Google Analytics

Ads + Attribution Models Google is Sunsetting in 2024

Sunset of the Standard Attribution Models, Alternative Attribution Models, + Getting Familiar with Data-Driven Attribution

Main Image Courtesy of Google’s Blog.

Attribution Models

An attribution model is a framework that assigns credit to various elements of a customer's journey in relation to a marketing campaign or channel. It helps determine how effective a particular campaign or channel is.

An attribution model can help businesses improve their marketing efforts and allocate resources more efficiently. It can also help them identify which elements of their customer's journey are most effective at driving conversions.

Campaign effectiveness can be measured using an attribution model. It enables businesses to identify successful tactics and channels.

Through an analysis of customer touchpoints, businesses can gain a deeper understanding of their customers' interactions with brands and improve their overall experience. Identifying which marketing channels and strategies are performing well can help businesses maximize their marketing budget.

Through the use of an attribution model, businesses can get a deeper understanding of their marketing performance. It can also help them make more informed decisions and improve their marketing results.

Earlier this year, Google explained that it would no longer support four attribution models in its Google Analytics and Google Ads platforms.

  • First click: The First Click attribution model assigns 100% of a customer's first touchpoint credit to the first stage of the customer journey.
  • Linear: The Linear model distributes credit evenly across all of the customer's touchpoints, ensuring that each interaction is highly impactful.
  • Time Decay: The Time Decay model distributes credit more evenly across the customer's various touchpoints, with fewer credit given to the earlier interactions.
  • Position-Based: The position-based model allots a majority of the credit to the initial and last touchpoints of the customer's journey, with the balance being distributed evenly across other interactions.

Sunset of Attribution Models

As the company has gotten rid of its previous four main attribution models from its framework, Google will now only support the last-click model, which provides credit to the customer's last touch point during the conversion process. Although this feature will remain available, companies will have to manually adopt it.

According to Google, the decision to phase out the rules-based models was due to the low adoption rate. It also noted that the majority of these models are not used to attribute Google Ads web conversions.

DDA graphic
Machine learning and the dynamic display ad format (DDA) are more indicative of the actual state of ecommerce websites. Image Courtesy of PPCexpo.

Most of the time, rules-based models undervalue certain elements while overvaluing others. For instance, first-click undervalues the various aspects of a customer's journey while overvaluing the last touch point.

Why are Attribution Models Changing?

Google is phasing out many of its ad models because they do not give advertisers enough credit. This is an issue that affects both the company and its users.

Lack of Accuracy 

For a long time, Google Analytics has used the Last-Click model as the default attribution method. But, this method doesn't take into account other factors that contributed to a conversion. The First-Click model, on the other hand, only gives credit to the first contact point. This method could not accurately reflect a customer's journey.

Although the Time Decay and Linear models provided more of a balanced approach to capturing the customer journey, they still fall short in capturing all of the elements of it. This would often lead to poor marketing decisions and inaccurate data.

Customer Behavior 

Over the past few years, the behavior of customers has changed significantly due to the various touchpoints that they have used to interact with businesses, such as social media, email, and search engines. For instance, if a customer found a business through a social media platform, then they would have been credited for the purchase even though they also engaged with the company through email.

Cross-Device Tracking

Due to the increasing number of devices that people use, it has become more challenging to track the whole customer journey. Traditional models only track activity on a single device, which can lead to poor marketing decisions and inaccuracies.

Real Time Data is Necessary

Due to the limitations of traditional attribution models, which rely on historical data, they are unable to provide businesses with the necessary insight into their customers' journey. With the help of ML algorithms, the GA4 model can provide businesses with real-time insight into their marketing spend.

3 Alternative Attribution Models

Google Ads has various attribution models that can help you track the effectiveness of your campaigns. Take a look at the remaining models to learn more about their capabilities!

First Click Attribution

The first click attribution model takes into account the ad or keyword that led to your website and converts it into a credit. This method helps marketers understand how their ads can generate interest and entice new customers.

Last Click Attribution

The Last-click attribution model provides credit for conversions that led to specific keywords or ads. This model helps marketers identify which ads can influence the most immediate actions, like signing up or making a purchase.

Data-Driven Attribution

A machine learning-based model known as a data-driven attribution is utilized in Google ads to determine and attribute conversions to various points in a customer's journey. It takes into account each marketing touchpoint's contribution to driving conversions based on historical data.

Data-driven attribution takes into account the distinct interactions between your ads and customers, which makes it different from other models that rely on predefined guidelines when it comes to distributing credit. By evaluating the impact of each touchpoint, the model can provide a more accurate representation of how your marketing platforms perform, allowing you to make more informed decisions regarding campaign optimization and budget allocation.

Issues Associated with Data-Driven Attribution

Due to the emergence of new obstacles, the data attribution models used to collect and interpret data have become less reliable and inaccurate. These models tend to overemphasize the user experience or assume that every touchpoint is relevant, which can result in undervaluing each interaction.

The concept of first click is considered an outdated model when it comes to assessing the various digital channels that users can interact with. For instance, they may be engaging with your brand through social media, video platforms, and smart TVs before they even visit your website.

Traditional attribution models are not able to properly reflect the complexity of the user journey due to the fragmentation of attention across various platforms.

The complexity of the user journey is also exacerbated by the fact that most people use different devices when accessing the Internet. For instance, a person can browse social media using their phone, look up a blog post, or watch a YouTube video on their TV. Each of these platforms may have ads that are related to the brand they are currently engaging with.

Data attribution models should also be able to take into account the various steps in the user's journey to determine the value that they are getting from their interaction. Image Courtesy of Google Analytics Blog

The increasing complexity of the user journey has caused many people to become more cautious when it comes to interacting with online platforms. They are more prone to doing their research and opting out of sharing information whenever possible. This can result in data attribution issues since models that rely on last-click don't take into account these changes.

In 2024, Google plans to phase out the use of cookies in its analytics tools. Most of the attribution models that were previously built using Google Analytics rely on the cookie for their measurement. As a result, the company is currently developing new systems that do not need cookies to perform attribution.

Privacy Regulations 

A privacy regulator's effect on a data-driven attribution model is significant. The General Data Protection Regulation (GDPR) and other cross-border legislation ensure that consumers are protected from unauthorized access and use of their data. 

In Canada, the Personal Information Protection and Electronic Documents Act (PIPEDA) also provides consumers with a level of protection regarding their online presence and data. Image Courtesy of TermsFeed

Although they do not explicitly prohibit the use of data-driven attribution, privacy regulators require companies to take the necessary steps to ensure that their practices are aligned with the principles of personal privacy. In most cases, implementing regulations will reduce the amount of information that is available to inform the model.

Future of Data-Driven Attribution

Machine learning will eventually replace the traditional attribution model in ecommerce. Many companies will start using this as their default model.

Through data-driven modeling, brands can now gain a deeper understanding of their marketing campaigns' real impact on their sites. This method eliminates the need to rely on inaccurate information and provides them with a more realistic view of what's happening in their sites. As a result, any disruptions in the current practices of ecommerce companies should be considered a sign that the era of traditional models is ending. Machine learning and DDA are pushing the envelope in terms of their capabilities.

One of the main advantages of using data-driven modeling is that it eliminates the need for brands to rely on a traditional model, which is why Google has forced many companies to use the DDA framework. Unfortunately, this method also exposes the downside of relying solely on Amazon and Google for marketing measurement. If you rely on these two companies for your measurement, they will decide how you view their data.

One of the biggest risks that brands face when it comes to adopting new technology is the loss of control over their attribution models. With the inclusion of first-click in this list, they are now giving up control over their data to third-party services such as Google Analytics.

Ecommerce Companies Must Keep Focus

Due to the widespread use of Google Ads and Google Analytics, many companies will be affected by the removal of the attribution models that are based on rules.

Before implementing the changes, it’s important that ecommerce companies thoroughly analyze the impact they will have on their ability to measure the effectiveness of their marketing channels and identify the areas where they will lose. 

In addition, it’s also important that ecommerce companies confirm that their ad interactions and conversions meet Google's minimum requirements to be eligible for the DDA. If not, they'll have to use a last-click model with no access to multi-touch capabilities.

To ensure that they are still able to implement the changes, brands should consider the various options available to them when it comes to adopting a multi-touch attribution model. Although it may be challenging to lose some of the models that are currently used, independent platforms can help them fill the gap. The future of attribution is data-driven and, as a result, companies that adopt this approach will be better positioned to adapt than those that continue to use outdated models.

Get Familiar with the Change in Attribution Models Today

Get the most out of your marketing budget by switching to data-driven attribution! This approach will allow you to get more precise and accurate information about how your campaigns are affecting conversions, and where your marketing budget stands at all times.

Get great insight from our expert team.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
By signing up you agree to our Terms & Conditions