Using AI to Optimize Your Google AdWord Bidding

May 8, 2018
by
· 4 min read

Digital advertising is no passing trend. According to a recent study by professional services firm PwC, companies spent over $20 billion on internet advertising in each of the first two quarters of 2017 — a dramatic increase from the first half of 2016.

 

In my experience, most companies lack an accurate way to estimate their marketing attribution per keyword. 

A big part of that ad spending goes to bidding on search terms through Google AdWords, and competing for high-traffic keywords is increasingly expensive. Some of the most competitive keywords can cost up to $50 per click – a hefty price to pay when it is so difficult to accurately determine the return on investment (ROI) of each individual marketing touchpoint (what we call “marketing attribution”).

In my experience, most companies lack an accurate way to estimate their marketing attribution per keyword. They rely on experience (guesses) to determine on which keywords to bid and for how much, which often results in wasteful spending on the highly competitive keywords. Essentially, that’s money being flushed down the drain. However, by applying machine learning and artificial intelligence (AI) to Google Adwords spending, companies can avoid this unnecessary drain on resources.

 

The Challenge: Achieving Target ROI on Google AdWords

Bidding on Google Adwords presents a double bind: bid too low and you lose advertising opportunities, causing your product and brand awareness to fall, dragging sales volume (and profit) down with them. Bid too high, and your marketing ROI will dwindle, making your efforts redundant if not actively detrimental.

This problem, painfully obvious when bidding for keywords, is true for all marketing activities. In the digital age, companies have to spread themselves over dozens of new communications channels, including: social media, digital advertising, television, print or radio, and more.

 

The secret to achieving target marketing ROI lies in consumer data: purchase behavior, overall consumer trends, seasonal patterns, demographics, and more.

“What we are seeing in today’s world of digital commerce is that the average transaction has more than 30 touchpoints in it,” said Lewis Brannon, paid search manager at CPC Strategy. With that many touchpoints, how can marketing departments hope to accurately determine which Google AdWords are actually driving sales? Not only that, marketers also have to take into account rapidly changing consumer behavior due to seasonal factors, competitor activity, and general trends in consumption.

The secret to achieving target marketing ROI lies in consumer data: purchase behavior, overall consumer trends, seasonal patterns, demographics, and more. AI and machine learning solutions allow companies to take advantage of all their available data – if, that is, they are able to afford, find, and retain the data scientists typically necessary to perform this type of analysis.

 

The Solution: Determining Optimum Bidding Prices with Automated Machine Learning

Automated machine learning, a type of AI that requires minimal human involvement, is helping companies overcome the data science skills gap. This new technology uses automation to quickly find patterns in historical data and build models to make predictions for future outcomes without requiring explicit programming.

 

By inputting past marketing activity and sales volume into automated machine learning software, companies build models that predict the sales volume attributable to each individual AdWord ad and therefore the maximum price to bid while still achieving their target ROI.

This has major implications for marketing strategy. For example, using automated machine learning, companies can determine the maximum and minimum prices they should bid on each individual Google AdWord in order to achieve their target marketing ROI and sales volume.

Each AdWord ad is only worth as much as the sales volume to which it contributes. To complicate matters, it shares the credit for those sales with all the other touchpoints a consumer experiences before the time of purchase. By inputting past marketing activity and sales volume into automated machine learning software, companies build models that predict the sales volume attributable to each individual AdWord ad and therefore the maximum price to bid while still achieving their target ROI.

 

Automated machine learning allows users of all skill levels to quickly and consistently develop and iterate on models that give insight into where Google AdWords spending is delivering results.

On the other end, to avoid bidding too low and losing the premium position for the AdWord ad to a competitor, a machine learning model must take into account the price elasticity of the ad by ingesting information about past winning and losing bids. This helps to determine the minimum price to bid on that AdWord.

Automated machine learning allows users of all skill levels to quickly and consistently develop and iterate on models that give insight into where Google AdWords spending is delivering results. DataRobot has customers who automatically rebuild their models every three days – a feat that would be impossible using traditional modeling methods.

By implementing automated machine learning, companies get real, actionable insights into where their marketing spending is doing the most good, allowing them to increase marketing ROI and focus on the keywords that really matter – without having to hire additional data scientists.

 

New Call-to-action

 

About the Author:
Colin Priest is the Director of Product Marketing for DataRobot, where he advises businesses on how to build business cases and successfully manage data science projects. Colin has held a number of CEO and general management roles, where he has championed data science initiatives in financial services, healthcare, security, oil and gas, government and marketing. Colin is a firm believer in data-based decision making and applying automation to improve customer experience. He is passionate about the science of healthcare and does pro-bono work to support cancer research.

 

 

About the author
Colin Priest
Colin Priest

VP, AI Strategy, DataRobot

Colin Priest is the VP of AI Strategy for DataRobot, where he advises businesses on how to build business cases and successfully manage data science projects. Colin has held a number of CEO and general management roles, where he has championed data science initiatives in financial services, healthcare, security, oil and gas, government and marketing. Colin is a firm believer in data-based decision making and applying automation to improve customer experience. He is passionate about the science of healthcare and does pro-bono work to support cancer research.

Meet Colin Priest
  • Listen to the blog
     
  • Share this post
    Subscribe to DataRobot Blog
    Newsletter Subscription
    Subscribe to our Blog