There have been some very bold statements made about how well one’s ads can perform via FBX, and, as marketers who have made use of it from the beginning, we have found there to be varying degrees of both truth and myth in some of the statements made about FBX.
What we love about FBX is that it provides marketers an opportunity to retarget users in a safe place. It also provides a LOT of inventory. With One TRILLION page views per month, this is, theoretically, even more than all other display exchanges combined.
Truth or Myth?
Here are four things we keep hearing about FBX:
- FBX has the cheapest CPMs ever seen
- FBX has really high CTRs
- FBX has really high conversion rates
- FBX generates really low CPAs
Since the start of the Acquisio Trading Desk, we’ve run more than 40 Facebook retargeting campaigns. We have done it over two continents, in 15 countries, and in most verticals. For these purposes, we selected three that would outline a variety of different goals and executions.
The first campaign is a very large ecommerce site similar to a Walmart or Zellers. The campaign was geo-executed in Russia. The strategy used was to retarget side-wide visitors, shopping cart abandoners, and converted users. The goal: generate as many orders as possible at a Cost per Order (CPO) of less than $11.
What we found in this campaign is that we had a huge number of impressions, with CTRs of between 0.05% and 0.26%. Conversion rates for site-wide visitors were only 1.3% with a CPA of $209.60 (not good!), but the shopping cart retargeting CVR came in at 8.3% with a CPA of $7.60.
While the retargeting of site-wide visitors was a disaster, the retargeting of those who had abandoned their shopping carts was fantastic, leading us to segment the campaign and target only those visitors.
The second is a sport clothing retailer in the UK. The strategy was to use retargeting only: FBX retargeting, site-wide retargeting on display exchanges, and first-party search retargeting on display exchanges. The goal was a $25 CPO.
FBX did have a lower CPM – almost three times lower than the other two methods on display exchanges. CTR across all channels was very low, from 0.06% for FBX to 0.10% for combined display retargeting efforts. In the end, CPO was lowest on both search ad retargeting at $20.50, and on FBX at $21.00.
With a better creative strategy, this campaign may have seen the CTR essential to growing an audience. While these levels were acceptable to the client, we still feel there would be more to gain from each of these channels and, given that the CPM was lowest through FBX, this would be an attractive channel to promote this product through.
For a shopping comparison website where conversions are actually a click to the affiliate offer, we attempted a target CPA of $0.52 through pure FBX retargeting, concentrating only on high-value products.
Again, limited creative was available. CPM was quite low at $0.83, with CTR at 0.07%. However, of that number, we had a great conversion rate of 33.9%, but still ended up with a CPA of $3.30.
Although CPM was very low, the CTRs on this campaign were well below what we normally see on other channels. And while the conversion rate of 33% looks quite good on paper, it is still half of what the client receives through other display channels.
So…Truth or Myth?
From our data, here’s what we can say:
FBX has the cheapest CPMs ever seen: TRUE.
Very low CPM helped to bring down the overall cost of the campaigns and, were it higher, we would have seen very different CPA results. Cheaper than what we normally see on RTB for display.
FBX has high CTRs: MYTH.
We saw the same results as most Facebook campaigns. Extremely low CTR as compared to display.
FBX has high conversion rates: MYTH.
Except in the last case, our conversion rates ranged from not bad to extremely low when compared to the conversion rates we obtain from retargeting on display exchanges.
FBX generates lower CPAs: SAME.
What we saw is that this tends to be about the same as other types of display retargeting campaigns.
There are methods that we used to help improve every aspects of these campaigns and we have conclusions and recommendations based on this data that we will share in Part 2.
For the complete data, please come see my presentation on this topic at SMX West on March 11 and get the full picture and learn more about the what is, and what is not possible through FBX campaigns.
- Marc Poirier