Google Ads: Smart-Bidding Best Practices

December 9, 2019

Google’s machine learning models for bid optimization are collectively called “smart bidding” Most of these models can be applied as “portfolio bid strategies,” allowing you to group together multiple campaigns into a single bid strategy with a common goal. All of Google’s automated bid strategies use a feature called “auction-time bidding”, meaning bids are optimized in each and every auction.

Each bidding strategy that is currently available is ideal for different scenarios. In some cases, multiple bid strategies work effectively, and there are instances where none of them work better than manual CPC. The only way to know if smart bidding is right for your account is through testing, so we recommend launching smart bidding tests via the Draft and Experiments feature of Google Ads to ensure performance improvement before moving your full campaign budget over to Google’s “smart bidding” machine learning models.

Below, we walk through each strategy and outline the recommended approach to effectively utilize each of the currently available “smart bidding” strategies– these are listed in approximate order of least to most advanced:

Enhanced CPC (enhanced cost per click)

Allows Google to help you get more conversions from manual bidding by raising your max CPC bids for clicks that seem more likely to lead to conversions and lowering your max CPC bids for clicks that seem less likely to convert.

Maximize Clicks

Google automatically sets bids to help get as many clicks as possible within your budget, with the option to set a max CPC bid limit to cap the cost per click. This is a safe starting point for trying out smart bidding earnestly. If your goal is volume then it will be a worthwhile test. And if your goal is cost efficiency then we recommend setting a max CPC bid limit just above your avg CPC to see how it performs.

Maximize Conversions

Google automatically sets bids to help get as many conversions as possible with your budget. This is a safe bet setting to use so long as your conversion tracking signals are high quality and there is also sufficient quantity to inform the models. If you are relying on smart goals or have very few conversions per day, this model won’t work as well for you.

Target Impression Share

Google automatically sets bids to increase your ads’ chances of appearing in the search page area you select. We recommend selecting “top of results page” for where you want your ads to appear and setting an attainable impression share percent to target (e.g. if your current Top IS is 30% consider targeting 50% as opposed to 100%). Incremental changes mitigate risk and avoid wasted spend. Here as well you can set a max CPC bid limit to keep costs per click under control.

Target CPA = Target Cost per Acquisition (cost per conversion)

Google automatically sets bids to help get as many conversions as possible at the target cost-per-action (CPA) you set. Some conversions may cost more or less than your target. This is a great strategy if you want to test accounts with a good amount of conversion volume where the primary goal is improving cost efficiency.

Maximize Conversion Value

Google automatically sets your bids to help you get the most conversion value within your budget. This a good strategy to test in accounts where eCommerce tracking or conversion weighting has been implemented so real (or at least directional) revenue can be attributed to keywords and ads. This is a great strategy to test in order to increase total revenue, but bear in mind this is only worth testing if your revenue tracking and attribution back to Google Ads is implemented and working properly.

Target ROAS

Google sets bids to help get as much conversion value as possible at the target return on ad spend(ROAS) you set. Some conversions may have a higher or lower return than your target. This is the ideal strategy for more advanced accounts that have accurate eCommerce tracking or directional conversion weighting implemented. Essentially, this strategy allows you to optimize towards the most qualified (high value) leads and away from the least valuable leads.

Overall, it’s worth testing Google’s smart bidding strategies. They have the potential to drive improved performance (if tested in carefully) and Google commits to improving them over time. In our experience, smart bidding strategies perform best in accounts with lots of historical data for the models to work with and plenty of room to improve. Where you’re most likely to see smart bidding come up short compared to manual CPC are in accounts that are fairly new and/or lacking substantive conversion tracking, or in accounts that have been heavily optimized for years and thus have little room for improvement.

If you’d like to learn more about smart bidding and what strategy is best for your business, drop us a line, we’d love to chat.

Ready to take your ads, and your business, to the next level? Get in touch with the DELVE team today.