Deciding on an attribution window to review your data is just as important as the KPIs you use to determine how effective your advertising is—in fact, it will directly affect how those KPIs look depending on which lens you choose to look at your data through.
We have a client who drives membership sign-ups and looks at their cost per acquisition to decide on how to invest budget into their various marketing channels. Their target CPA was $80. In this blog post, you will see 3 examples of how they could look at attribution and which we chose and recognize to be the most accurate to calculate their CPA, resulting in them to scale their advertising campaigns.
As you can see below, the number of membership sign-ups they received in 2019 via Facebook Ads looks quite different depending on the attribution window you choose.
These numbers are even less when you look at their membership sign ups in Google default last-click attribution because it doesn’t account for view through conversions, stripped UTM parameters, or assisted conversions from Facebook Ads.
There are several attribution window combinations you can choose to use, but some will be more accurate than others. We’re breaking down three common attribution windows, listing the pros and cons and comparing results from the same account in the same time period!
Google Analytics Last-Click
This view only looks at conversions in Google Analytics associated with coming directly from a Facebook ad.
Facebook Spend / Google Analytics Membership Signups from Last-Click Facebook source/medium
You’ll have a good sense of your direct response traffic from your specific ad—that is, you’ll know how many people purchased/converted directly from a click. This view allows you to see how traffic from a channel interacts with your website through bounce rate, average time on page, average number of pages viewed, etc.
This attribution model only tracks the very last thing a user does, completely ignoring the earlier part of the customer journey. It doesn’t have all the conversions that came from the Facebook ads because of issues like UTM parameters typically dropping off when someone clicks an ad in the Facebook app and it redirects to their web browser.
Only looking at last-click conversions avoided any overlap with other channels, however, it didn’t show all of the membership sign-ups driven from their Facebook campaigns.
Facebook 28-Day Click/1-Day View Default Attribution
This view looks at all conversions in Facebook ads that viewed or clicked the ad within the automatic Facebook attribution window (28-day click, 1-day view).
Facebook Spend / Membership Signups from anyone who clicked an ad in 28 days or viewed an ad within one day
This view accounts for all conversions that interacted with your Facebook campaigns. Many agencies and brands use this attribution view because it seems like it’s the most profitable.
This view is too broad and oftentimes will cause a large overlap in reporting when you are running multiple advertising channels, making it hard to actually determine how large of a role Facebook played in the customer journey. For example, maybe the last thing they clicked was a Google ad, but they clicked a Facebook ad 28 days ago and now the sale is attributed to both.
This view was too broad and made the CPA look very low in comparison to Google’s numbers. Additionally, we were not able to discern how much of an impact Facebook actually had on converting these users and how much channel overlap was being reported on.
7-Day Click/1-Day View Blue Wheel Attribution
This view looked at all conversions in Facebook ads that clicked an ad within 7 days or viewed an ad within 1 day and converted.
At Blue Wheel, we always report on Facebook ads using this attribution view. As you’ll see, we believe that this view gives the most accurate representation of how many conversions Facebook is actually driving.
Facebook Spend / Membership Signups from anyone who clicked an ad in 7 days or viewed an ad within 1 day.
This accounts for the first interaction with a Facebook ad and within one week of clicking an ad to minimalize overlap with other channels without completely ignoring people who don’t convert immediately.
Our brands frequently want to compare this view with Google Last-Click and, as stated above, Google Last-Click doesn’t account for view-throughs, UTM stripping, or traffic that converts outside of the last-click.
With the narrow view, it appeared we were drastically unprofitable, and with the broad view, we were over-reporting on conversions coming in from other channels.
This view gave us a better understanding of how much we could scale based on a CPA that was in between the broad and narrow views. Using this attribution window, we were able to forecast budget according to how much we could spend while maintaining their CPA goal.
Choose the Right Attribution for You
The client we reviewed in this post had a very short sales cycle. That being said, something like a 28-day click attribution may make more sense for a brand that has a longer time-to-close.
Whatever attribution you decide, make sure you decide which is the most accurate to your business and use that view consistently to set your benchmarks. It’s crucial to view all your Facebook campaigns through the same model so you can better understand how to optimize!
Keep in mind that you should only use the spend on a specific campaign to calculate that campaign’s CPA. Total spend isn’t an efficient number to use when you have other campaigns with different goals, such as awareness, lead gen, etc.
Check back soon for the final part of our ad series, which will focus on everyone’s favorite topic—testing! Be sure to follow us on LinkedIn for the latest.