The Metrics of Success in Facebook Advertising
Published: October 23, 2013
Author: Franco Puetz
I’m sure you’ve heard the news of yet another revision to Facebook’s ad creation interface – all designed with the hope of making it easier for Facebook advertisers (and users who want to take a stab at Facebook advertising) to reach their target audience.
So what’s Facebook’s incentive to continually make it easier for users to create ads on their platform? That’s easy – an increase in advertisers generally correlates with the number of ads served on the platform. If you consider the fact that there are a set number of ads that can be shown at any given time across the platform, you can relate all of this back to economics 101. If there is a larger demand for the impression space (and a limited supply), Facebook will raise the prices of these ad impressions/clicks to acquire a higher margin. Hopefully you weren’t under the impression that these interface updates were coming straight from the kindness of Facebook’s heart. 😉
All of this considered, I figured it would be a good idea to touch base with other Facebook advertisers on the topic of what metrics we should be tracking and some of the more common approaches to measuring campaign performance. We can expect the cost of Facebook ads to increase over time. Whether or not this recent update will have a noticeable impact is still to be determined, but knowing how to optimize our campaigns will help us stay well-equipped to tackle an influx of demand for impression space while simultaneously keeping our campaigns affordable and effective.
Cost Per Conversion (CPA)
Paying attention to how much you are spending on your ad campaign is critical for most business objectives because they allow you to relate your costs back to a return on your investment. Even more important in campaign management is being able to tie those costs back to a particular conversion. Some of the more common conversions seen on Facebook are page likes, user engagement, leads, or purchases. Facebook makes the process of measuring all of this information incredibly easy to track through the interface, and the conversion metric seems to be sufficient for a number of advertisers.
Unfortunately, if you’re measuring based on this metric alone, you could really be working your way into more complex issues down the line. The biggest issue that comes to mind is volume. I can almost guarantee you will see a drop in CPA if you go into your account and split your bid in half on your ads, but doing so will likely take toll on the volume produced.
Click Through Rate (CTR)
How clickable are your ads? This is a great metric to follow if you are simply trying to drive clicks to your ads, but for most advertisers it’s what happens after the click that really matters. If you want a high CTR, give something away for free – consumers love freebies. This is a dangerous metric to rely too heavily on if you are bidding on a CPC model. Having a high CTR is a great thing, but if you can’t deliver on the promise made in your ad creative on your landing page, this high CTR could bring your campaign to a screeching halt. This brings us to one of my favorite metrics, the conversion rate.
Conversion Rate (CR)
What is the likelihood that a user will complete the desired action after engaging with your ad? This is determined by the conversion rate. A good way to ensure you have an effective conversion rate is to be proactive in your efforts to enhance the continuity between your ad and your landing page. If the messaging doesn’t line up between the two, it will be much harder to trigger the conversion. The landing page will be a huge proponent of acquiring an effective conversion rate, but you can optimize your conversion rates even further by being explicit in your creative as to what the user can expect upon clicking. If you are bidding on a CPC basis, a higher conversion rate it slightly more valuable than a higher CTR, due to the fact that an increase in clicks will cost you more, whereas an increase in conversion rate will yield a higher income at the same cost. Always make sure you know the reasoning behind why you are optimizing for which metrics.
Conversions Pre Impressions (CPI)
This is my preferred KPI for any display campaign I’ve run on the web, and it is an incredibly powerful way to measure the performance of your Facebook advertising campaigns. This metric can be found by simply dividing your conversions by the number of impressions served. Alternatively you can pull the same metric by multiplying your CTR and CR together. This is not only taking into account how clickable your ads are, or a user’s propensity to convert after clicking your ad – this metric is literally telling you how many conversions you can expect per impression shown on Facebook. If you have a higher CPI and an audience that is scalable, you’ve struck a Facebook advertising gold mine.
Now I’m sure many of you are concerned about the lack of cost information included in this metric, but in my opinion this cost obsession should be secondary when optimizing your ads. You can always run back through your campaigns to adjust your bidding preferences, but finding a high CPI is much more challenging. Wouldn’t you prefer to share your ad with users who are more likely to convert, than to users who may be captured at a lower cost per click but are lacking in the conversion department? A lot of cheap clicks can add up over the long run, and optimizing on a cost basis could put your account health in jeopardy over the long run.
Hopefully this information has cleared up much of the ambiguity surrounding Facebook ad campaign optimization and performance measurement. I understand that everyone operates in different niche categories that will require unique approaches to performance measurement – my hope is that this post has opened up a few new ideas for you to consider the next time you’re mining through all of your extensive ad data. If you have any specific questions regarding your unique ad campaigns, feel free to reach out to me. Good luck on your next round of measurement and optimization!