. An incrementality holdout can put one channel's true return at 1.1×, which is the leak the blend was hiding. Your own blended ROAS Revenue ($) Ad spend ($) Enter both figures to read your own ratio. It is your number, not an audited one. See it in a case: [DTC / Ecommerce, $53k a year of wasted spend cut in 7 days](/cases/the-agency-s-winning-campaign-was-the-biggest-loser/) [Read the full Blended ROAS page →](/glossary/blended-roas/) 02 ## MER (Marketing Efficiency Ratio) Total revenue divided by total marketing spend, a blended, platform independent efficiency measure. Harder to game than per platform ROAS because no single channel can take credit twice. MER = **total revenue** ÷ **total marketing spend** Worked example *Illustration, not a client figure* $1,200,000 of revenue on $300,000 of total marketing spend is an MER of 4.0. If the platforms together report $1,860,000 of attributed revenue against that same spend, they are counting sales more than once. Your own MER Revenue ($) Total marketing spend ($) Enter both figures to read your own ratio. It is your number, not an audited one. See it in a case: [B2B SaaS, $74k a year of wasted spend cut in 7 days](/cases/a-1-2m-ad-budget-hid-a-40-wasted-impression-rate/) [Read the full MER page →](/glossary/mer/) 03 ## CAC (Customer Acquisition Cost) The fully loaded cost to win one customer: acquisition spend divided by new customers. Understated whenever it leaves out creative, tools, onboarding and team time, which is most of the time. CAC = **total acquisition spend** ÷ **new customers** Worked example *Illustration, not a client figure* $90,000 of media for 300 new customers reads $300. Add the $30,000 of creative, tools and team time the deck left out and the true CAC is $400. See it in a case: [Marketplace, +29% blended margin in 7 days](/cases/cac-looked-fine-until-we-split-it-by-cohort/) [Read the full CAC page →](/glossary/cac/) 04 ## Contribution margin Revenue minus every variable cost per order or unit: cost of goods, shipping, payment fees, returns. It is the real profit a marketing dollar defends, and the number blended ROAS quietly ignores. Contribution margin = **revenue** − **variable costs** Worked example *Illustration, not a client figure* A $100 order with $42 of goods, $9 of shipping, $3 of payment fees and $6 of returns allowance contributes $40. A 2.5× ROAS on that order spent $40 to win it, so the sale earned nothing. See it in a case: [DTC / Ecommerce, $59k a year found in 6 days](/cases/bestsellers-were-priced-below-their-true-landed-cost/) [Read the full Contribution margin page →](/glossary/contribution-margin/) 05 ## Break even ROAS Break even ROAS is 1 divided by your contribution margin rate: the return your ads need before they add a dollar of profit. Break even ROAS = 1 ÷ **contribution margin rate** Worked example *Illustration, not a client figure* If an order of $100 leaves $40 after its variable costs, every $1 of ads must bring back $2.50 of revenue before the order adds any profit: a break even ROAS of 2.5×. See it in a case: [DTC / Ecommerce, $134k a year found in 6 days](/cases/scaling-spend-was-hiding-a-134k-annual-margin-leak/) [Read the full Break even ROAS page →](/glossary/break-even-roas/) 06 ## Incrementality The share of conversions that happened because of the spend, against those that would have occurred anyway. High reported ROAS with low incrementality means you are paying for sales you already had. Incremental share = (**exposed conversions** − **holdout conversions**) ÷ **exposed conversions** Worked example *Illustration, not a client figure* A geo holdout: exposed regions convert 1,000 times, matched holdout regions 850. Only 150 of the 1,000 conversions, 15%, were incremental. The platform reported all 1,000. See it in a case: [Retail, $88k a year of wasted spend cut in 7 days](/cases/online-ads-were-paying-for-in-store-demand/) [Read the full Incrementality page →](/glossary/incrementality/) 07 ## Non incremental spend Ad budget that buys conversions you would have won without it, classically branded search against customers already intending to purchase. The most common leak the audits surface. Non incremental spend = **spend** × (1 − **incremental share**) Worked example *Illustration, not a client figure* $40,000 a month of branded search with a 20% incremental share means $32,000 a month bought demand that already existed. Non incremental paid spend appeared in 68% of the 74 audits, the most common finding in the corpus. See it in a case: [Pharma, $78k a year of wasted spend cut in 7 days](/cases/unbranded-search-was-paying-for-traffic-the-brand-site-alrea/) [Read the full Non incremental spend page →](/glossary/non-incremental-spend/) 08 ## Attribution inflation When several platforms each claim the same sale, so attributed revenue adds up to more than 100% of actual revenue. It over funds whichever channel claims most aggressively. Attribution inflation = **sum of platform attributed revenue** ÷ **actual revenue** Worked example *Illustration, not a client figure* Meta claims $300,000, Google $250,000 and email $100,000 against $500,000 of actual revenue: 130%. $150,000 of the attributed revenue never existed. Attribution inflation appeared in 54% of the 74 audits. See it in a case: [DTC / Ecommerce, $31k a year of wasted spend cut in 5 days](/cases/retargeting-was-paid-twice-for-the-same-customer/) [Read the full Attribution inflation page →](/glossary/attribution-inflation/) 09 ## Margin leak A recurring, usually invisible loss of marketing margin, whether wasted paid spend, mispriced offers or retention drop off, that a healthy looking blended metric conceals. Margin leak = **recurring annual loss** an audited fix would stop Worked example *Illustration, not a client figure* A $3,000,000 revenue business leaking 18% of a $600,000 marketing budget is losing $108,000 a year while every blended report reads green. Across the 74 audits a median 18.7% of marketing spend was leaking, inside an 11% to 28% band. See it in a case: [DTC / Ecommerce, $134k a year found in 6 days](/cases/scaling-spend-was-hiding-a-134k-annual-margin-leak/) [Read the full Margin leak page →](/glossary/margin-leak/) 10 ## Dunning The process of retrying failed recurring payments. Weak dunning turns solvable payment failures into involuntary churn, quietly erasing retention gains. Involuntary churn = **failed payments not recovered** ÷ **active subscribers** Worked example *Illustration, not a client figure* 2,000 subscribers, 120 failed payments a month, 40 recovered: 80 lost, a 4% monthly involuntary churn before anyone chose to leave. See it in a case: [Subscription, $52k a year found in 5 days](/cases/weak-payment-recovery-was-leaking-revenue-as-involuntary-chu/) [Read the full Dunning page →](/glossary/dunning/) 11 ## Take rate The percentage a marketplace keeps from each transaction. A take rate set without passing landed cost through to pricing is a structural margin leak. Take rate = **marketplace revenue** ÷ **gross merchandise value** Worked example *Illustration, not a client figure* $150,000 of fees on $1,000,000 of GMV is a 15% take rate. Refund the buyer in full while keeping the fee only on completed orders and a 12% refund rate turns 15% into 13.2%. See it in a case: [Marketplace, $44k a year found in 7 days](/cases/take-rate-was-leaking-through-unmanaged-refunds/) [Read the full Take rate page →](/glossary/take-rate/) 12 ## Payback period The time for a customer's contribution margin to repay their acquisition cost. Almost always longer than the board believes, because CAC excludes onboarding and blended metrics flatter it. Payback period = **CAC** ÷ **monthly contribution margin per customer** Worked example *Illustration, not a client figure* A $400 CAC against $25 of monthly contribution margin pays back in 16 months. The deck that used $50 of monthly revenue instead of margin said 8. See it in a case: [Fintech, +19% CAC payback accuracy in 6 days](/cases/cac-payback-was-twice-what-the-board-believed/) [Read the full Payback period page →](/glossary/payback-period/) 13 ## LTV:CAC Lifetime value to customer acquisition cost, a headline health metric that misleads when LTV is built on revenue instead of contribution margin. LTV:CAC = **lifetime contribution margin per customer** ÷ **CAC** Worked example *Illustration, not a client figure* $1,800 of lifetime revenue at a 40% contribution margin is $720 of lifetime margin. Against a $400 CAC that is 1.8 to 1, not the 4.5 to 1 the revenue version shows. See it in a case: [Subscription, +22% LTV:CAC in 5 days](/cases/growth-spend-was-outrunning-retention/) [Read the full LTV:CAC page →](/glossary/ltv-cac/) 14 ## Wasted ad spend Media budget producing no incremental margin. Quantifying it, not just naming it, is the core output of a marketing audit. Wasted ad spend = **non incremental spend** + **spend that converted nothing** Worked example *Illustration, not a client figure* Four locations bidding on the same keywords against each other raise the price of every click. The spend converts, so no dashboard flags it, and the margin it should have produced never arrives. 22% of audited SMB marketing spend was wasted paid spend; only 62% was effective. See it in a case: [Services, $39k a year of wasted spend cut in 5 days](/cases/every-location-bought-the-same-keywords-against-each-other/) [Read the full Wasted ad spend page →](/glossary/wasted-ad-spend/) 15 ## Marketing & margin audit An independent, evidenced review of marketing spend and unit economics that finds where budget leaks, quantifies the margin at stake and hands over a prioritized fix list. Audit output = **where it leaks** + **the annual figure** + **the fix list, in priority order** Worked example *Illustration, not a client figure* A fixed fee, findings in 5 to 7 working days, one auditor with read only access, and a guarantee floor of findings per tier or the fee comes back. 74 audits across 9 industries; the median highlight finding is worth $58k a year. See it in a case: [The benchmark](/benchmark/) [Read the full Marketing and margin audit page →](/glossary/marketing-and-margin-audit/) ## The terms, against the corpus What 74 audited SMB marketing budgets looked like once every term above was measured properly. Only **62%** of audited SMB marketing spend was effective: 22% was wasted paid spend, 9% mispriced offers and 7% leaky retention. The median highlight finding is worth $58k a year. [74 cases](/findings/) [The benchmark](/benchmark/) Cite this figure Only 62% of audited SMB marketing spend was effective: 22% was wasted paid spend, 9% mispriced offers and 7% leaky retention (MarginFix, 74 marketing margin audits). MarginFix, marginfix.ai/glossary/ ![Portrait photograph of Cristian Bragau](/assets/founder/cristian-bragau-2026-10.jpg) The auditor's note Every term here decides whether a report is telling you the truth about margin. Blended ROAS and platform attribution are where the audits most often find the gap: the top line reads healthy while one channel leaks. Learn the terms, then check your own numbers against them. **[Cristian Bragau](/about/cristian-bragau/)** Founder & Lead Auditor · 17 published cases · [LinkedIn](https://www.linkedin.com/in/cristian-bragau-828b8526/) ## Know the terms. Now find where they leak in your own numbers. Vocabulary is not the audit. Every term above, measured against your own numbers, in 5 to 7 working days. $10k to $50k of findings, or your money back. [Book the audit](/book/) [See the benchmark report](/benchmark/) Fixed fee. No retainer. NDA first. Go deeper ## Read on from the glossary - [All 74 findings Ranked by annual value, $58k median, each one a real case.](/findings/) - [The benchmark Where 74 SMB marketing budgets leaked, by share of spend and by industry.](/benchmark/) - [Margin leaks The five leak patterns in depth: what each one is and how an audit finds it.](/margin-leaks/) - [Marketing audit checklist Thirty checks across paid media, attribution, pricing, retention and reporting.](/marketing-audit-checklist/) - [Insights Where SMB budgets leak, the ROAS field guide and how to audit your agency.](/insights/) - [What an audit costs The 2026 market ranges with sources, against the fixed fee tiers.](/marketing-audit-cost/)