Where is Google Attribution?
In May 2017, Google announced the launch of a free version of Google Attribution. In October of that year, it said the tool had rolled out to “hundreds more” advertisers. But now, more than a year and a half later, it has stayed under the radar and has yet to fully roll out.
It’s not uncommon for Google — and other tech firms — to announce something and then never release it or release it much later than anyone expected. So where does Google Attribution stand? It’s still alive, and it’s still in beta.
The company says it is continuing to collect customer feedback and does not have any updates to share at this time. Agency marketers that have had clients testing it say they have been giving Google their input. In the course of those communications, one marketer heard about a tentative timeline of late 2019 for release, but said that was not definite.
What we’re waiting for
The big selling point of Google’s free attribution tool is to help marketers make more informed bidding decisions in Google Ads campaigns by letting them capture an ad’s contribution at any point along the conversion path, and not just when it’s the last click, as highlighted in a case study of Nordic Choice Hotels from September.
It pulls in data from Google Analytics, Google Ads and Google Search Ads 360 (formerly DoubleClick Search) and applies the advertiser’s chosen attribution model, including Google’s machine learning-powered model called data-driven attribution, across channels and devices. That data can then get fed back into automated bidding strategies in Google Ads or Google Search Ads 360.
Presumably, non-brand search and display campaigns that tend to be higher funnel will be likely to get more credit when looking at the full journey, and with Google’s automated Smart Bidding strategies, bids will be adjusted accordingly.
Some advertisers who are testing it out said they are worried that Google may favor its own channels. Several marketers in the beta spoke to us about their thoughts on the tool on the condition of anonymity. Because there’s no way to see exactly how much credit Google is assigning various touch points, marketers are left in the dark about the weighting formulas. Google can say it treats all touch points equally, but it’s hard to tell marketers to toss aside skepticism when they can’t see the data for themselves.
Other feedback from those testing the free tool is that actionable insights aren’t easily surfaced and still require digging to find.
Another agency executive said it’s definitely a work in progress with mixed results, but that Google Attribution is still promising. That team has been providing Google with feedback on issues, and currently recommend clients use it directionally.
At the enterprise level, Google deprecated the paid Attribution 360 digital attribution beta in October. There are attribution features in beta, including Model Explorer and ROI Analysis, available in Google Analytics 360. Attribution 360 is separate from Google’s TV Attribution product, which aims to show the cross-channel (i.e. digital and search) impact of television campaigns. Google is still focusing on cross-media metrics and working on evolving TV Attribution into a holistic video measurement solution that measures both TV and online video.
Google, of course, is not the only ad seller working on attribution tools to get marketers away from last click models. Amazon has an attribution tool of its own in beta. Facebook made its attribution tool available to all advertisers in October, after beta testing started in March 2017. Marketers have reported mixed reviews.
Resources on attribution and paid media
- Enterprise Paid Media Campaign Management Platforms: A Marketer's Guide (Sponsored)
- Pixel problems: Troubleshooting your paid media tracking implementation
- 5 takeaways from Google's Q4 2018 earnings for search marketers
- LinkedIn's 'Interest Targeting' lets advertisers target ads based on users' professional interests
This story first appeared on Marketing Land. For more on digital marketing, click here.