Stop Trusting Blended ROAS
Blended ROAS is the most over-cited metric in performance marketing. Stop using it as a decision input. It conflates organic and paid revenue in a way that makes marketing look more effective than it is, particularly for brands with growing organic traffic. Here is what to measure instead, and how to build a defensible ad cost per unit when the platforms disagree with each other.
Why blended ROAS lies
Blended ROAS divides total revenue by total ad spend, including organic revenue that would have happened without any advertising. A growing brand with strong SEO, word of mouth, and repeat purchase behaviour will show excellent blended ROAS — not because advertising is efficient, but because organic revenue is growing and diluting the ad cost denominator. The inverse is equally misleading: a brand that cut organic investment and scaled paid will show worsening blended ROAS even if paid advertising is performing well.
The test is simple. If you turned off all paid advertising tomorrow, how much revenue would you still generate from organic channels within 30 days? That number is your organic revenue baseline. Subtract it from total revenue before calculating ROAS on paid channels. Only then do you have a number that tells you anything true about advertising effectiveness.
Marketing Efficiency Ratio as a better top-level metric
Marketing Efficiency Ratio (MER) — total revenue divided by total marketing spend including agency fees, creative production, influencer payments, platform ad spend, and attribution tooling — is more honest than blended ROAS as a portfolio-level efficiency metric. It captures the full cost of your marketing operation against the revenue it generates, without the platform-attribution distortions that make channel-level ROAS unreliable.
The limitation of MER is that it does not help you make channel-level allocation decisions. A healthy MER of 4.5x tells you your overall marketing operation is working. It does not tell you whether to shift budget from Meta to TikTok or whether your Google Shopping campaigns are overfunded. For those decisions, you need channel-level metrics — but you need to measure them correctly.
The number that matters: ad cost per unit shipped
Compute ad spend per unit shipped: take total ad spend on a channel in a period and divide by units sold through that channel in the same period. This is your real customer acquisition cost per unit. It is not glamorous, it does not have a clever acronym, and it does not impress investors. It is, however, the number that tells you whether your advertising is actually profitable at the unit level.
Plug this CAC per unit into the NetSellerProfit calculator as your ad spend input. If the channel net profit per unit is positive after your real CAC, the channel works. If it is only positive when you use the platform-reported attributed revenue number (which in most cases is higher than reality for Meta and TikTok), the channel is losing you money and the dashboard is hiding it.
The iOS attribution discount
Meta, TikTok, and Snapchat all over-report attributed revenue relative to independently measured revenue post-iOS 14.5. The magnitude varies by brand, category, and audience composition (iOS vs Android mix), but a 20–35% haircut on platform-reported attributed revenue is a reasonable conservative assumption for most direct-to-consumer brands in 2026.
Apply a 25% haircut to platform-reported attributed revenue until you have data to do better.
Until you have incrementality test data specific to your business — from a geo holdout test, a conversion lift study, or a media mix model — the most defensible approach is to apply a conservative discount to all platform-reported numbers and make decisions on the discounted figure. This will cause you to scale back channels that are marginally profitable on paper. That is the correct outcome. Marginally profitable on paper is often loss-making in reality.
Post-purchase surveys as ground truth
The most underused attribution tool available to e-commerce operators is a post-purchase survey with a single question: "How did you first hear about us?" The answers — which you collect from real buyers who have already converted — are not subject to pixel loss, browser restrictions, or attribution window disputes. They are, however, subject to recall bias and are skewed toward the most memorable touchpoints rather than the most recent ones.
Run post-purchase surveys continuously and compare the channel mix they report against your platform-reported attribution. The gap between the two is your attribution distortion coefficient. Use it to adjust your media mix model and bidding decisions. Over six to twelve months, this data becomes one of the most valuable assets in your marketing operation.