Curriculum
37 docsROAS Is the Devil: Why In-Platform Metrics Lie
ROAS Is the Devil: Why In-Platform Metrics Lie
Module: Google and YouTube Experts Instructor: John Moran Revenue Rush University
The Problem with ROAS
I charge $2,500 an hour and the first thing I tell every client is this: please, ignore in-app ROAS. It's like a broken swing — it's only going to let you down.
Here's what 400% ROAS usually means: warm traffic, branded traffic, or existing customers. Google takes credit for the sale. Facebook takes credit for the same sale. Your customer was going to buy anyway because they saw your product three times across four platforms. Both platforms claim the conversion, and when you add up all your platform-reported revenue, it's 40-60% more than your actual bank deposit.
You can't trust Google Ads data. You can't trust Meta data. The only number you can trust is the one in your bank account.
MER: The Metric That Actually Matters
MER = Total Revenue / Total Marketing Spend
That's it. Total revenue from ALL channels (Shopify tells you this number). Total marketing spend across ALL platforms. Divide.
Example: - Total Shopify revenue this month: $50,000 - Total ad spend (Meta + Google + email tools): $15,000 - MER = $50,000 / $15,000 = 3.33
This number doesn't lie. It doesn't double-count. It doesn't have attribution windows or view-through credits or data modeling. It's money in divided by money out.
How to Use MER
MER as your decision framework:
| Scenario | What It Means | Action |
|---|---|---|
| MER stable, Meta ROAS dips | Platform misattributing, business is fine | Don't panic. Monitor for 3-5 days. |
| MER stable, Google ROAS dips | Same — platform noise | Don't panic. |
| Meta ROAS up, MER down | Meta is STEALING credit from organic/email | Investigate. Platform looks good but business is worse. |
| MER up, all platform ROAS stable | Something else is driving growth (organic, email, referral) | Find it and double down. |
| MER down, all platform ROAS stable | Rising costs or declining organic | Investigate cost structure and organic traffic trends. |
The most dangerous scenario: Meta shows improving ROAS while MER declines. This happens when Meta's pixel "gets smarter" at claiming credit for existing customers while you're actually acquiring fewer new ones. If you only watched Meta, you'd think things are improving. Your bank account tells the truth.
nCAC: The Other Metric You Need
nCAC = New Customer Acquisition Cost
Platform-reported CPA includes EVERYONE who purchased — new and returning. If your retargeting brings back 50 existing customers cheaply, your blended CPA looks amazing. But you didn't acquire 50 new customers — you just reminded 50 existing ones to reorder.
nCAC counts only genuinely new customers.
nCAC = Prospecting Spend / New Customers Acquired
This is the number that tells you whether your growth is real or an illusion.
nCAC Benchmarks by Vertical (DTC)
| Vertical | Healthy nCAC | Danger Zone |
|---|---|---|
| Supplements / Wellness | $25-45 | >$60 |
| Skincare / Beauty | $30-50 | >$70 |
| Apparel | $20-40 | >$55 |
| Food / Beverage | $15-35 | >$50 |
If your nCAC is above the danger zone AND your LTV doesn't justify it (LTV:nCAC ratio below 3:1), you're buying customers you'll never profit from.
CAC Payback Period
CAC Payback = nCAC / Average Monthly Revenue Per Customer
This tells you how many months it takes to recoup the cost of acquiring a customer.
Example: - nCAC: $60 - Average monthly spend per retained customer: $35 - CAC Payback: 1.7 months
If your CAC Payback is under 3 months, you can scale aggressively — the customer pays for themselves quickly. If it's 6-12 months, you need strong retention or you're building a house of cards. Over 12 months? Stop scaling until you fix retention.
The nCAC Ceiling: Your Emergency Stop
Set this number before you spend a dollar. The nCAC Ceiling is the absolute maximum you're willing to pay for a new customer, based on your unit economics.
nCAC Ceiling = (Average LTV × Target Margin) - COGS - Fulfillment
If your average customer LTV is $180, your target margin is 30%, COGS is $25, and fulfillment is $13: nCAC Ceiling = ($180 × 0.30) - $25 - $13 = $54 - $25 - $13 = $16
Wait, that seems low. And it IS. That's the point — it forces you to be honest about your economics. If your LTV can't support your acquisition cost, either improve LTV (retention, AOV, email) or accept that your current nCAC is burning cash and you're betting on future improvements.
The nCAC Ceiling isn't aspirational. It's the line where you pull the emergency brake on spend.
What This Means Practically
-
Set up MER tracking immediately. One spreadsheet. Weekly. Total Shopify revenue divided by total ad spend. If you do nothing else from this module, do this.
-
Track nCAC separately from blended CPA. Use Shopify's "First-time vs returning" customer filter plus your prospecting spend to calculate real nCAC.
-
Never make a scaling decision based on platform ROAS alone. Always cross-reference with MER. If platform ROAS says increase budget but MER is declining, trust MER.
-
Run your own incrementality tests. The simplest: pause a channel for 5-7 days and see what happens to total revenue. If you pause Google and total revenue barely moves, Google was taking credit for organic conversions. If revenue drops 20%, Google was doing real work. Now you know.