THE ROAS NUMBER THAT’S LYING TO YOU

Every week I speak to brands running paid media who are happy with their ROAS. They're hitting a 4x, 5x, sometimes 8x return on ad spend as reported in Meta Ads Manager or Google Ads. Business is good. The ads are working.

Then I look at their P&L. The numbers don't match. Revenue is growing but margin is flat or declining. CAC is creeping up quarter over quarter. And when we trace it back, the problem is almost always the same: platform ROAS and real ROAS are not the same number.

Why Platform ROAS Is Misleading

Meta and Google measure revenue attribution within their own window — typically 7-day click or 1-day view for Meta, last-click for Google. This creates several structural problems:

• Double-counting — A customer who saw a Meta ad, clicked a Google ad, then converted through direct traffic might be counted by both platforms. Your combined reported ROAS of 6x might be a real ROAS of 3x.
• View-through inflation — Meta counts conversions from people who saw your ad but didn't click it. These are often people who would have converted anyway.
• Branded search attribution — Google often takes credit for conversions that came from branded searches driven by Meta's top-of-funnel spend. Both platforms claim the same customer.
• Return blindness — Platform ROAS is calculated on gross revenue. If your return rate is 30%, your real revenue is 30% lower than reported.

How to Calculate Your Real ROAS

Real ROAS = Net Revenue Attributable to Paid / Total Paid Media Spend

Net revenue means after returns, after refunds, after discounts applied. Total paid spend means including agency fees and creative costs — not just the media spend you put into the platform.

For attribution, use incrementality testing where possible. If you can't run a holdout test, use a media mix model or at minimum a last-touch MTA that de-duplicates across platforms. Don't trust any platform's native attribution as your source of truth.

The Three Numbers That Actually Matter

• blended CAC — total marketing spend / new customers acquired (all channels, all costs)
• LTV:CAC ratio — if this is below 3:1, the business model doesn't support the acquisition cost
• Contribution margin after acquisition cost — what's left after COGS and marketing per order?

The Three Levers That Actually Move Real ROAS

1. Improve Post-Click Conversion

The fastest way to improve ROAS is to convert more of the traffic you're already buying. A landing page improvement from 2% to 3% conversion rate is a 50% ROAS improvement with zero additional spend. Most brands under-invest here relative to the return.

2. Improve Average Order Value

ROAS is a ratio — the denominator is fixed (your spend). Increasing average order value through bundles, upsells, or subscription conversion directly improves the ratio without changing media efficiency.

3. Reduce CAC Through Creative Diversification

Creative fatigue is the most common driver of rising CAC in mature paid media accounts. Systematic creative testing — not guessing what will work, but building a creative framework that generates and tests hypotheses at scale — typically yields 20-40% CAC reduction within two quarters.

Platform ROAS is a useful operational metric. It tells you whether an ad set is working relative to others. It is not a business health metric. Conflating the two is how brands end up growing revenue and destroying margin simultaneously — and not understanding why until it's too late.

Know your real number. Manage to that.