On Friday October 19th, Facebook released their attribution tool after a year of alpha and beta testing their first iteration. This tool will help solve cross-channel, device, browser and online and offline measurement challenges by using a data-driven incrementality methodology powered by machine-learning. Understanding attribution is important as it helps advertisers make more informed decisions in regards to budget allocation – ultimately allowing brands to maximize ROI by shifting ad dollars to the best performing channels and platforms.
How it Works
“Facebook Attribution, built with a foundation of people-based insights, gives you a more complete picture of your marketing performance across publishers, on and off Facebook, and across channels and devices. See a more complete picture of your cross-device and cross-publisher impact. Assign credit to marketing touchpoints along the consumer journey to understand what's driving your business outcomes.”
We dug into this further to get a better understanding of this new tool. Using a combination of Facebook pixel, SDK, Offline Conversions and existing UTM parameters, Facebook is able to track a consumer’s conversion path, combining it with a variety of models to understand which touchpoints should get credit for a given conversion. For advertisers who do not have UTM parameters already implemented on their landing page, URL tagging options are available, though there are limitations to the data Facebook can access from larger publishers like YouTube and Snapchat. Ultimately, these touchpoints are combined with Facebook pixel data to de-dupe reach and provide people-based measurement and attribution data (something not possible for cookie-based tracking and tags). Finally, across Facebook-owned properties, this Attribution Tool can leverage pixel-based conversion events to create statistical attribution models that measure incremental lift driven by your Facebook campaigns.
It’s important to note that while Facebook offers this product for free, they are also integrated with several third-party measurement partners who offer similar Attribution Tools for a fee. For advertisers who are already working with these partners, it may be worth exploring these existing relationships to explore more robust and accurate attribution models and move beyond last touch attribution.
Although Facebook states that Facebook Attribution “is built with a foundation of people-based insights that gives you a more complete picture of your marketing performance across publishers, on and off Facebook, and across channels and devices,” they don’t provide a comprehensive explanation of exactly how this works outside of the Facebook family of apps and services. It’s easy to understand how Facebook connects the dots within their own ecosystem using the Facebook Pixel, SDK, impression and click data, UTM tags and online and offline conversion events by connecting these actions to individual identities. However, it’s not clear exactly how they’re measuring attribution across other channels beyond referral source links. This issue becomes even more questionable when you consider offline measurement in other channels such as Google, which measures store visits based on post-click vs. post-impression data.
It is also important to note that unlike Google’s Attribution Tool, Facebook’s Attribution Tool doesn’t currently allow advertisers to set up automatic optimizations to Facebook campaigns or inform automatic bidding. Its capabilities are currently limited to reporting only.
Given that Facebook Attribution is free, uses an unbiased methodology, and the barrier to entry is relatively painless, it is worthwhile for advertisers to test this attribution platform. Google’s version of a cross-channel cross-device attribution tool is still in beta and there are other platforms likely working on similar tools. As any media buying platform launches their tool, advertisers should test each to determine the variance in reporting. Since any current tool will likely use data-driven incrementality methodology, in theory, all of their attribution reports should be similar.
It’s commonly observed that today’s marketers are drowning in data, but starved for insights. Facebook’s new attribution tool offers the promise of helping marketers uncover valuable insights about how their initiatives work together, and smart advertisers will test this tool to determine its place in their measurement ecosystem.
Nadia Samuylina, Associate Director, Paid Social, also contributed to this blog post.