Take Advantage of News Feed Average Position Data
August 22, 2012
Facebook recently released an exciting new data point showing the average position in which your ad appeared in the news feed. News feed-eligible ad types include Page Post Ads and Sponsored Stories targeted for news feed desktop and mobile placements. The default includes all placements (and can be changed within Power Editor).
So why does this matter? Much like search engine marketing optimization, you now have an additional data point to determine how to optimize your ads and exactly what position is best for your goals. For example, search engine marketers long ago discovered that contrary to popular belief, it doesn’t always pay to be number one in search rankings. In fact, frequently, ad positions 3 and 4 at the top right of the search results page generate the best value in terms of volume and cost per click (CPC) without the premium pricing it often requires to win the number 1 slot.
This is unlikely the same case in Facebook because being in the news feed is a prime position for an ad. What’s important is that you now have the access to data to analyze the relationships between your cost per click (CPC), click through rate (CTR), volume, and average position.
First, you’ll need to know where to find this data.
Under your Reports tab is a new report called “News Feed.” API partners also have access to an “Avg Position” field. Take a look at the News Feed Impressions, Clicks, and Average Position. From here, you can calculate your news feed specific click through rate (CTR).
We took a look at 3 data sets to determine if there is a correlation between CTR to avg. position and found that in there is, in fact, a distinct relationship.
Facebook’s ad rank algorithm is similar to Google AdWords and is based on an eCPM auction. Therefore, the higher your effective CPM (CTR x max bid CPC), the better chance you have of ranking higher against other advertisers.
Here you can see the correlation in effect from a data set which includes sponsored stories with like and page post ads:
eCPM = CTR x CPC. As CTR increased, we can see that the average position achieved is closer to 1. Several outliers exist, but the cluster of data from positions 1.6 to 3.7 are indicative of a clear trend.
In a second data set, we isolated the CTRs and average positions specifically for Page Post ads that were run only on Mobile placements. Here we see a bit more noise that could be attributed to a wider range of CPCs, but the general trend still shows better average position with higher CTRs.
In this 3rd data set, we are achieving a vast majority rank of 1 for a set of sponsored stories where our CTRs are exceptional and our eCPMs are very high – winning the bid auction for most ads.
The takeaway here is that our ads are saturating the top position for our targeted audience. Our next step here could be to see if we are over-paying for our ad position by bidding too heavily. If we can reduce our CPCs and test targeting several different ad positions, we may achieve greater cost effectiveness.
If we are looking for volume, we can see that we are already at the top position, and achieving greater volume cannot be achieved through a better position.
The key takeaway is to discover performance distinctions that lead to opportunities to optimize.
As an advertiser, some questions you can begin to answer are:
- How do I rank against my competitors?
- Am I at the best position for my volume goals?
- Am I at the best position for my performance (cost effectiveness) goals?
- If I am only at position 1, am I overpricing my ads? You can test this by testing reduced CPCs and determining the effects of volume and resulting cost towards your goals.
News Feed ads are shown in prime real estate directly in front of your audience. News Feed ads for desktop placements demonstrate CTRs from 2-20x better than equivalent ads that appear on the right side bar. Mobile ads in the news feed also show 6-8x CTR. Regardless of position, are you utilizing news feed ads?
This article was originally published on Fbppc.com.