THE CURRICULUM / CH 3 — Marketing Mastery

3.4Understanding Metrics

Measured against modelled, the funnel as a diagnostic, and break-even derived from your own margin.

5 min read

It is Sunday night. Say the ads manager reads 3.1 return on ad spend and forty-eight purchases, your store admin shows thirty-one orders, and your bank balance is lower than it was on Monday. All three of those are correct, and you have to decide tonight whether Monday spends anything.

Two kinds of number

Measured

  • A click. An impression.
  • Orders in your store.
  • Money in the bank.

Modelled, in part

  • A purchase in a later session, credited to an earlier ad.
  • Anything matched across devices, or estimated where consent is missing.
  • View-through credit.
Three tiers of number ranked by how much you can trust them. Measured: money that landed in the bank, orders in your admin, invoices you paid. Modelled: platform-attributed conversions, anything with a lookback window, blended estimates. Unknown: what you assumed, what you were told, what you hope. This is a rank, not a score.
The same two, with a third tier underneath them.

Two platforms with different attribution defaults can both credit the same order, which is how reported revenue exceeds real revenue with nobody lying.

One purchase, counted by two different ledgers. Two things decide whether the platform records it. First the window: how long after the ad a purchase still counts, so a purchase made inside the window is counted and one made after it closed is not. Second the device: whether the half that saw the ad and the half that bought can be joined up, so a purchase on the same device is joined up and one on another device may or may not be. The platform counts events; your bank counts money. No figures appear.
One purchase, two ledgers. The platform counts events; your bank counts money.

THE BLENDED FIGURE CANNOT DOUBLE-COUNT ITSELF.

total revenue over total ad spend

THE FUNNEL AS A DIAGNOSTIC

The five funnel stages from impressions to purchases, each with a narrowing bar, and between them what a drop at that step points at: the hook or audience, a page that does not match the ad, the offer or price or proof, and checkout or shipping or trust. No conversion rates are shown.

Where it drops

StageA collapse here means
Barely deliveringBudget, bid, audience size or review status.
Impressions to clicksThe creative, or its match to the audience.
Clicks to page viewsTechnical. Slow page, broken link, tracking.
Views to add to cartThe page and the offer.
Cart to purchaseFriction. Forced accounts, late costs, payment methods.

Which definitions decide anything

MetricWhat it decides
CTRLink clicks over impressions. Creative and audience fit.
CPC and CPMDerived, and a market signal. Neither is a target.
AOVRevenue over orders. The lever beginners ignore.
CPA and ROASReadable only against your margin.

a vanity metric changes no decision.

A VANITY METRIC CHANGES NO DECISION.

Reading the pattern, not the number

SignalHealthyAverageConcerning
Link click-throughAt or above your own rolling averageIn line with your historyFalling while frequency rises
Cost per purchaseBelow break-evenHovering at break-evenAbove it for a full window, no trend
  • Link click-through — Somebody else screenshot is not a benchmark.
A store daily dashboard: total sales, 159 orders, a 6.19 percent conversion rate, 15 live visitors.
One store, one day. Everything that decides the question is missing.
Gross revenue falling to net profit. The top bar is gross revenue, drawn full width. Each line below removes a slice of it in order: refunds and chargebacks, cost of goods, shipping and fulfilment, payment and platform fees, advertising, and apps, tools and subscriptions. The solid bar in each row is what is still left at that point and the dotted piece beside it is what that line took. Every slice is drawn the same width on purpose, because no cost structure is being claimed. What remains at the bottom is net profit. No amounts, percentages or currency appear.
Gross revenue at the top, and every line that takes a slice on the way down.

BREAK-EVEN, FROM YOUR OWN MARGIN

A worked example of one order at seventy nine dollars ninety five drawn as a single vertical bar cut into slices, each sized by the dollars it takes. In order down the bar: the ATO takes seven dollars twenty seven in GST, the factory eleven forty, freight and customs three sixty, the courier twelve ninety nine, the payment processor one seventy, refunded buyers one twelve, the platform and apps one eighty seven, the ad platform twenty eight dollars, and the seller keeps twelve dollars. The ad platform slice is thirty five per cent of the order and the seller's slice is fifteen per cent. The nine slices add to seventy nine dollars ninety five exactly. It is a worked example, not a quote and not a result.
One order at a stated price, and every hand that takes a piece of it before you do.
  1. The price a customer actually pays, after discounts.
  2. Subtract the cost of goods.
  3. Subtract shipping and fulfilment, including what you absorb.
  4. Subtract fees, packaging, app charges, an observed refund allowance.
  5. What remains is contribution margin per order. Divide by price for the ratio.
A worked example deriving break-even return on ad spend from one product's contribution margin, step by step. Step one, the price is seventy nine dollars ninety five. Step two, everything except advertising takes thirty nine ninety five. Step three, contribution is forty dollars. Step four, contribution divided by price, forty over seventy nine ninety five, is nought point five zero zero. One divided by nought point five zero zero is two point zero zero, so break-even is two dollars of revenue for every dollar of advertising. Below, the same reciprocal is drawn at other margins as proportional bars: thirty per cent needs three point three three, forty per cent needs two point five, fifty per cent needs two point zero, sixty per cent needs one point six seven, and seventy per cent needs one point four three. On this price a thirty per cent margin leaves twenty three ninety nine to buy a sale while a sixty per cent margin leaves forty seven ninety seven. It is a worked example, not a quote and not a result.
The same worked example, carried through to the return ad spend has to make.

Break-even cost per purchase is your contribution margin per order, and nothing more complicated than that. Which is why a figure somebody else calls good is not information about your account.

Contribution per order plotted against cost per order. The line falls in a straight diagonal because every extra unit of cost comes straight out of contribution, and it crosses zero at one point: above that point every sale pays you, below it every sale costs you. The axes carry no magnitudes.
Break-even on the order. Software, salaries and your time sit outside.

TAKEAWAYS

  • Compare ads with platform figures. Judge the business on blended revenue over blended spend.
  • Fix the first stage of the funnel that collapses. Everything after it is distorted.
  • Break-even ROAS is one divided by your contribution margin ratio. Arithmetic, not a benchmark.
  • Decide what would make you act before you look at the number.

CHECK YOURSELF

Your contribution margin ratio is one quarter of the selling price. Your break-even ROAS is:

Clicks arrive on the page as expected, but add to cart has collapsed while your ads are unchanged. Where is the problem?

DO THIS BEFORE THE NEXT MODULE

  1. 01Work out the contribution margin per order on your main product, fees and refund allowance included.
  2. 02From it, calculate your break-even cost per purchase and put it where you will see it before a report.
  3. 03Put one week of platform purchases, store orders and bank deposits side by side, and explain the gap in a sentence.
Enrol free