Your ROAS Is Wrong: How to Calculate Real Return on Ad Spend
Google says your ROAS is 4.2x. Meta says 3.8x. Your bank account says neither is true. Here's how to calculate real ROAS — and why the difference averages 25-40%.
Your Google Ads dashboard says your ROAS is 4.2x. Your Meta Ads dashboard says 3.8x. You're feeling good.
Then your accountant closes the quarter. Revenue is 31% lower than what the ad platforms reported. You had returns. Chargebacks. And both Google and Meta claimed credit for the same customer.
Welcome to the ROAS lie. It affects every business running paid advertising, and it's costing the industry $37 billion annually in misattributed spend.
The Real ROAS Formula
Platform-reported ROAS:
Reported ROAS = Platform-Attributed Revenue / Ad Spend
Real ROAS:
Real ROAS = (Collected Revenue - Returns - Chargebacks) / Total Ad Spend
The difference between these two numbers averages 25-40% for most e-commerce businesses. Here's why.
Why Platform ROAS Is Always Higher Than Reality
1. They Count Revenue That Never Collected
A customer clicks your ad, adds items to cart, and completes checkout. Google counts that as a $200 conversion. But:
- The payment fails (3-5% of all transactions)
- The customer returns the item (15-30% in e-commerce)
- The customer disputes the charge (0.5-1.5% of transactions)
Google still shows $200 in attributed revenue. Your bank got $0.
2. Both Platforms Claim the Same Conversion
A customer sees your Meta ad on Monday, clicks your Google ad on Wednesday, and buys on Friday. Both platforms count the full revenue. You spent $50 across both platforms and made one $200 sale, but your dashboards show $400 in total attributed revenue.
This is called double attribution, and it inflates your total reported ROAS by 20-60% depending on how many channels you run.
3. They Take Credit for Organic Sales
Most platforms use long attribution windows (Google: 30 days, Meta: 7 days click / 1 day view). A customer who saw your Meta ad once, forgot about it, and found you through Google search two weeks later? Meta claims that sale.
Studies show 15-30% of platform-attributed conversions would have happened organically.
Worked Example: The ROAS Reality Check
Let's say you run an e-commerce store:
| Metric | Value | |--------|-------| | Total ad spend (Google + Meta) | $10,000 | | Google-reported conversions | $28,000 | | Meta-reported conversions | $18,000 | | Total platform-reported revenue | $46,000 | | Platform-reported combined ROAS | 4.6x |
Now let's calculate real ROAS:
| Adjustment | Value | |------------|-------| | Actual Shopify orders from ad-attributed customers | $32,000 | | Minus returns (22%) | -$7,040 | | Minus chargebacks (1.2%) | -$384 | | Minus orders that would have happened organically (est. 20%) | -$4,915 | | Actual collected revenue from ads | $19,661 | | Real ROAS | 1.97x |
Platform-reported: 4.6x. Reality: 1.97x. The platforms overstated ROAS by 133%.
If your target ROAS is 3x, you'd think you're crushing it. In reality, you're losing money on every ad dollar.
How to Calculate Your Real ROAS
Step 1: Get Your Actual Revenue
Pull collected revenue from your payment processor (Stripe, Shopify Payments), not your ad platforms. This is the money that actually hit your bank account.
Step 2: Subtract Returns and Chargebacks
Your payment processor tracks refunds and disputes. Subtract these from collected revenue.
Step 3: Estimate Organic Overlap
This is the hardest part. Methods:
- Holdout test: Turn off ads for a geographic region and measure organic sales
- Incrementality test: Randomly hold out a percentage of your audience and compare
- Conservative estimate: Assume 15-25% of attributed conversions were organic
Step 4: Divide by Total Ad Spend
Use your total spend across all platforms — not what each platform reports individually.
Step 5: Compare to Platform Numbers
The gap between your real ROAS and platform-reported ROAS is your attribution inflation rate. Track this monthly. If it grows, your platform data is becoming less reliable.
The Cross-Channel Problem
Running ads on multiple platforms makes this worse, not better. Each platform optimizes for its own attribution:
- Google optimizes for last-click attribution (takes credit for the final click)
- Meta optimizes for view-through attribution (takes credit for ad impressions)
- TikTok uses a 7-day click / 1-day view window
- LinkedIn has a 30-day attribution window for lead gen
None of these platforms have incentive to share attribution fairly. They all want to show that their platform drove the most revenue — because that's what keeps you spending.
The only source of truth is your payment processor. Money either arrived in your bank account or it didn't. Work backward from there.
How NuMoon Solves This
NuMoon connects your ad platforms directly to your payment processor. It:
- Pulls actual payment data from Stripe/Shopify — not platform-reported conversions
- Matches customers across platforms using email and transaction data
- Deduplicates attribution so each sale is counted once, not per-platform
- Subtracts returns and chargebacks automatically
- Calculates real ROAS per channel daily, not monthly
- Alerts you when a campaign's real ROAS drops below your target
The result: you know exactly which ad dollars make money and which ones burn it — in real time, not after your quarterly close.
Frequently Asked Questions
How much does ROAS inflation vary by industry?
E-commerce typically sees 25-40% inflation. SaaS sees 15-25% (fewer returns, but longer sales cycles muddy attribution). Local services see 30-50% (heavy reliance on phone calls that platforms can't track).
Should I stop trusting platform data entirely?
No. Platform data is useful for relative comparisons (Campaign A vs Campaign B within the same platform). It's unreliable for absolute numbers (total revenue driven by ads). Use platform data for optimization, payment data for evaluation.
How often should I calculate real ROAS?
Weekly at minimum. Daily if you're spending more than $5K/month on ads. The faster you catch a campaign with declining real ROAS, the less money you waste.
What's a good real ROAS target?
After accounting for returns and organic overlap:
- 3x+ real ROAS: Profitable after COGS for most businesses
- 2x real ROAS: Breaking even after COGS (margin-dependent)
- Under 2x: Likely losing money on ads
Can I calculate real ROAS without connecting tools?
Technically yes — download data from each platform, your payment processor, and reconcile in a spreadsheet. Practically, this takes 3-5 hours per week and most founders stop doing it after month 2. Automation is the only sustainable approach.
Stop Trusting the Scoreboard. Check the Bank Account.
Every ad platform is incentivized to make their numbers look good. It's not malicious — it's their business model. They want you to spend more.
Your job is to know the real numbers. Connect your ad spend to your actual revenue. Calculate real ROAS. Make decisions based on money collected, not money reported.
Take the free health scan to see how much your ROAS numbers might be inflated — and how to fix it.