We use cookies. You have options. Cookies help us keep the site running smoothly and inform some of our advertising, but if you’d like to make adjustments, you can visit our Cookie Notice page for more information.
We’d like to use cookies on your device. Cookies help us keep the site running smoothly and inform some of our advertising, but how we use them is entirely up to you. Accept our recommended settings or customise them to your wishes.

2021 in Review: Google’s Deliberate Path to Full Automation

This year, Google made a slew of announcements about changes in their products and platforms. Individually, they were not surprising or revolutionary to day-to-day campaign management, but all together they solidify a future where leveraging automation is no longer an edge, it’s a must-have to not fall behind.

A recap of the major announcements this year

2021 google automation timeline

  • We started February with removal of broad match modifier (BMM) and refinement of broad match. While there was a lot of chat about BMM, the broad match changes actually posed opportunities to marketers to expand coverage and capture additional relevant traffic. Early testing has shown positive results with broad match when used in combination with auction-time bidding, showing improvement over historical performance for pure broad. More and more advertisers are interested in testing broad match, especially with recent competitive pressures and CPC increases.
  • On the creative side, in August we learned that responsive search ads (RSA) will be the only ad format starting next summer. With this format, we will no longer control what the final message is to each and every user, but will instead spend time creating more options for the machine-learning algorithm to tailor copy for each auction.
  • Performance max, announced in May 27 and now globally available as of November, introduced a fully automated campaign type across search, shopping, display, and video. This builds on previous launches of Smart Shopping and Discovery campaigns, which were the first foray into cross-channel ad formats.
  • In September, we also heard about data-driven attribution (DDA) becoming the default attribution model in Google Ads. Google is really pushing advertisers to better value upper funnel interactions and leverage modelling to understand incrementality of each touch.
  • Google also started to aid in account optimizations with the introduction of Recommendations and Optimization Score, but in April they took is a step further with auto opt-in of recommendations.

When combined with some previous changes, such as auction-time bidding and expansion of Optimization Score across formats (on video as well), after this year’s announcements, there are no elements of paid search where machine learning is not used.

list of search elements

What does the future look like?

None of the changes seem surprising, but what does that mean for the advertisers in the long run? Digital has gotten us addicted to granular insights and campaign tweaks, but as we progress down the path of the automation, the day-to-day execution will be less about detailed optimizations, and more about strong inputs to feed the machine.

  • Measurement will be a key pillar, as it always has been. Leading advertisers will invest time in making measurement more robust and moving beyond purely tracking online transactions. With additional signals, machine learning tools will handle more of the fine tuning that we are used to managing in day-to-day execution.
  • Attribution is another key part of the measurement pillar. While DDA might now be the default in Google Ads, advertisers should look for advanced attribution models across their stack, from bidding tools to overarching analytics packages.
  • Data feeds will also play a key role in smooth execution in an automated advertising world. More feed-based solutions are expected to surface, with some already starting to roll out in 2021 (such as Discovery ads with feeds). Data feed quality and a strong ability to manipulate and test fields that are customizable will be crucial in 2022.
  • And, of course, first-party data. Continued product evolution to make first-party data ingestion easier is expected with new ad formats, which will enable advertisers with a strong data strategy to differentiate from their peers.

All of these changes and the expectations for the new year also tie nicely into the developing privacy landscape. Privacy changes are forcing us to get more comfortable with modelled measurement and targeting options – and in a similar vein, the product changes for execution are pushing us to lean into tools that might seem harder to manipulate, yet still drive really strong results. Every year, we evolve and shift our best practices and 2022 definitely has the promise to challenge our thinking.