Marginal Incremental ROAS (MIROAS): What is it? And why does it matter to marketers?
Imagine this scenario: You are managing Meta campaigns of an eCommerce business, and have just been given more budget. You need to decide how to invest it. Luckily, you have recently transitioned from Last-Click attribution to measuring Incremental ROAS (iROAS). Here's what your are seeing:
- Ad Set A: 5.0 iROAS
- Ad Set B: 6.0 iROAS
Which one should you spend more on? Your gut tells you to invest more in B, because it has higher iROAS. This could be a smart move, but it could also be costly mistake.
Why? Because you are looking at the wrong KPI.
All spend optimization decisions should be made based on Marginal Incremental ROAS (MIROAS), not based on iROAS.
What is Marginal Incremental ROAS (MIROAS)?
Marginal Incremental Return on Advertising Spend, or MIROAS, tells you the return for the next invested dollar. In other words, it answers the question: If an additional dollar is invested into the channel, campaign or ad set, how much additional revenue will it generate?
MIROAS is always dependent on the current spend level for the specific marketing activity:
👉 When your spend level is low, you typically have lots of room scale, and MIROAS can be high. High MIROAS means that the next additional dollar drives high amount of additional revenue.
👉 When your spend on an activity grows, you will reach a point where the channel starts to saturate, and MIROAS starts to decrease. Low MIROAS mean that the next additional dollar drives little additional revenue.
How is MIROAS Different from iROAS?
The distinction between MIROAS and iROAS is critical, because making spend optimization decisions based on iROAS is as good as flipping the coin.
iROAS provides you the average return your investments have generated within a certain historical timeframe. It's a backward-looking metric that evaluates your average performance across different spend levels and historical events.
iROAS cannot be used in spend optimization, because it does not provide the return for the next additional dollar spent on the channel/campaign/ad set.
Why is Marginal Incremental ROAS (MIROAS) So Critical?
MIROAS is the only KPI that provides the expected the return for the next invested dollar. By comparing MIROAS across all of your channels, campaigns and ad sets, you can identify marketing activities which deserve more budget, and ones which you can scale to a lower level without losing much revenue.
Let's discuss this in practice by getting back to the example that started this blog. Before choosing where to spend the additional budget you were given, you have decided to take Sellforte into use. This was a smart decision, because, Sellforte provides you the following data on the two Ad Sets:
The data tells you that while Ad Set B has had high average iROAS (6.0), its MIROAS is only 0.5. This means that investing one dollar more in Ad Set B delivers only $0.5 of incremental sales.
At the same time, Ad Set A has slightly lower average iROAS (5.0), but still has MIROAS of 3.0. This means that the next dollar spent on Ad Set A delivers $3.0 of incremental sales.
👉 The obvious conclusion is to invest more in Ad Set A, because it has higher MIROAS.
This example illustrates a key challenge in media spend optimization: High iROAS does not necessarily mean that you should increase your spend. It could be that you have lots of room to scale (high MIROAS), but it could also be that you are well past the saturation point (low MIROAS).
How is Marginal Incremental ROAS (MIROAS) Calculated?
MIROAS can be calculated as a derivative of the response curve for the channel, campaign or ad set.
💡Marginal Incremental ROAS (MIROAS) Formula:
MIROAS = Δ Revenue / Δ Spend
where
Δ Revenue = Incremental revenue driven by an additional unit of spend
Δ Spend = Additional unit of spend
The image below illustrates this visually. In the visual representation, it's also clear how MIROAS is a function of your current spend level. In this image response curve, MIROAS is very high at low spend levels, but decreases as spend level grows.
Response curves, in turn, are estimated by Marketing Mix Modeling tools.
Which Tool Should I Use for Measuring MIROAS?
To get most out of your marketing budget, you should use a tool that fulfills two criteria.
1. You should choose a tool that provides MIROAS for each campaign and ad set. This enables you to transition from unreliable iROAS-based optimization to MIROAS-based optimization on a very granular level.
2. You should choose a tool that translates MIROAS-based optimization recommendations automatically to bidding recommendations for each campaign and ad set. This saves hours of manual work from your team, trying to understand how certain budget change recommendations are translated into Target ROAS values for a specific campaign or Ad Set.
There is only one tool in the market currently fullfil the two criteria above: Sellforte. Below is a screenshot from Sellforte Performance, which provides bid value recommendation for each campaign and ad set, based on MIROAS.
Other alternatives tools include:
- MMM tools that provide iROAS on campaign-level, but don't provide MIROAS: You cannot optimize campaign spend with them due to lack of MIROAS.
- Traditional MMM tools that provide MIROAS on channel-level: You cannot optimize campaign performance with these tools due to lack of campaign level MIROAS.
Ready to get started?
If you’d like to start MIROAS-based optimization, book a demo with us today.
Authors
Lauri Potka is the Chief Operating Officer at Sellforte, with over 15 years of experience in Marketing Mix Modeling, marketing measurement, and media spend optimization. Before joining Sellforte, he worked as a management consultant at the Boston Consulting Group, advising some of the world’s largest advertisers on data-driven marketing optimization. Follow Lauri in LinkedIn, where he is one of the leading voices in MMM and marketing measurement.
You May Also Like
These Related Stories
What is Causal Attribution in Marketing?

Marketing budgeting season is here with five best practices to follow.
