Google is bringing more AI into its ad business with the introduction of a new ad unit for AdSense. The new release aims to use the power of machine learning to try and optimise ad placement.
The new Auto Ads uses machine learning to “read” webpages and work out what the most appropriate places to put ads might be, and how many ads should be run. Publishers can activate the feature by adding a single line of code to their pages.
The service first appeared in a quiet, limited beta in in the second quarter of 2017. Google claims that publishers taking part in the beta saw an average revenue lift of 10%, and revenue increases ranging from 5% to 15%.
Benefits
AdSense is already a fairly automated service, but so far it has been up to web publishers where they want ads to be placed. Google has then selected the ads based on analysis of the page that matches the content to relevant ads.
The service is incredibly important to Google’s parent Alphabet, accounting for a whopping $27 billion of its $32 billion ad revenue in Q4 2017.
Auto Ads takes the responsibility for ad placement away from web publishers. Google claims that the service will only show ads when they are likely to perform well and provide a good user experience. The use of AI to work out where to place ads based on how they are likely to perform is an interesting development.
AI will also estimate how many ads should be on your page to increase revenue. Whether this leads to web publishers logging on to site to find it filled with ads. This post seems to indicate to that some beta testers were not exactly elated at the number of ads that appeared on their sites. It will be interesting to see if the AI can balance revenue with user experience.
