In advance of Apple’s promise to crack down on user tracking on iOS, Facebook has begun implementing a series of updates to the platform that will forever change the way advertisers optimize and collect data.
Historically, Facebook has used a default 28-day click, 1-day view attribution window, while offering advertisers the ability to select a shorter window if desired. This means Facebook had the ability to track users who engaged with or saw any Facebook ad within a 28-day window. When using a 28-day click attribution window, if a user clicked an ad but left your website and later came back to purchase 8 days later, you would receive attribution for this conversion in Facebook’s Ad Manager.
On January 19th, many brands woke up to suspiciously high cost per purchases and lower ROAS. It turns out that on the 19th, Facebook had finally begun to roll out their new default window and removed the 28-day click, 1-day view default window from a large portion of accounts. In the near future, all advertisers will no longer be able to track users outside a 7-day click, 1-day view attribution window. If a user clicks your ad and purchases 8 days later that conversion will not be reported.
So what does this mean for advertisers like you and I? For starters, it means we’ll have to start looking at attribution differently, as we understand our Facebook attributable reporting is giving us an even smaller picture of what performance actually looks like. KPIs will have to be adjusted in order to optimize, but the good news is we’re not flying blind. As Facebook historically gave advertisers the ability to view conversions from all offered attribution windows, we can use the historical difference in reported attribution between windows to set strong and informed new KPIs.
Below is a look at the increase in cost per purchase for smaller attribution windows across 10 anonymous direct to consumer ecommerce brands from Q4 of 2020. All of the following brands spend the majority of their budget in North America and I’ve rounded up spend to the nearest one hundredth thousandth. Column 3 shows the increase in CPA when advertisers switch from a 28-day click, 1-day view to a 7-day click, 1-day view window. Column 4 shows the increase in switching from 7-day click, 1-day view to 7-day click window, and the last column shows the increase when switching from a 7-day click to a 1-day click window.
As shown above the biggest decrease in attributable performance is not when advertisers switch from a 28DC|1DV window to a 7DC|1DV - but rather when view through conversions cease to be included in attribution. While Facebook’s new default window of 7-day click does not include view through attributions, advertisers will still have the ability to select a 7-day click, 1-day view attribution window after the iOS14 update is in effect.
This loss of attributable performance when excluding view through attribution is even more evident when looking specifically at subscription brands. Below is a look at the increase in cost per purchase across attribution windows for a specific set of subscription brands at varying spend levels. Once again the below brands focus the majority of their spend in North American and spends have been rounded to the nearest one hundredth thousandth.
While it would be convenient to see a correlation in the increase of CPA across brands, as displayed above the increase is very specific to each brand. Rather than there being a common average in loss of attribution for DTC brands, the severity of loss seems to predominantly revolve around the average consideration time each brand's average customer takes before pulling the trigger and buying the product.
For example, a consumer deciding to buy a $3,000 piece of furniture is going to take a significantly longer amount of time to convert after seeing your ad for the first time, than a user who’s considering a purchase of $100 shoes. For brands with those users who take longer to convert, the loss in attribution will prove to be more detrimental in determining ad effectiveness.
While it’s fascinating to look from a higher level at the impact of attribution on CPA across brands, ultimately as I make the decisions on future CPA & ROAS goals for my clients I’m looking mostly at their own historical attribution data. Every brand in Ads Manager has access to their historical attribution data by utilizing the “Customize Columns” and “Comparing Windows” features which will allow you to look at attribution for every available attribution window. The percentage change from window to window can serve as a guide for the percentage that CPA goals should increase and ROAS goals should decrease.
Despite these ongoing changes, one thing is for certain: we’re losing a significant level of Facebook attribution. It’s time to get ahead of the curve and pull your historical attribution data to determine your new goals.
At Thesis, we’ve developed a testing methodology that generates more learnings, quicker, while also protecting our core scaling campaigns from creative flops. We’ll show you exactly how we set up these creative tests and generate more creative wins for our clients.
A detailed look at how Ad Relevance Rankings correlate with actual performance.