Performance Marketing and ROAS

ROAS, defined with a worked example

ROAS = attributed revenue ÷ ad spend. If you spend USD 1,000 on ads in a month and attribute USD 3,000 of revenue to them, your ROAS is 3.0 (often written 3:1 or 300%). That's the whole formula — the complications are in the words "attribute" and "revenue."

Your breakeven ROAS comes from your margin

A ROAS of 3.0 sounds great until you do margin math. Worked example: say your gross margin is 40% — of every USD 100 in revenue, USD 40 is left after the cost of goods. Breakeven ROAS = 1 ÷ margin = 1 ÷ 0.40 = 2.5. At ROAS 3.0 you're making money, but modestly: USD 3,000 revenue × 40% margin = USD 1,200 gross profit, minus USD 1,000 spend = USD 200 actual profit. At a 25% margin the same "great" 3.0 ROAS is losing money (breakeven would be 4.0). Compute your own breakeven before judging any campaign — and remember the agency management fee belongs in the cost side too (see management fees vs ad spend).

Why platform-reported ROAS overstates

• Attribution windows: platforms claim credit for purchases that happen days or weeks after a click — including buyers who would have purchased anyway.
• View-through credit: some setups count people who merely saw an ad and later bought through another door.
• Double counting: run Google and Meta together and both may claim the same sale. Add their dashboards and you'll "earn" more than your bank account did.

None of this makes platform ROAS useless — it's a fine relative signal for comparing campaigns within one platform. It's just not your profit statement.

MER: the honest number for small accounts

For most small and mid-size accounts, the cleaner check is blended measurement — MER (marketing efficiency ratio): total revenue ÷ total marketing spend, no attribution games. Worked example: USD 20,000 total monthly revenue ÷ USD 4,000 total marketing spend = MER 5.0. Watch the trend: if spend rises and MER holds or climbs, the marketing is working; if spend rises and MER sinks, the platform dashboards are flattering you. Small accounts rarely have the data volume for fancier attribution to mean anything — blended truth beats granular fiction.

The ROAS-chasing trap

The classic mistake: retargeting campaigns always show the prettiest ROAS — they harvest people your prospecting already warmed up. Cut prospecting to feed retargeting, and ROAS rises while revenue falls; eventually the retargeting pool empties because nothing refills it. Judge prospecting and retargeting as one system, not as competing line items. The same trap appears as over-cutting: maximum ROAS is usually achieved at minimal spend — and minimal profit. You're optimizing total profit, not the prettiest ratio.

Measurement setup, briefly

• Conversion tracking verified before spending — test a purchase or lead end-to-end and watch it land in the platform and your analytics.
• Consistent UTM tagging so your analytics can tell channels apart.
• Consent and signal loss: cookie consent and ad blockers mean platforms see less than reality — another reason blended MER matters. Server-side tracking can recover some signal; weigh its complexity against your account size.

A monthly reallocation routine

• 1. Pull spend, platform conversions, and blended revenue for the month.
• 2. Compute MER and compare against last month and your breakeven math.
• 3. Within each platform, kill the bottom creative/audiences and feed the winners.
• 4. Shift budget between platforms only on blended evidence, in modest steps — big swings reset learning phases.
• 5. Write down what you changed and why; next month's review starts from that note.