Pmax automated: Structuring campaigns for maximum ROAS performance

Why “ROAS” could be the best strategy to go with

The goal of a ROAS-based campaign is simple: allocate more budget to high-performing products and reduce or eliminate spend on those that underperform. However, while this approach is effective, it does come with some potential drawbacks.

Benefits

• Focuses on top-performing products: Maximizes ad potential by prioritizing products that consistently deliver high revenue with lower costs.
• Reactive structure: Instead of making speculative decisions, it relies on historical data, making it a more accurate, data-driven strategy.
• Customizable timeframes: Can be built on performance data from custom periods, such as 30, 60, or 90 days, to fit your campaign needs.
• Scalable: As performance data evolves, this flexible structure allows you to adjust ROAS targets or budget allocation according to new trends or seasonal changes.

Downsides

• Data dependency: If you previously used only a single Performance Max campaign and many products were overlooked, there may not be enough historical data to effectively set up this structure.
• Requires active monitoring: ROAS performance will fluctuate over time, so close monitoring and adjustments to rules are necessary for campaigns to adapt quickly.

TIP: To keep each product at least somewhat profitable, you can use an automation tool like Dotidot to set a minimum ROAS. This minimum ROAS acts as a limit, making sure a product doesn’t go below a certain profitability level.

How structuring by ROAS look like

When segmenting campaigns by ROAS, the structure should reflect the performance range of your products and align with your overall marketing goals. Based on conversion volume and budget, here are three primary segmentation approaches:

2-Campaign Structure:

• High ROAS campaign: For products that consistently deliver strong returns, set a higher ROAS target. This ensures that these products get the majority of the spend.
• Low ROAS campaign: Allocate a smaller budget for products that have historically struggled. Set a lower ROAS target to capture any potential conversions without cutting into your profitable campaigns.

3-Campaign Structure:

• High ROAS campaign: Prioritize top-performing products with a high ROAS target.
• Medium ROAS campaign: Include products that perform moderately well. With a balanced ROAS target, this middle tier allows for better experimentation and visibility.
• Low ROAS campaign: Keep low-performing products here, with a budget and ROAS target that enables visibility without overspending.

4-Campaign Structure with “Zombies” Campaign:

• High, Medium, and Low ROAS Campaigns: Structure as above to maximize budget control and reach based on performance.
• Zombies Campaign: For products that are underperforming or not performing at all, set this campaign to "Maximize Conversion Value" with no specific ROAS target. By allocating a small daily budget, you give these products a chance to gain visibility and potentially rebound without impacting the main campaigns.

TIP: For more information about the “zombie” campaign strategy, read our article that covers this strategy in detail.

Revenue from GAds or GA4?

The choice depends on your business goals and how your attribution modeling or reporting aligns with them. Keep in mind that tracking and channel attribution have limitations. You have these options:

• Google Ads revenue: This focuses entirely on maximizing the potential of your product within Google channels. However, it may overlook the product's influence on overall performance.
• Google, Meta, and other advertising channels: This considers the cumulative performance of your product across all advertising channels.
• Google Analytics 4 revenue: This accounts for the overall revenue of your product, regardless of channel attribution.

For our example, we will use Google Analytics 4 revenue.

Setup in Dotidot

STEP 1 - import additional product data

To build a ROAS-based PMAX structure, you first need to calculate the metric, as it cannot be directly imported from Google Ads. Instead, you will need to:

1. Import Revenue from Google Analytics
2. , it will automatically create a variable called _ga4_last_30d_item_revenue_
3. Import costs from Google Ads, it will automatically create a variable called _ads_last_30d_price_

Both metrics can be imported using various time ranges (e.g., 30, 60, or 90 days).

Tip: For the most accurate "bucketing," it is recommended to use a 30-day range. However, if you have a low volume of conversions, consider increasing the time range.

STEP 2 - calculate the ROAS

Next, you need to create a new variable that calculates ROAS from the additional data. Here it depends if you prefer a simple ROAS coefficient or ROAS in %.

Basic ROAS formula: Revenue / cost = ROAS
ROAS as a percentage: (Revenue / cost)*100 = ROAS %

Let’s use ROAS % since it’s the preferred format for most marketers. 

With this in mind, create ROAS % variable:

1. Navigate to your data source
2. Go to Variables section and click “+”
3. Choose Numeric variable
4. Name it appropriately so you can easily identify it later in your data source. We’ll use “ROAS_percentage”
5. In the Input field, replicate the formula above using data from your source
   (See the image below for how it should look)

STEP 3 - segment your products

Depending on the ROAS ranges and the number of conversions, divide your products into different buckets. Each bucket should have at least 30 conversions (though 50+ is recommended) over the selected time period. For example, let’s create three buckets:

• For products with a ROAS below 300%, create a product set called "Low ROAS."
• For products with a ROAS between 301% and 499%, assign them to the "Medium ROAS" product set.
• For products with a ROAS of 500% or higher, create a bucket called "High ROAS."

To set this up in Dotidot, follow these steps:

1. Navigate to or stay in your data source
2. Go to the Product Sets section and click "+" to create a new set
3. Name the set (e.g., "High ROAS" or any name you prefer)
4. Select the variable containing the calculated ROAS value
5. Set the appropriate condition. For instance, to include products with a ROAS of 500% or higher, select "greater than or equal to" and input the value 500

Now, repeat this process for all three buckets, changing only the name and values in the final step.

STEP 4 - Build the campaigns

The final step is the simplest: build your campaigns based on profit margin groups. Set the target ROAS for each campaign using these guidelines:

• High ROAS campaign: Set a target ROAS of at least 500%, or higher if aiming for more ambitious goals. Allocate the largest budget to this campaign.

• Medium ROAS campaign: Set a target ROAS between 301% and 499%. Aim for the lower end of the range for more conversions or the higher end for better efficiency. Assign a moderate budget.

• Low ROAS campaign: Set a target ROAS of 300%. The goal is to avoid significant losses while encouraging the algorithm to improve performance and move items out of this bracket.

Finally, product selection is a critical part of the setup—ensure that you include the corresponding product sets for each campaign.

‍