Last-click vs Data-driven: Never-ending confusion on attribution!

From the beginning of attribution modelling, we have been relying heavily on Last-click attribution. However, since Google has been pushing Data-driven attribution modelling to report conversions, now we are in bit of bothering position to pick one, having never-ending confusions to decide which one should we go for?

I'm guessing your Google Ads account is recommending you switch your conversion tracking to data-driven attribution. Like most people, you are likely using last-click attribution at the moment and have no freaking clue what the difference really is. That is a pretty normal feeling. Let's dive deeper and help you make a more informed decision.

Data-driven Attribution Model?

Data-driven attribution, also known as DDA, is the newest attribution model and one that Google recommends adopting, providing your account meets certain criteria. But you may be wondering how Google Ads data-driven attribution model gives credit for conversions.

Data-driven attribution uses advanced machine learning to analyze data and decide how important each touchpoint is in a customer's journey. Conversions are broken up and attributed to each touchpoint based on its influence and impact on a customer converting.

Clicks and video engagements are analyzed across Search (including Shopping), YouTube, Display, and Discovery ads in Google Ads to identify patterns that lead to conversions. When using automated bidding, not only do these patterns support DDA to assign conversions, but they will also help the bid strategy leverage data and patterns that lead to conversions to find customers that behave in a similar way. This is what makes data-driven attribution the most advanced attribution model.

Perfect for businesses with complex conversion paths and those that have multiple touchpoints as well as any eligible business with an abundance of data that would like to benefit from machine learning. Since it uses advanced algorithms to decipher data and attribute conversions, DDA can provide better clarity over a campaign, ad group, keyword and ad performance making it a good choice for most accounts.?

• Pros: Uses machine learning to assign credit to touchpoints based on their impact on conversions. This means it provides a more accurate view of the customer journey.
• Cons: Requires a lot of data to function and its fundamental that conversion tracking is accurate. This may prevent businesses with little conversion data and accounts with tracking issues from adopting this attribution model.

Last Click Attribution Model?

Last-click attribution, as the name suggests, gives all the credit to the last touchpoint before converting. Last-click attribution is straightforward and commonly used, however, there has been a shift in recent years for the need to focus on more than just the last click, taking into account the multiple touchpoints throughout a customer’s journey.

For example, a conversion path might consist of multiple touchpoints, starting with generic keywords, followed by Display and Video ad interactions, and ending with a conversion taking place from branded keywords. In this example, the brand keyword will get all of the credit. However, you could argue the generic keyword that introduced the customer to the business played a role in the conversion or is equally as important as the brand keyword the conversion is attributed to. The same could be said for the video and display interactions.

Perfect for businesses that have few touchpoints with users before a conversion takes place, such as e-commerce businesses with a short sales cycle.

• Pros: Simple and easy to implement. This model provides insight into how channels perform on a basic level.
• Cons: Ignores all touchpoints except the last one. For this reason, it may not provide a comprehensive overview of the customer journey and the value of how other channels and campaigns contribute to conversions.

So? Which one is SUPERIOR?

Last-click attribution will always assign credit for a sale to the last ad that a user clicked. By contrast, data-driven attribution will spread credit for a sale across every ad campaign that the user interacted with over the span of the conversion, with fractional credit being assigned to each campaign based on Google’s algorithmic analysis of the campaigns.

“Conversion modeling powered by machine learning allows you to preserve measurement even when cookies or other identifiers aren’t present. Data-driven attribution in Google Ads takes this a step further. It uses advanced machine learning to more accurately understand how each marketing touchpoint contributed to a conversion, all while respecting user privacy.”

— Vidhya Srinivasan, VP/GM Buying, Analytics and Measurement, Google Ads

We know that buyer paths for sales can have multiple steps, with users interacting with multiple ads, ad types and devices. By taking advantage of data-driven attribution, we can more accurately gauge the effectiveness of all campaigns and may discover that some lower-performing areas of the account at the top of the funnel actually have a greater overall value than expected.

From my industry experience, let me give you a quick example-

Think of the case of a business traveler. This person needs to travel to Singapore for business purposes and is looking for flight options. He googles for suitable flight options and gets to know about GoZayaan from paid search campaign. He clicks the ad, visits website and checks all available options, and then moves out without booking. Later, he notices another ad while reading an article about promotional offers for booking Singapore flights. He clicks the ad, gets to know about the offer, and again moves out without booking. Later, while watching a video on youtube, he sees another ad talking about the benefits of booking flights from GoZayaan app. This time, he clicks the ad, download the app, and book the flight using the deals.

Here, the action path looks like this-

Click on a Search Ad>Click on a Display Ad>Click on a YouTube Ad>Conversion!

DDA may create?a model that distributes credit as follows?(the?percentages below are just examples and may be very different for your specific campaigns):

Display: 10% Generic:?30% Brand:?60%

Here are 3 case studies of real businesses using data-driven attribution:

1. Medpex, the largest mail-order pharmacy in Germany, used data-driven attribution together with smart bidding. This resulted in a +29% increase in the number of conversions and a -28% decrease in cost per acquisition.??

2. Select Home Warranty is a provider of household warranty for repair projects in the United States. Using data-driven attribution, they saw a +36% increase in leads and a -20% decrease in CPA.?

3. H.I.S. is a global travel agency that operates in over one hundred cities around the world. Using DDA, Smart Bidding and Dynamic Search Ads, H.I.S were able to drive a +62% increase in the number of conversions at the same CPA.

Conclusion:

For quite a while, the last-click attribution model was all we had to go on. We simply didn’t have a way of seeing how each touch contributed to the ultimate conversion. But a number of tools and capabilities now exist to give us a more holistic view, or what’s known as data-driven attribution models.

First-touch, linear, position based, and time decay are all rules-based attribution models. They’re about aligning philosophically within an organization and deciding that this model is the lens you want to apply to the conversion journey.

But data-driven attribution is a wholly different way of looking at it. Powered by machine learning, it takes all the guesswork and philosophizing out of the equation, and looks at conversions objectively based on trends. Going back to the example of the couple’s conversion journey, it will take into account every marriage on record and find patterns — in cases where the couple didn’t go to the movies, were there fewer weddings in the end? If so, then that movie date in the middle starts to get a bit more credit for the conversion.

Measurement platforms with data-driven attribution capabilities, such as Google Analytics 4 or Google Analytics 360, use machine learning to look at every conversion and all the elements that led to it, to determine which steps needed to be there and which were less important.

Regardless of which attribution model you use, your goal should be to get as big a scope as possible on all the information in order to drive better, faster decisions. Attribution will allow you to answer key questions such as: how much does each touchpoint contribute towards driving a conversion? Do you have all the information about what’s driving those conversions? Are you looking at that information through the right lens?

If you can find the answers to those questions, your measurement tools and your business will have a happy and successful marriage.