The ROAS Trap: Why a 6x Campaign Might Be Worth Less Than a 2x Campaign

The obvious move is to shift budget from the low-performer to the high-performer. You do it, and your total revenue drops. This is the ROAS trap, a scenario where refining for a seemingly positive metric leads to negative results because the metric itself is flawed. You are making decisions based on a metric that is fundamentally broken, a vanity metric that feels good but says nothing about profitability or growth.

The Problem: Your ROAS is a Lie

Unlike causal metrics that identify which sales were a direct result of ad spend, ROAS simply shows an association, which is a flawed way to measure marketing effectiveness. For ecommerce brands, this distinction is the difference between scaling profitably and burning cash.

Its simplicity is its greatest flaw. ROAS, as calculated by ad platforms like Meta and Google, is a correlation metric, not a causal one. It shows you what revenue is associated with a campaign, not what revenue was caused by it. This is a critical distinction that many marketers fail to grasp.

Think about it. A customer sees your TikTok ad, browses your site, and leaves. A week later, they see a retargeting ad on Instagram, click it, and make a purchase. The Instagram ad platform, using a last-touch attribution model, will claim 100% of the credit for that sale. The reported ROAS for that retargeting campaign will be sky-high. But did that ad cause the sale?

This is the core of the ROAS trap. You are refining for credit, not for cause. You are rewarding the channels that are best at being the last click, not the channels that are best at creating new customers.

1. You see a high ROAS on a retargeting campaign. 2. You shift more budget to that campaign. 3. Your prospecting efforts, which feed the retargeting campaign, suffer. 4. Your pool of potential new customers shrinks. 5. Your overall revenue stagnates or declines, even as your reported ROAS stays high.

You are stuck in a loop, pouring money into a campaign that is simply harvesting the demand created by other channels. You are a victim of cannibalistic channels, and your marketing attribution platform is sending you the wrong signals.

The Slow Bleed of Misleading Metrics

Misleading metrics like ROAS cause a slow, silent erosion of your growth potential by encouraging investment in low-impact campaigns while starving high-impact ones. Unlike metrics that measure incremental lift, ROAS does not distinguish between sales that were caused by ads and those that would have happened anyway. This leads to a cycle of refining for vanity metrics instead of actual business growth.

Every decision you make based on misleading ROAS data is a step in the wrong direction. You are flying blind, and the consequences are severe.

Consider the 6x vs. 2x scenario. The campaign is simply there to capture the final click. Its incremental impact, the sales that would not have happened without it, is close to zero. This is why many are now arguing to move beyond ROAS to iROAS (incremental ROAS).

Its reported ROAS is lower because it is doing the hard work of demand generation. Many of the customers it influences will not convert immediately. They will enter a consideration phase, be touched by other channels, and eventually convert through a different touchpoint.

You are refining for the short-term illusion of efficiency while sacrificing the long-term health of your business. You can calculate your potential losses with our waste calculator.

Escape the Trap with Causal Inference

Causal inference is the solution to the ROAS trap because it measures the true cause-and-effect relationship between your ads and sales. Unlike traditional attribution that relies on correlations, causal inference determines the incremental sales generated by each campaign. This allows you to calculate a Causal ROAS, a metric that reflects the actual profit generated from your ad spend.

To escape the ROAS trap, you must stop measuring correlation and start measuring causation. As Hal Varian, Google’s Chief Economist, wrote, the key is to compare actual outcomes to counterfactual outcomes [1]. This is the foundation for understanding the true impact of your marketing.

This is the concept of incremental sales. Incremental sales are the sales that were directly caused by your marketing efforts. They are the sales that would not have occurred otherwise. This is the true measure of a campaign’s value. You can learn more about how to measure this in our guide to incrementality testing.

Incremental Sales = Total Sales - Sales That Would Have Happened Anyway (Baseline Sales)

By measuring incremental sales, you can calculate your true, causal ROAS:

Causal ROAS = Incremental Revenue / Ad Spend

This is the metric that should guide your budget allocation. It tells you the real return on your investment, free from the distortions of attribution models. Suddenly, your decision is clear. You can model different scenarios with our ROAS calculator.

Geo-lift tests involve showing ads to a specific geographic region while withholding them from a similar control region.

FAQ

What is the ROAS trap?

The ROAS trap is the common mistake of refining marketing spend based on the reported Return on Ad Spend from ad platforms. This metric is often misleading because it measures correlation, not causation, leading marketers to over-invest in campaigns that capture existing demand and under-invest in campaigns that create new demand.

Why is a high ROAS misleading?

A high ROAS is misleading because it does not account for incremental sales. A campaign can have a high ROAS by targeting users who were already going to buy, effectively taking credit for sales that would have happened anyway. This inflates its perceived value while its true, causal impact on revenue is low.

How do I calculate my true ROAS?

To calculate your true ROAS, you must measure the incremental revenue generated by a campaign and divide it by the ad spend.

What is a better alternative to ROAS?

The best alternative to traditional ROAS is Causal ROAS, which is based on incremental sales. This metric provides a true measure of a campaign’s effectiveness by isolating the revenue that was directly caused by your advertising efforts, allowing for much smarter budget allocation and a clearer understanding of marketing’s impact.

How is Causal ROAS different from blended ROAS?

Blended ROAS is a simple calculation of total revenue divided by total ad spend. While it can be a useful health metric, it does not provide any insight into the performance of individual channels or campaigns. Causal ROAS, on the other hand, uses causal inference to isolate the incremental impact of each marketing activity.

Find your true ROAS.

Find your true ROAS.

References

[1] Varian, H. (2016). Causal Inference in Economics and Marketing. Proceedings of the National Academy of Sciences, 113(27), 7310-7315. Available at: https://people.ischool.berkeley.edu/~hal/Papers/cause-PNAS4.pdf

[2] Gordon, B. R., Zettelmeyer, F., Bhargava, N., & Chapsky, D. (2019). A Comparison of Approaches to Advertising Measurement: Evidence from Big Field Experiments at Facebook. Marketing Science, 38(2), 193-225. Available at: https://pubsonline.informs.org/doi/abs/10.1287/mksc.2018.1141