1. Ad Analytics Help Center
  2. About
  3. About Attribution Methodology

About Modeled vs. Unmodeled Attribution

Spotify Ad Analytics models attribution results in order to provide a more comprehensive view of the impact of your campaign.

Before you dive in:

It's important to understand what data Spotify Ad Analytics is able to use for attribution and how we analyze attribution using a device graph

What are modeled results vs. unmodeled results? 

Using a device graph, Spotify Ad Analytics confirms the impressions in a standard attribution window based on confirmed household IP addresses. Additionally, IP addresses that are not ‘noisy’ (i.e. noisy IP addresses are those using a cell-tower or commercial connection) are analyzed in order to determine if they should be considered a household or not. When Spotify Ad Analytics determines impressions are not from a household, results are subsequently modeled so that your campaign's results provide additional insight into the effectiveness of its full reach and impact.

Screenshot 2023-05-19 at 2.45.50 PM

How does it work? 

  1. Based on the total IP space captured through impressions/downloads, we project a multiple of additional conversions that would have also likely happened on the advertiser’s website had we been able to run standard attribution across all impressions.
  2. The multiplier will vary for each line-item as it is based on the amount of non-noisy IP data we can use for attribution and the amount of noisy IP data discarded. 
  3. The multiplier is calculated by taking the unique number of total IP addresses and dividing it by the unique number of household IP addresses.