What a budget ladder does to incremental ROAS
Step a campaign budget up in stages and reported ROAS can hold steady while incremental ROAS quietly falls. The gap between the two numbers is where wasted spend accumulates.
Three platforms, three answers, one budget to defend. Your GA4 export already holds the answer, and the first finding is free and never leaves your browser. Get your first finding free
By Joris van Huët, Founder & CEOUpdated 3 min read
Run the numbers for your store: the free break-even ROAS calculator.
Reported ROAS can stay flat across four budget steps while incremental ROAS falls the whole way. Both numbers are real. They are answers to different questions, and only one of them is about your budget.
Why the ladder hides its own cost
A budget ladder steps spend up in stages, holding at each level until the reported return looks stable, then stepping again. It is a sensible way to scale and a poor way to learn.
At a low budget, a platform spends on the people easiest to convert. Many of them were going to buy. The platform sees the conversion, claims it, and reports a high return.
Raise the budget and the platform has to reach further. The newly reached audience converts at a lower rate, but the platform is still claiming the easy conversions underneath. The average holds. The margin does not.
Compare what each channel reports against what the total tells you, on the same window, and start with the channels where the two disagree most.
| Signal | What it suggests |
|---|---|
| Reported ROAS flat as budget rises | The platform is re-claiming conversions it already had |
| Total revenue rising slower than spend | The new spend is buying less than the old spend |
| A channel's reported revenue exceeding total revenue when summed across channels | Overlapping claims on the same orders |
The third line is the one people find first and dismiss. If every platform's self reported revenue is added together and the sum is larger than the store's actual revenue, at least one platform is counting an order somebody else also counted. That is not a bug. It is what platform attribution overcounting does by design.
The step that actually answers it
The ladder has a free experiment built into it that most teams throw away: the moment before each step and the moment after. If nothing else changed in that window, the difference in total revenue against the difference in spend is a crude read on the marginal return of that step.
Crude, but honest, and it needs no new tooling. What ruins it is changing three things at once, which is what a scaling week usually looks like.
What a causal read adds
The ladder read gives you one number per step, with no interval and no way to separate the step from the weather.
A channel that cannot be resolved should be reported as unresolved, not given a number that looks like the others.
A read costs 99 euro once and is refundable within 30 days, no questions asked.
Related answers
Frequently asked questions
What's the fastest way to identify and cut wasted ad spend across all my marketing channels?
Compare each channel's self reported revenue against the store's actual total on the same window. Where the summed platform claims exceed real revenue, at least one platform is double counting. Start cutting where the disagreement is largest, then confirm with a holdout before the cut becomes permanent.Why does reported ROAS stay flat when I raise the budget?
Because the platform keeps claiming the easy conversions it was already getting while the newly reached audience converts worse. The average of a strong base and a weak margin can look unchanged even as the marginal return falls.How do I know if the next budget step is worth taking?
Read the change in total store revenue against the change in spend across the step, with nothing else changed in that window. It is crude and it is honest. Changing creative, audience, and budget in the same week destroys the comparison.
Go deeper: Causal attribution, explained.
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
Have an audience of ecommerce operators? Earn 20% recurring commission by recommending the €99 read.
Keep reading
Terms in this article
- AttributionAttribution identifies user actions that contribute to a desired outcome and assigns value to each. It reveals which marketing touchpoints drive conversions.
- ConversionConversion is a specific, desired action a user takes in response to a marketing message, such as a purchase or a sign-up.
- RevenueRevenue is the total income generated by the sale of goods or services related to a company's primary operations.
- ShopifyShopify is an ecommerce platform for creating online stores and selling products. Attribution modeling shows which marketing channels drive traffic and conversions within Shopify.