Foundations & core concepts: CAC, LTV & ROAS

Ask four marketing leaders what their CACCACCustomer Acquisition Cost (CAC) is the total sales and marketing spend divided by the number of new customers gained in a period. It measures how efficiently you grow.View full definition → is and you will get four numbers built from four different denominators. One counts media spend only. One adds the agency retainer and the creative studio. One divides by every new customer, including the ones who arrived through organic search and never saw an ad. One divides by gross signups and forgets the people who cancelled in week two. The arithmetic in all four cases is a division. The money is lost in the definitions. This lesson fixes them: what belongs in the numerator of CAC, whose lifetime LTV actually measures, and what ROAS counts and refuses to count. Everything later in this module (cohort curves, payback windows, incrementalityincrementalityThe share of results (sales, conversions, revenue) that only happened because of a marketing action, not what would have occurred anyway.View full definition → testing, board-level arbitration) assumes these three objects are unambiguous. They usually are not.

Core concepts

CAC, Customer Acquisition Cost, is everything you spend to turn a stranger into a first-time paying customer, divided by the number of people who became one in the same period.

The numerator, done properly, holds paid mediapaid mediaVisitors arriving via paid ads or sponsored placements, where you pay a platform to display your message rather than earning visits organically.View full definition →, creative production, agency fees, martechmartechThe connected set of software tools a marketing team uses to plan, run, measure and automate campaigns across channels.View full definition → subscriptions, sales and marketing salaries and commissions, referral bounties, and acquisition discounts (the free month, the €30 off the first box). It excludes money spent on people who already bought: account management, retention emails, loyalty programmes, customer support.

The denominator is new paying customers. Not leads, not signups, not trial starts, not free-tier accounts. If someone has never paid you, they have not been acquired; they have been reached.

One structural wrinkle: spend and customers do not arrive in the same month. In a business with a ninety-day sales cycle, March media produces June customers, so a month-by-month CAC swings for reasons that have nothing to do with efficiency. Longer periods, or spend lagged against the cohort it actually bought, fix this.

LTVLTVLifetime Value: the total revenue (or profit) a customer generates throughout their entire relationship with your business.View full definition →, Customer Lifetime Value, is the total value you expect from one customer across the whole paying relationship. The starting formula: average order value multiplied by purchase frequency multiplied by expected lifespan. A customer spending $200 per order, three times a year, for four years, gives $2,400.

Two things about that number. First, $2,400 is revenue, and revenue is not what pays back CAC; gross profit or contribution margin is. Second, "lifetime" is a forecast. Until a cohort is fully churned out, every LTV figure is a projection built on an assumed churn ratechurn rateChurn rate is the percentage of customers or revenue lost over a period. It measures how fast a business loses its existing customer base.View full definition →, and in subscription businesses the implied lifespan is usually just 1 divided by the monthly churn rate. A 4% monthly churn implies 25 months. Nudge the churn assumption to 5% and you have deleted five months of revenue from every customer you own.

Whose lifetime also needs a ruling. The seat or the account? The individual shopper or the household that shares one HelloFresh box? Whatever you pick, the same unit has to appear in the CAC denominator, or the ratio between the two is nonsense.

ROASROASReturn on Ad Spend (ROAS) measures the revenue generated for every unit of currency spent on advertising, calculated as revenue divided by ad cost.View full definition →, Return on Ad Spend, is revenue attributed to advertising divided by the advertising spendadvertising spendAny media you pay for: display ads, search ads, social ads, and sponsorships. You buy access to someone else's audience on a per-click, per-impression, or flat-fee basis.View full definition → that earned it. Spend $100,000 on Google Ads, get $400,000 in attributed revenue, ROAS is 4x.

What ROAS does not include: cost of goods, shipping, payment processing, returns and refunds, discounts, agency fees, salaries, or any spend outside the platform being measured. It is top-line, gross, and attributed by a model you chose. It answers one narrow question (did this ad spend produce revenue) and cannot answer whether the company made money.

Key sub-concepts

1. The LTV:CAC ratio

The comparison only works if both sides are measured in the same currency: gross-profit LTV over fully loaded CAC. Revenue LTV divided by media-only CAC will flatter you by a factor of three or more. A 3:1 ratio is the conventional floor for subscription and DTC businesses; the module's frameworks lesson deals with how to compute it properly and when to distrust it.

1. Payback period

The number of months of gross profit from one customer needed to recover the CAC that bought them. CAC of $600 against $100 of monthly gross profit gives six months. Payback and LTV:CAC answer different questions: one is about cash timing, the other about eventual return. A business can have a healthy ratio and still run out of money waiting for it.

1. Blended CAC versus channel-level CAC

Blended CAC puts all acquisition spend over all new customers, including the ones who came through word of mouth or direct traffic and cost you nothing. That makes it a real number for the finance team and a misleading one for budget allocation. Channel-level CAC divides one channel's spend by the customers that channel produced, which requires you to decide, in advance, how a customer touching four channels gets counted.

1. ROAS versus MER

MER, Media Efficiency Ratio, is total revenue divided by total ad spend across every channel. It ignores attribution entirely, which makes it blunt and hard to game. A brand doing $10M on $2M of ad spend has an MER of 5x. When channel ROAS climbs while MER falls, the attribution model is reallocating credit, not creating growth.

Real-world cases

CASE 1: HubSpot and the denominator question

HubSpot sells the CRM and marketing software many teams use to produce exactly these reports, and it publishes the raw material for its own: total customers (past 200,000) and average subscription revenue per customer, disclosed quarterly. Anyone can divide sales and marketing expense by net new customers and get a defensible CAC. The interesting part is the denominator. HubSpot runs a free CRM tier with millions of users. Counting those as acquired customers would drive CAC toward zero and make the number meaningless. The definitional line, paid conversion, is what keeps the metric honest.

CASE 2: HelloFresh and whose lifetime you are measuring

HelloFresh acquires with a heavily discounted first box, which means the opening order is frequently sold below cost. The acquisition is only paid for by the third, fifth, tenth delivery. The company reports active customers, orders per customer and contribution margin, and contribution margin is the correct base for LTV here: revenue per box tells you almost nothing once ingredients, packaging and refrigerated logistics are removed. A second definitional trap sits in "active": a customer who pauses for four months and comes back is not a new acquisition, and the spend that woke them up belongs to retention, not CAC.

CASE 3: Dollar Shave Club and what ROAS cannot see

The 2012 launch video cost roughly $4,500 to produce and drove around 12,000 orders in the first 48 hours. Judged on immediate ROAS, those orders were close to worthless: the entry plan was a dollar a month plus shipping, so attributed revenue against production and hosting looked thin. The whole business rested on refill subscriptions running for years, which no ROAS figure captures because ROAS only sees the transaction inside the attribution window. Unilever bought the company in 2016 for a reported $1 billion. Read the campaign on ROAS and you kill it; read it on lifetime value and you buy more of it.

CMO action items

• Write down the CAC inclusion list and publish it. Every ambiguous line (brand campaigns, sponsorships, the discount on the first order, the growth engineer's salary) gets a ruling in writing, with a date. Revisit it once a year, not once a quarter, or your trend line becomes uncomparable.
• Define the customer whose lifetime you measure, and the events that open and close it. Free users, trialists and repeat households each need an explicit ruling, and the same unit must appear in the CAC denominator.
• Require every ROAS figure that reaches an exec meeting to state its revenue basis and attribution window on the same slide. A 4x with no window is not a number, it is a mood.

Common mistakes

Mistake 1: A media-only numerator

Most reported CAC covers ad spend and nothing else: no salaries, no creative production, no tooling, no agency retainers, no acquisition discounts. Fully loading those costs commonly moves CAC up by a third or more, and in sales-assisted B2B it can double. A company confident in its $150 CAC may be operating at $240. Every allocation decision built on the smaller number is wrong in the same direction.

Mistake 2: Measuring LTV in revenue

A meal-kit customer generating $2,400 of revenue at 25% contribution margin gives you $600 to pay for acquisition, service and overhead. Compared against a $500 CAC, the revenue version says you are printing money and the margin version says you are barely covering the cost of the customer. Pick the margin version and say so on the label.

Mistake 3: Reading ROAS as profitability

ROAS is a revenue multiple on ad spend and nothing more. It ignores COGS, fulfilment, returns and overhead. A 4x on a product with 25% gross margin loses money on every incremental dollar. Convert to gross-profit ROAS (gross profit generated divided by ad spend) before any scaling decision, and know your break-even multiple by product line before the campaign goes live.