Google’s Ad Auction: Why Target CPA Creates a Premium Bidding Problem

TL;DR

Introduction

Google has long been heralded as the gold standard in performance marketing thanks to the high intent of search queries. For those looking to sell a product, there’s no better way to find prospective buyers than to target search terms related to that product. It’s the main reason Google has been able to achieve astronomical growth as a business in the last 20 years.

However, this reliance on an ad revenue model only works if advertisers are seeing monetary benefits and choosing to invest more money into the platform. To this end, Google has always developed more ways to deliver value to its customers (advertisers), most of which have genuinely improved the ad experience.

Google’s Evolution in Advertising

From its origins as a simple search engine, Google’s ad auction has evolved dramatically. By integrating YouTube, expanding the Display Network to Pixel phones’ Discover feed, and adding ads to Google Maps, they’ve created diverse advertising opportunities for businesses of all sizes.

Behind these product expansions lies Google’s core strength: data. As one of the world’s largest data companies, Google’s vast user insights have enabled the development of sophisticated conversion-optimized bidding strategies. But this has come at a cost, as we’ll discuss the problem with Target CPA and its impact on advertisers.

Understanding Google’s Data Infrastructure

The Scale of User Data

Google’s data graph, or individual data points on each user can be measured in gigabytes. To put that into perspective, you could fit roughly 100,000 pages of a word doc into 1 GB. Depending on account history, 1GB of data is a very conservative estimate. If you’re curious to see for yourself, you can download the data linked to your account at this link.

This level of tracking spans all Google products, including but not limited to:

Impact on Advertising Capabilities

In short, this means Google’s ad targeting capabilities are second to none (maybe Meta). At the very least, Google is able to get a pretty accurate picture of who you are, what your interests are, whether you’re thinking about buying those shoes you liked, or are at the point of having your credit card in hand.

Advertisers looking to sell products or get in front of the right decision makers can make use of this data, for a price of course. The result is hyper-targeted ads to individuals that are searching for your products, browsing competitors, or otherwise research/considering your product or service. Sounds pretty great, no?

Before we jump to conclusions, let’s take a step back and review how ad auctions have worked historically.  

How do Google’s ad auctions work?

Second-price auctions

Ad auctions have traditionally been second-price auctions, wherein the advertiser who bids the highest pays the second highest advertisers’ bid plus 1 cent. For example:

Goodway explains this well in their blog.

Google’s Quality Score Enhancement

Google’s ad auction takes this one step further by incorporating what they call a quality score, which is derived from three factors:

This is Google’s way of ensuring that it delivers the best experience to the end user. The result is a system called ‘Ad Rank’, which multiplies an advertiser’s bid and Quality Score (rated 1-10).

Let’s look at an example:

Advertiser B could pay up to $3.01 depending on whether there is a third advertiser in the auction, primarily due to their below-average Quality Score. This demonstrates how Google’s system rewards engaging ads and landing pages while penalizing poor user experiences – effectively allowing advertisers to “buy” better positions through either higher bids or better quality content.

Conversion-Optimized Campaigns

Google’s conversion-optimized campaigns use AI to predict how likely each user is to make a purchase. Think of it like a heat map – users who’ve been researching shoes across multiple sites like Nike, Adidas, and Foot Locker light up as ‘hot’ prospects with a high probability of buying. They’ve shown clear shopping intent through their behavior. Meanwhile, someone just casually searching shoes for the first time appears as a ‘cool’ prospect, with a lower probability of converting.

This predictive scoring completely transforms how the ad auction works. Instead of just bidding on keywords, advertisers are essentially bidding on the likelihood of a sale. The result? Access to those high-probability users becomes premium inventory, commanding higher prices in the auction.

This shift from basic keyword bidding to conversion probability transforms Google’s ad auction into an entirely new dynamic.

Google’s New Ad Auction Formula

Ad Rank = (Target CPA x Conversion Probability x Quality Score)

A desired CPA (Target CPA) is set, or left unrestricted, indicating willingness to pay whatever is necessary to win the auction within budget constraints. Google assigns a conversion probability based on user intent signals, while quality score continues to function as it did in the traditional model.

The fundamental change in Google’s ad auction is clear: rather than bidding for clicks, advertisers now effectively bid for conversions. While the second-price auction rule still applies ($0.01 more than the next highest bid), CPC bids can vary dramatically based on conversion probability.

The result is a tiered system where high-probability auctions become premium inventory, requiring significantly higher bids for entry. And therein lies the problem with Target CPA: advertisers are essentially forced to use conversion-optimized models to access these high-intent users.

Strategic Implications

This model presents distinct opportunities and challenges:

Advantages:

Limitations:

Recommendations

The current landscape demands a more sophisticated, multi-channel approach that:

For a practical example of this multi-channel strategy in action, see CDP Marketing Strategy: Growing Leads with LinkedIn and Youtube Ads.

Conclusion

While Google’s ad auction system promotes conversion-optimized campaigns as the path to success, the reality is more nuanced. High-intent targeting through Target CPA comes at a premium price, and relying solely on these automated bidding strategies can limit growth. True success in digital advertising demands a balanced approach: one that combines strategic high-intent targeting with broader marketing initiatives that build awareness and engage users throughout their buying journey.