Introduction
If you’re starting out with Facebook Ads for the first time or launching new campaigns with different objectives, it can be very hard to know what your budgets should reasonably look like.
Facebook’s growing popularity means more and more advertisers are using the platform, leading to a squeeze on the available inventory and, in turn, increasing advertising costs. Having a solid understanding of your own data is essential, allowing you to accurately forecast performance or be confident enough to change tactics to deliver on your revenue targets. But what is a good CPC, CTR or CPM these days? We’ve pulled the data to find out.
At Croud, much of our Facebook Ad experience falls within retail, ranging from luxury fashion through to gardening. Using the extensive data at our disposal, we’ve pulled out some benchmarks that should help you glean how your own performance metrics compare. The results are interesting and show how many possible variations must be accounted for when planning campaigns.
*The data is from Q1 2017, averaged and anonymized across all Croud retail clients. All data is based on Facebook’s standard attribution model of 28-day post click, 1-day post view. Let’s dive in.
Average click-through rates
Whilst CTR is a great indicator of how well your ad is resonating with an audience, it’s not necessarily the best indicator of campaign performance. Let’s say you’re running ads with a ‘Brand Awareness’ or ‘Reach’ objective, your KPI should be to make people aware of and be able to recall your ad or brand. Therefore, CTR and clicks are secondary metrics. Of course, if you’re looking to drive Conversions or Leads then CTR is extremely important. It stands to reason that our data set shows the highest CTRs across those objectives which demand a response or action. Conversion campaigns have a CTR of 1.2% and Dynamic Product Ads one of 1.4%, but interestingly the campaign objective which drives the highest CTR is Lead Generation, often used for early bird offers or newsletters.
The average click through rate across all our retailer data was 0.75%.
What can we take from this?
It’s important to remember that your Facebook campaign objective should be aligned with your overall marketing objectives and your reporting be setup to reflect that. CTR is an excellent metric to understanding the impact of your creative’s performance or the ability of your campaign to drive clicks. But it may not always reflect the overall performance of a campaign, such as if your set aim is to speak to as many people as possible with a reach objective. If your CTR is significantly lower than these benchmarks, it can be prudent to conduct a content audit and ensure that all your creative is relevant, exciting and still fresh to an audience that can be quickly fatigued.
Average CPM
Moving on, let’s take a look at one of the more interesting metrics for understanding both market forces and value, namely Cost per thousand impressions or Cost per Mille (CPM). We’ve seen huge fluctuations in CPMs, particularly around large shopping events like Black Friday or Thanksgiving. At certain points in November 2016, we even saw CPMs rise as high as $50. The dust seems to have settled in 2017 and we’re seeing CPMs normalize again. You can see from the data below that the highest CPMs tend to be seen when your aim is for the user to perform an action, or if you’re targeting a granular audience segment. App Engagement CPMs are the most expensive ($11.38) across our data set; a powerful tactic that uses 1st-party data to target certain cohorts of customers. You can see that Brand Awareness has one of the cheapest CPMs at $1.58, we’ll often use video here to target people who are more likely to pay attention to and remember your ad.
The average CPM across all our retailer data is $5.24.
What can we take from this?
Facebook’s algorithms are pretty smart and it might be naive to think we can do a better job optimizing performance ourselves. However, that doesn’t mean you can’t be creative in testing different auction strategies to achieve your end goal. Our data shows us that bids for conversions are 77% higher than for clicks, and in some cases, clients can’t justify the inflated cost. That’s when it may be worth testing different campaign objectives, like clicks. Using a simple framework, like the one below, you can see how a click campaign could effectively outperform a conversion campaign in terms of CPA. 
Average conversion rates
Similar to the above, we see the best conversion rates when campaign objectives target people at the bottom of the funnel, like remarketing or re-engaging current customers. Dynamic product retargeting (9.2%), Conversions (6.85%) and App engagement (5.65%) show strong results across the data. As you move up the funnel to leads, reach or video views the conversation rate is understandably low (under 1% for all).
The average Conversion rate across all our retailer data is 5.62%.
What can we take from this?
Facebook will always tell you to set up your campaigns with the end objective in mind, and 90% of the time we agree. If your aim is to get people to purchase a wooly sweater, it makes sense to optimize for conversions. However, what do you do on the occasions when the cost proves prohibitive and high CPM for Conversions make the campaign unprofitable?
With a few of our lower product margin retailers, we decided to switch up the standard approach and in the weeks leading up to Thanksgiving 2016, tested a few different auction strategies. One that yielded excellent results was using brand awareness and link-click campaigns to drive cheaper traffic. Our objective was to build audience data and retarget users at key buying times, knowing the CVR would be significantly higher. We took insights from Google, Facebook and the client’s 2015 trends to retarget customers with Black Friday offers when they were in purchase mode. For one client, when we combine both the prospecting and retargeting together, we outperformed YoY revenue for November by 73%.
Average costs-per-click
CPC is another interesting metric for tracking value and effectiveness, and it deserves analysis beyond what it first seems to reveal.
Unlike Google Adwords, where the CPC is incredibly important to everything that follows, with Facebook, the weight of its significance very much depends on the campaign objective, as seen below.
If you’re running a Reach or Brand awareness campaign, the objective isn’t to get clicks and the CPC might, therefore, be quite high, as seen in the figures for 'reach' data below. It also depends a lot on the messaging, are you asking the user to click on the ad? What is the specific call to action? Are you using the right ad format for the objective?
The average Cost per click across all our retailer data is $0.70
What can we take from this?
Effectively, the data backs up some of the analysis above, demonstrating that there are cheaper CPCs out there if that is indeed your objective.
Overall, we’d encourage retailers to start thinking of Facebook as a channel where they can deliver full funnel strategies, engaging users at every stage. Facebook has been working hard to factor this full journey into their products, from Canvas ads for awareness to Dynamic Product Ads for conversion. As marketers, this is incredibly exciting; what other channels can deliver that full funnel as effectively? It’s then up to us to use these tools proficiently and most importantly, report on performance correctly.
One thing to approach with caution is the expectation that just throwing some direct response dollars into the platform will deliver a ton of last click sales in Google Analytics (GA). Facebook can be highly effective, but let’s not fool ourselves. If you’re going to use the channel give more thought to the way you’ll attribute activity beyond GA last-click, or you’ll stifle your potential.
Hopefully, some of this data has helped frame your current performance and highlighted some of the ways you can use data to adapt your approach to Facebook advertising.
If you have any questions on the above, or how you might adapt your current Facebook strategy please do get in touch.
