Data-Driven Attribution for Google Search

The following is one of my favorite Harry S. Truman quotes, and for some, words to live by.

However, in search advertising, when it comes to maximizing ROAS (Return On Ad Spend), knowing which ads to credit for conversions is critical. That’s where using data-drive attribution for Google Search comes in.

This post will focus on using data-driven attribution for Google Search. To get started, check out the following short video from Google:

Determining the Purchase Contribution of Each Google Search

Data-Driven Attribution for Google search- pathway to decision — Data-Driven Attribution for Google search- **pathway to decision**

Businesses grow and expand online primarily by driving relevant traffic to their site. The ultimate end goal for most ad campaigns is to get a user to follow a link to your website, and to make a purchase or perform some other valuable action for your business. In this sense, digital marketing is all about leaving a trail of breadcrumbs for your customers to follow.

But how is an advertiser to know which ad clicks are the most relevant? This question is more important now than ever before, as the challenge of optimizing increasingly long customer journeys to a conversion looms. For example, 90% of web users are likely to switch between different devices multiple times throughout the day. In addition, customers typically make more than one search before making a purchase, to ensure that what they’re seeing is absolutely what they want. Studies have shown, for instance, that users make one hundred and thirty nine searches before purchasing an automobile!

Data-Driven Attribution Automobile Example — Data-Driven Attribution ***Automobile Example***

The Limits of Conversion Tracking

Though conversion tracking is useful in determining which ads are directly leading customers to making a purchase on your site, they don’t paint a full picture of the steps that customer took to arrive at their decision. Instead, all “credit” for that sale is given to the last ad the user clicked on. That misses the bigger picture.

As an example of this line of thinking: say that you want to book a vacation for you and your family. You might search on Google for “vacation ideas – southwest” and then click on an ad for California’s tourism site. After browsing a bit, you see pictures of Death Valley and instead get the inspiration to bring your family to the Grand Canyon. Then, you make a second search the next day to look for the latter, eventually clicking on an ad for trips to the Grand Canyon and purchasing tickets. How is Google supposed to decide how much credit for your purchase is given between the two ads on which you clicked?

The answer: data-driven attribution.

Data-Driven Attribution - Chart — Data-Driven Attribution – ***Chart***

Google takes a look at factors which include your unique account data and machine learning technology. They then compare the path you took to a purchase, with one another user took that did not convert. Using this method of comparison enables Google to put together a road map of how you got from point A to point B and decided to make your purchase.

This is great for advertisers. Not only does this attribution enable you to have a greater understanding of how your customers think and purchase from you, but when switching to a data-based, automated model, research has shown that your total number of conversions increase while your cost per conversion remains constant. Not a bad deal at all!

This awesome new tool is available in Google AdWords. Some accounts however, may not have enough data collected to enable it at first. In the meantime, Google suggests that advertisers still select an attribution model that gives credit across multiple clicks. If you do this, you’ll be giving yourself a more complete picture of how your customers operate. This enables you to tweak and optimize your ads to suit their patterns and needs.

Happy marketing!