How Google rewards Relevant Ads

First, we need to understand how ad actions work.

The position your ad ends up in the auction (or if it shows at all) and how much you pay for a click is determined by its ad rank which depends on two components: Your Bid (max CPC) and Relevance. In this context, relevance is how relevant your ad experience is to a Google User.

Let’s step back and think of Google Search as a business: it has two dichotomous objectives:

A) Serve up good relevant interesting results to users (including paid results)
B) Sell Ad slot to make revenue in action.

While this would never be explicitly mentioned in the Google help articles. It would be best to incentivize advertisers that help with objective A and which is what Ad Rank is all about.

If we read the following official Ad Rank documentation we could see how it’s all about resolving the above objectives.

    • Auction-time ad quality (expected clickthrough rate CTR, ad-relevance, and landing page experience),

All of the above are good proxies of relevance.

    • Expected impact of extensions and other ad formats.

Generally, information-rich ads are a better experience.

    • Context of the person’s search (for example, the person’s location, device, time of search, the nature of the search terms, the other ads and search results that show on the page, and other user signals and attributes),

Again, This is all about the users if the ad itself is relevant

    • Ad Rank thresholds

This is the component that actually stops ads from even appearing if it is deemed too “low-quality!”

Actual Definition of Quality Score

Quality Score is an estimate of the quality of your Ads.
The most important thing to note is that it is  not used at auction time to determine Ad Rank. However, it is intended to give you a general sense of the quality of your ads.

This could be broken into three components

  • Expected clickthrough rate
  • Ad relevance
  • Landing page experience

All of which we know influence Ad Rank. Higher quality ads can lead to lower prices and better ad positions.

Using Quality Score

Google Ads help details usage of this metric as a diagnostic tool here.

The key challenges are:

1) The need to identify high value areas to fix

2) Ensure the most competitive ads for each auction

3) Improving ad quality

  • User Intention
  • Device Experience

As the above is best done at an auction level (ie, keyword) which would lead to overwhelming amounts of data points. Action areas also become diffused if we work down a list of bad quality score keywords (ie, creating multiple landing pages and adcopies to address each class of keywords)

Swarmplots to the Rescue

Swarmplots plot allows information dense chart of QS of individual keywords to be generated.

This allows us to quickly visualize which theme needs the most work on and most importantly if we need to make our structure and experience more granular.

The method is fairly straightforward – We tokenize keywords to determine the top themes driving our campaginsThe individual keywords QS is plotted to show the spread of QS.

With the above we could quickly identify actionable scopes of optimization and pin point which experiences we need to alter to improve the account.

Swarmplots example: Apple Product Retailer

Here we could quickly identify that this retailer has generally done well on QS for most themes.

The more problematic keywords are contained in the Ipad and Iphone theme.

Digging deeper into the keywords we see that the low QS iphone/ipad keywords were of older models

by altering adcopies and landing page experience we could quickly improve the relevance of these keywords.