Your agency reports a 10x ROAS on your Meta campaigns. Sounds incredible, right? In some cases, it is. In many others, that number is a carefully crafted illusion — and understanding the difference could save your business tens of thousands of ringgit per month.
This article isn't about bashing agencies. We are one. It's about helping marketers understand the metric that actually determines whether their advertising is working — and why the most commonly reported number is often the most misleading.
The Problem with Platform-Reported ROAS
When your agency shows you a "10x ROAS" figure, they're almost certainly using the number reported by the ad platform — Meta Ads Manager, Google Ads, TikTok Ads, etc. Here's why that number is almost always inflated:
1. Attribution Windows Overlap
A customer sees your Meta ad on Monday, clicks your Google ad on Tuesday, sees a TikTok ad on Wednesday, and buys on Thursday. All three platforms claim credit for the same sale. Your "combined ROAS" across platforms adds up to something absurd — because you're counting the same revenue three times.
Here's how this typically plays out: A Malaysian e-commerce brand's Meta ROAS shows 5.2x, Google ROAS shows 4.8x, and TikTok ROAS shows 3.1x. Added together, that implies 13.1x total return. But the actual blended ROAS — total revenue divided by total ad spend — is only 3.8x. The gap? Double and triple-counted conversions across platforms.
2. View-Through Attribution Inflation
Meta's default attribution window includes people who merely saw your ad (didn't click) and then purchased within 7 days. For brands with existing organic traffic and brand awareness, a huge chunk of these "view-through conversions" would have happened anyway. The ad didn't cause the sale — it just took credit for it.
Here's what an incrementality test typically reveals: when Meta ads are turned off in one state for two weeks while running everywhere else, revenue in the holdout region often drops by only 20-25%, despite Meta reporting that it was "driving" 60% of conversions. That can mean over half of Meta's claimed conversions were not incremental — those sales would have happened regardless of the ads.
3. Branded Search Cannibalization
This is the most expensive mistake in digital marketing. Your agency runs Google Search ads on your brand name. These ads appear above your organic listing for people who were already searching for your brand. Of course they convert well — the customer already knew you!
Brand search campaigns often report 15-25x ROAS because they're capturing demand that already existed. Turning them off would shift most of that traffic to your organic listing at zero cost. But they look amazing in reports, so agencies keep running them.
The Metric That Actually Matters: Blended ROAS
Blended ROAS is dead simple:
Blended ROAS = Total Revenue ÷ Total Marketing Spend
No platform attribution. No view-through credit. No double counting. Just: how much money came in, and how much did you spend on marketing? This is the number your CFO cares about — and it's the only number that reflects reality.
If your total marketing spend is RM50,000/month and your total revenue is RM200,000/month, your blended ROAS is 4x. Period. It doesn't matter what Meta, Google, or TikTok claim individually.
How to Calculate True Incremental ROAS
Blended ROAS tells you overall efficiency, but it doesn't tell you which channels are actually driving growth. For that, you need incrementality testing. Here are three recommended methods, ranked from simplest to most rigorous:
Method 1: Geo-Holdout Tests
Turn off ads in one geographic region while keeping them running elsewhere. Compare revenue changes. This mirrors the incrementality test example above. It's not perfect — regions aren't identical — but it's practical and directionally accurate.
Method 2: Media Mix Modeling (MMM)
Statistical models that analyze how changes in spending across channels correlate with changes in revenue, controlling for external factors like seasonality, promotions, and competitor activity. This requires 12+ months of historical data to be reliable, but it gives you a channel-level view of true contribution.
Method 3: Conversion Lift Studies
Meta and Google both offer randomized controlled experiments where your audience is split into exposed and holdout groups. The difference in conversion rates between the two groups shows the true incremental impact of your ads. These are the gold standard for incrementality, but they require significant spend and statistical sophistication to interpret correctly.
What to Do About It: A Practical Framework
- Start tracking blended ROAS today. Pull total revenue and total ad spend into a single spreadsheet. Update weekly. This number is your north star.
- Audit your branded search spend. If you're spending more than 5% of your budget on brand terms, run a holdout test to see how much revenue you'd lose by turning them off. The answer is usually "less than you think."
- Switch Meta attribution to 7-day click only. Removing view-through attribution from your reporting gives you a cleaner (though still imperfect) picture of ad-driven conversions.
- Run a geo-holdout test on your biggest channel. Pick your highest-spend platform, turn it off in one state for two weeks, and measure the revenue impact. The results will either validate your spend or reveal significant waste.
- Ask your agency hard questions. "What's our blended ROAS?" "How much of our Google spend is on brand terms?" "Have you run any incrementality tests?" If they can't answer these questions clearly, that's a red flag.
Why Blended ROAS Belongs in Every Performance Report
Blended ROAS should be the first metric in every performance report. Not because it looks impressive — in fact, blended ROAS is always lower than platform-reported ROAS, which makes the numbers look worse on paper compared to agencies that cherry-pick platform metrics.
But it's the honest number. And the job of any good agency isn't to generate impressive-looking reports — it's to grow your business. When you measure what matters, you make better decisions about where to invest, what to cut, and how to scale.
Want to know your true blended ROAS? Book a free strategy session — we'll audit your current reporting, identify where numbers are being inflated, and show you a clear picture of your real marketing performance.