Modeled conversions use data that does not identify individual users to estimate conversions that Google is unable to observe directly. This can offer a more complete report of your conversions.
When cookies aren’t observable, it may not be possible for you to measure some conversions. This can happen when there are cookie restrictions in browsers or blind spots from cross-device conversions, leading to industry-wide gaps in observability.
Modeled conversions use machine learning to quantify the impact of marketing efforts when a subset of conversions can’t be observed.
How it works
Google Ads currently models website conversions that should be attributed to clicks and views, but may go unobserved due to various limitations. Through modeled conversions, these unobserved clicks and views are added to your account’s reports, along with the observed conversions.
If conversions can’t be observed due to partial or missing conversion data, Google models conversions by:
- Identifying portions of traffic with known observed conversions.
- Applying statistical techniques to portions of traffic where data is missing or partial.
Modeled conversions use current observable signals such as device type, date and time, and conversion type to create an accurate view of user behavior.
Google’s approach is different from fingerprinting technologies, which typically rely on signals such as IP addresses that identify users across various touch-points and generate a “fingerprint ID” to identify the user across future interactions. We are not generating such IDs or attempting to identify individual users. Instead, Google is aggregating non-sensitive data such as historical conversion rates, device type, and time of day to predict the likelihood of conversion events across the set of users who viewed or clicked on an ad.
Benefits of modeled conversions
- Holistic measurement across all your ads traffic: Gain a more accurate picture of your advertising outcomes, and a complete picture of the conversion path.
- Efficient campaign optimization: Modeled conversions help you push your campaign more effectively.
- More effective omni-channel marketing by understanding how users convert across multiple devices.
- For example, when a user starts their journey on one device and completes the conversion on another, it may not be possible to attribute the conversion. Google observes data from the large number of signed-in users on Google properties to extrapolate similar behavior from signed-out users.
- Accurate conversion measurement while only reporting on data that does not identify individual users. This privacy-centric approach provides you with a holistic view of your users' conversions, ensuring that your measurement isn’t impacted when conversions are not directly observable.
Tip: To ensure the most accurate conversion modeling for your account, implement the global site tag or Google Tag Manager. This ensures that we get as many observable conversions as possible in order to apply our statistical modeling techniques. The best way to measure online conversions is to use tagging tools that set cookies in the same domain as your website (known as first party cookies). For App measurement, we recommend using the latest version of Google Analytics for Firebase.
How to view your modeled conversions data
Google will include modeled conversions in the "Conversions" column in your reports. The “Conversions” column shows you the number of conversions you've received, across your conversion actions and may include modeled conversions in cases where you are not able to observe all conversions that took place.