Be Smart About Smart Bidding
This past March several of us at J Miller Marketing were invited down to Google’s Chicago headquarters to learn about their recent advances in automation and how machine learning is being utilized via their Smart Bidding options to improve digital marketing campaigns. During the course of the day we gained vital insight on how to best use these advanced Smart Bidding features to drive results for our clients. We also learned that while it is a versatile and incredibly effective tactic, it takes a mindful marketer to identify the right situation, choose the correct Smart Bidding option and properly asses its performance.
Why Smart Bidding?
Smart bidding can be advantageous over other bidding strategies through its ability to adjust bids to the ideal value in real time. The smart bidding algorithm considers 72 million signals such like a user’s previous behavior or even the weather!
There are seven unique smart bidding strategies available for Google search:
Target Search Page Location – bids to ensure your ads show up on a specified location on the SERP (typically position 1).
Target Outranking Share – bids to ensure your ads show up above a specified competitor.
Maximize Clicks – bids to deliver the most clicks within your daily budget.
Maximize Conversions – bids to deliver the most conversions within your daily budget.
Target CPA – bids to deliver conversions at a set cost per conversion.
eCPC (effective CPC) – considered Target CPA lite or the “gateway” to Target CPA. Bids to deliver the most conversion but isn’t constrained by a target CPA. Used to discover what an achievable CPA would be for your campaign at which point you can transition to Target CPA.
Target ROAS – bids to maximize your return on ad spend rather than number of conversions.
Each of these strategies have their own strengths and weaknesses as well as baseline requirements for them to be effective. Having an agency familiar with these allows you to extract the most value from Google’s Smart Bidding strategies.
A Word of Caution on Smart Bidding
Many of the conversion based Smart Bidding tactics require a minimum amount of conversions before they can be implemented. In cases where you have little or no conversions the machine doesn’t have enough information to make decisions and your Smart Bidding strategy will fail. Providing the largest amount of data possible to the machine is the key to performance.
Because the machine needs data to inform its bid changes it takes time for results to develop. During the 1-2 week learning phase results can vary wildly as the machine learns what does and does not work. Seeing such drastic swings in traffic could led a cautious marketer to make changes but that is exactly what you shouldn’t do. Any changes restart the learning phase!
Conclusion
Machine learning and Smart Bidding has the potential to out perform manual bidding adjustments but that doesn’t mean it is always the right choice. A campaign that fails to meet the eligibility requirements or uses the wrong bidding strategy can cripple performance.
Keeping up with the latest innovations and features is not about being the first to implement them, rather it is about understanding them and properly testing which features are in the best interest of the client. That is the driving force behind everything we do here at J Miller Marketing.
Follow J Miller on Facebook, Twitter, Instagram, and LinkedIn. If you believe expert assistance will be a helpful addition to your marketing goals, contact us today.
