Achieving high KPIs like ROAS traditionally looked good on   marketing  performance reports, but more and more companies (such as Uber and Airbnb) have discovered that much of their spending had no impact. Because of this, today we will explore one key topic in marketing measurement: Incremental ROAS.

This article will answer the following questions:
What is incremental ROAS?
What is the difference between ROAS and incremental ROAS?
Does ROAS bring ROI (Return on investment)?
How to calculate ROAS vs incremental ROAS?
How brands measure incremental ROAS?
How are we measuring incrementality?
 

What is Incremental ROAS?

Incremental ROAS (iROAS) measures the additional revenue generated from your advertising efforts beyond what would have occurred without the ads. In simpler terms, incremental ROAS measures the true contribution of your ads to your overall revenue.

The difference between ROAS and Incremental ROAS

Unlike traditional ROAS, which simply calculates the return on the total ad spend, incremental ROAS focuses on the extra lift (contribution) produced by the ads.
Think of it like this: standard ROAS tells you how much revenue you earned for every dollar spent on ads, but it doesn't account for how much of that revenue would have been obtained anyway. Incremental ROAS addresses this by focusing solely on the true contribution (revenue) generated by your ad campaign.
Incremental ROAS is also referred to as marginal ROAS.
 

ROAS vs incremental ROAS - simple example

The difference between incrementality and attribution is often exposed when launching a new channel:

 Why is this happening?

In this case, launching a new channel, such as a paid media channel, cannibalizes the organic results that would have occurred anyway even without the customer seeing the paid ad. However, using last-touch attribution, sales will be attributed to the last channel that interacted with the user, even though the user would have come organically anyway.

Does ROAS bring ROI (Return on investment)?

Because ROAS considers the results reported by the channel or by third-party attribution, ROAS is biased, missing the obvious question:

• Are these results incremental or redundant?
• Do the results actually add value that we were not getting already?
   

The most sophisticated advertisers were duped to rely on ROAS, only to discover that a positive ROAS could still lead to a negative ROI. Because of this, any marketer should think about incremental ROAS (ROI) and not ROAS.
The unfortunate way most companies figure this out is either when the budget “runs out”, or by intentionally reducing ad spend to measure the incremental value of ad spend, as seen in the example below:

(This is an example from our budgeting handbook, which also presents budgeting principles and strategies for improving incremental results. You can find it here.)

Common causes of non-incremental returns

• Cannibalization: This is the number one reason for advertising budget waste.
• Last touch attribution: This is a common reason for over attribution. It’s similar to attributing the victory only to the players that scored a goal in a football game.
• External factors: The interplay of various factors that influence sales, such as the external factors, seasonality and market changes.
• The effect of diminishing returns

ROAS calculation

Let’s shortly look at how ROAS is calculated first. Here is the formula:

For example, if you spend $10,000 on ads and generate $40,000 in revenue, your ROAS is 4. This indicates that for every 1 dollar spent, you earned 4 dollars back.
However, ROAS assumes that the ad campaign is solely responsible for all revenues.
 

Incremental ROAS Calculation (iROAS)

Incremental ROAS (iROAS), on the other hand, isolates the impact of your advertising efforts by measuring the additional revenue generated solely due to the ads. Here is the standard formula:

True incrementality is measured as total ROI - Return Over Investment, ensuring that advertising activities are producing additional value. In other words, incrementality helps you identify areas where you may be overspending and allows you to adjust your spend accordingly.

Here is a simple example:
Imagine you invest $50,000 in an ad campaign promoting a new product launch, and your reporting shows that the campaign generated $150,000 in revenue. A quick ROAS calculation (150,000 / 50,000) gives you a ROAS of 3.0, suggesting a good return on investment.
However, as a savvy marketer, you know there’s more to the story. When you examine your overall sales data, you find that your total sales for the period only increased by $70,000. This reveals a significant discrepancy.
This discrepancy highlights the importance of iROAS in understanding the true impact of your advertising. While the initial ROAS of 3.0 might seem like a success, the iROAS of 1.4 provides a more modest, yet accurate, picture of your campaign’s effectiveness.
Moreover, this discrepancy underscores a common issue: your campaign may be taking credit for sales that would have occurred anyway, thereby inflating the perceived ROAS. Ad platforms are often incentivized to present their performance in the best possible light, which can lead to over-attribution.
 

Calculating Incremental ROAS for multiple channels or campaigns

Calculating iROAS correctly using a simple formula works only if "all else is unchanged" (the ceteris paribus assumption from economics), which is rarely the case at a larger scale, such as when you have multiple channels in your marketing mix. In such situations, it is not sufficient to look at just positives and negatives to understand incremental revenue.

How brands calculate incremental ROAS

A very popular, though limited, method is Media Mix Modeling (MMM). We have written several articles about MMM on our website. Here is a summary:

This method was developed in the 1950s, at a time when traditional media channels like TV, radio, and print advertising were predominant. Back then, larger brands needed to analyze the impact of different campaigns and marketing changes on their sales performance. Today, Media Mix Modelling it’s still a good tool (depending on the brand and the historical data) for scenario planning, allowing brands to simulate various marketing strategies and forecast their potential outcomes.

However, MMM falls short in execution and practical application. Over the last three years, driven by changes such as the introduction of ATT (App Tracking Transparency) and the deprecation of the IDFA, hundreds of advertisers have tested MMM. Unfortunately, many found the results to be unsatisfactory and confusing, and did not complete the MMM calibration process.

If you want to read more about MMM, here are our 2 articles:

What is Media Mix Modeling? 
The Rise and Fall of Media Mix Modeling

At INCRMNTAL, we recognize the limitations of MMM but we are not anti-MMM. Our method is different and has been successfully adopted by hundreds of advertisers. While our platform is often compared to MMM, there is no straightforward bucket to place INCRMNTAL in.
 

Our approach: Causal AI incrementality measurement

After 20+ years of adtech experience, our founders saw how limited marketing measurement was to basic tracking between clicks and installs, or clicks and conversions. On top of that, in 2019,  AirBNB & UBER found that +80% of their spend was redundant.

Our vision was that true measurement needs to focus on measuring the additional value derived from marketing investments. Solving this was a significant technological challenge, especially with user tracking set to be deprecated. INCRMNTAL was going to offer a solution.

Our original idea was to build the most advanced multi-touch attribution solution, using a clean room shared with advertisers and publishers and their first  party data. 

We shifted our focus to incrementality very quickly, realizing that we did not want to build better attribution, but solve the measurement problem.

Today, we are using our own causal AI incrementality measurement to help lots of brands measure the true value of their marketing efforts and get actionable insights for them. Here is a summary of what INCRMNTAL does:

• Measuring the incrementality of any marketing channel or campaign, offline or online
• Analyzing both marginal and absolute values.
• Finding the optimal spend level before passing the point of diminishing returns
• Measuring the impact of external events out of your control: seasonality, weather, trends and more
• Measuring advertising cannibalization (For example: Based on  our measurements in some cases branded campaigns can be incremental)
  
   

All these with easy integration, no SDK/code needed and 100+ integrations with ad platforms, data warehouses, MMPs, Analytics platforms, and TV & CTV, measurement taking only 2 minutes.

Here is a measurement example from one campaign using INCRMNTAL. This reported:

• Incremental conversions
• CPI and Marginal CPI
• ROAS and marginal ROAS
• And also recommended which channels should be scaled up and which ones scaled down.
  
   

In the end, understanding the differences and recognizing common causes of non-incrementality such as cannibalization or last touch attribution is crucial. Focusing on incrementality ensures that marketing efforts truly enhance ROI, driving real growth and more effective resource allocation.
 

You can find more about our platform by scheduling a live demo with us to see how INCRMNTAL can enhance your marketing measurement.