Budget optimiser is a tool that provides easy access to insights around the best way to allocate marketing budget. In a nutshell, It works by looking at historical data, figuring out the estimated return on investments at different levels of ad spends, and then suggesting the distribution of ad spend that gives the most return on investments for the given period.
Before starting to use Budget optimiser, it is important to consider several things:
1. Historical data analysis
It’s important to know that Budget optimiser is looking at historical data and is not predicting future performance. Therefore Budget optimiser results should be used as a validation and insights tool about the past for guiding the future.
For example, if you’re looking at the results for September and October, and it’s saying you should have spent 25% more on a Facebook campaign, it would be wise to remember that November is different, and as such you probably shouldn’t follow the recommendation exactly, but use it as a recommendation to spend more on Facebook than you thought you should have previously.
2. Picking the right attribution model
Keep in mind that the entire ‘return on investment’ logic is based on the attribution model that determines the output of different campaigns.
For example, the output of Budget optimiser would differ drastically if you use Alvie regression-based attribution model as compared to the Google Analytics 4 Last non-direct click attribution model.
When working with the Budget optimiser, the output is only as good as the input. In the case of a budget allocation algorithm like this, it’s important to be able to trust what is deemed as the ‘return’ for a given channel.
For example, if the input data says that €5 was spent on Google Ads and €100 was earned, there’s a very good chance that the model would recommend spending more there. The 'earned' part of that sentence is what is important here. What was earned becomes clear by attributing some value to the channel. That is in essence why attribution models are important. The output Budget optimiser would be very different if the used attribution model would say that Google Ads earned €20 in the above example.
To accommodate different situations, Budget optimiser allows the use of both Market attribution models and Alvie regression-based attribution model as input for the calculation:
When you create an Attribution model configuration within Alvie, the Budget optimiser configuration will be automatically created and will utilize this attribution model as the input.
If you want to run Budget optimiser using one of the Market attribution models, simply create a Budget optimiser configuration and select the required Market attribution model as explained in Step 1.
We recommend taking the time to understand the specifics of all available models, including Market attribution models and Alvie regression-based attribution model, in order to choose the model that best suits your business case.