Google Ads Bidding
Can’t trust Smart Bidding for big decisions? Outperform it with an out-of-the-box approach.
Google pushes Smart Bidding to the Google Ads accounts to help you improve your campaigns. For most people, it’s a better solution than setting Cost-Per-Clicks (CPCs) manually with no help needed. The best part? It’s all free!
Do you have big budgets within your Google Ads account? Understandably, you’re probably looking for more sophisticated approaches that are tailored for your business. Let’s have a look at the weaknesses of Smart Bidding and the approaches you can use to outperform it.
LIMITATIONS OF SMART BIDDING
The following aspects aren’t the only limitations of Google Smart Bidding. You can see them as challenges in general when it comes to a bidding solution.
Smart Bidding and other 3rd party bidding tools have to work for many companies out there.
You will outperform those systems when bringing your business knowledge to the bidding models.
Sparse data
If the data isn’t sufficient, it’s hard to make good decisions, especially for long-tail keywords. When numbers aren’t stable from a statistical standpoint, you shouldn’t use them for your bid calculation.
High variance in basket sizes
At some point, you’ll use the basket size to calculate a value per click. But which numbers should you use? A single conversion can bring up your total basket size average. Is it a reason to raise the bids for that keyword for the next months? We don’t think so.
No external data sources
Google uses the data that is found in your Google Ads account. They don’t know about the overall transaction numbers of your e-commerce site. However, it makes total sense to use them for bid estimations and low sample size product keywords. What are your competitor prices look like today? It’s just an example where more data means better results for your bidding.
Short term seasonalities
If you run many sales actions for a limited time, it’ll take a while for Google to realize changes in conversion rate. The sale may have already finished when bids were adjusted.
Optimizing on margins
In the end, you’ll look at the profit you’re making as a company. However, how can we handle that in Google Ads when it comes to bidding? Just sending the basket size as a conversion value doesn’t tell the full story. You’ll also struggle when margins per products change over time.
Information of order cancellations
When you have many returns in your business, you should also consider this for your bidding approach. Of course, you can work with global percentages to make high-level adjustments. In reality, cancellations won’t be distributed in the same way.
DOES SMART BIDDING ALSO FAILS IN YOUR ACCOUNT?
When analyzing the status quo of the bidding solution, we look at many aspects to quantify the weaknesses of the current solution.
Let’s have a look at two simple bidding audits you can run easily on your account.
What is the highest CPC in your account?
If you follow Google’s recommendation, you may have not set upper bounds for your bids. The trouble often begins here. When Google’s machine learning algorithm finds various good signals, you’ll probably observe single click prices that are higher than your CPO target. You don’t understand machine learning to realize that there is something wrong. You’ll be surprised upon seeing what Google paid for a single click.
How does the 4th CPC quartile perform?
When your bidding has trouble with low sample sizes or outliers in your basket values, it causes wrong decisions. Historical data can’t be used for predicting the future performance. If a keyword triggered a big order randomly, it doesn’t mean you should bid more in the future. For that reason, you should have a close look at the top end of your most expensive clicks. What is the aggregated performance? The sneaky thing: You won’t realize it when you look at average numbers.
HOW TO BUILD YOUR OWN CUSTOM BIDDING SOLUTION
Yes, it requires a lot of work. If you spend huge budgets on Google Ads, it can be one of your best investments. Here are things to focus on some principles and techniques that will help you to get ideas for the components of your bidding system.
Spend your time on feature engineering
Machine learning isn’t a competitive advantage anymore. The guys who found the smartest ways for feature creation won the Kaggle challenges. It’ll make your model superior to others. What are generic approaches to generate new features? Firstly, creating N-grams on top of your keywords. Even better? Using an entity database for your business is also another example. The rest depends on your business —use your knowledge to make smart clustering, reduce dimensionality for attributes or use third-party data to create totally new data points.
Choose the right machine learning model
Of course, you can use stacked models that give you some more improvements in model quality. The downside is that they’re difficult to interpret and explain to other people. For that reason, we suggest Random Forest Regression —it’s possible to explain the most important variables.
