PerformanceININside Performance Marketing
JOIN THE PERFORMANCEIN Join the PerformanceIN Partner Network
Google Launches Machine Learning Attribution Product as “Alternative” to Last Click

Google Launches Machine Learning Attribution Product as “Alternative” to Last Click


Google has promised a free tool for its advertisers to measure the performance of their marketing across its products.

The Alphabet-owned search giant Google has today announced Google Attribution, a new way for its advertisers to measure the effects of their marketing, making use of the company’s acquired AI and machine-learning technology.

Baked into Analytics, the product also integrates with DoubleClick and AdWords, bringing together data from across Google to provide a broad picture of overall performance. It works with technology it acquired from attribution solutions company Adometry three years ago.

According to one of Google’s senior product managers, Babak Pahlavan, the free tool will measure clicks from “all sources”, including email and social - as long as the data is deployed by Google Analytics.

The machine learning aspect works by analysing each account’s unique conversion patterns, comparing the paths of those who go on to convert with those that don’t, and determining how much credit to assign each step in the consumer journey. These tailored results are available immediately for “reporting, updating bids or moving budget” between related channels.

For more complex requests, Google has also launched an enterprise solution called Attribution 360; "This connects to additional data sources such as Doubleclick ad manager, which handles non-Google sources, and produces more sophisticated reports," Pahlavan said.

Combating complexity

The search engine’s vice president of product management, Jerry Dischler, commented that an age where brands can access customers at any moment as they flit between tablets, mobile and desktop is “exciting”, but has created “unprecedented complexity”.

“There is more data, more platforms, more channels. On a daily basis, Google Analytics processes half a trillion digital moments across devices. What we are trying to do is use tech and machine learning to help make ad platforms more useful, easier and better connected.”

Google has heralded the beta launch as an alternative to on-the-market attribution tools, which it said are hard to set up, easy to lose track of between devices, and difficult to integrate with ad tools - therefore difficult to take action with.

“As a result, many marketers are stuck using last-click attribution, which misses the impact of most marketing touch points,” said Google.

Online-to-offline measurement

Alongside the launch of Attribution, Google also announced advancements into its offline attribution capabilites.

This is building on its store-visit technology released over two years ago, which it claims has measured over five billion store visits across 17 countries and is already being used by its advertisers to inform spend on various channels.

The upgrade to its “deep learning models” and mapping technology, however, will allow advertisers to measure store visits in much more challenging scenarios, such as multi-story malls in cities where business locations are densely packed, such as Tokyo and São Paulo.

While its store visit technology is already available for Search, Shopping and Display, it will be adding this capability to YouTube Trueview campaigns, allowing advertisers to measure the impact of video ads on foot traffic.

Perhaps the most significant update is that in the coming months Google plans to roll out “store sales management”, allowing advertisers to measure in-store revenue delivered by Search Shopping ads, in addition to just visits.

Google’s Attribution announcement today follows not long after Facebook’s most recent blunder which saw it erroneously charging advertisers for clicks on its carousel ads. The result: calls from the industry for large ad techs to stop “marking their own homework” and involve third-party analytics companies to verify the results of the ads they’re serving.

Continue the conversation

Have something to say about this article? Comment above, share it with the author @markjpi or directly on Facebook, Twitter or our LinkedIn Group.

Mark  Jones

Mark Jones

Mark manages all aspects of editorial on PerformanceIN as the company's Head of Content, including reporting on the fast-paced world of digital marketing and curating the site’s network of expert industry contributions.

Going by the ethos that there is no 'jack-of-all-trades' in performance marketing, only experts within their field, Mark’s day-to-day aim is to provide an engaging platform for members to learn and question one another, helping to push the industry forward as a result.

Originally from Plymouth, Mark studied in Reading and London, eventually earning his Master's in Digital Journalism- before making his return to the West Country to join the PI team in Bristol.

Read more from Mark

You may also like…