When Should You Serve Facebook Ads. AKA We Tried to Outsmart Facebook
We’ve written before about how you should send your emails at specific times of the day to get the most out of them, but we recently decided to run a test on whether there is a best time to advertise online. It seems obvious that there would be an advantage to send an email at a specific time - people aren’t usually checking their email at 2 AM - but we wanted to see if timing created differences in a world where prices are usually determined by an auction (of one form or another).
A quick aside. In this article we’re mostly going to look at three major ways to measure performance and cost of our ads. CPM (Cost-Per-Mille), CPC (Cost-Per-Click), and CTR (Click-Through-Rate). If you feel like you need a refresher on these terms take a quick read through of our crash course on CPM, CPC, and CPA.
We love Facebook ads here at FranchiseHelp. They’re a great tool to use to reach a very specific audience with a more narrative message than you can usually accomplish with search ads. But beyond that we love the details you can get out of the data and the amount you can tweak your campaigns to take advantage of all sorts of small details. That’s why we figured that Facebook would be the perfect place to run this experiment. We created a set of ten different ads and then ran each ad in 24 different adsets, one for each hour of the day. We let these ads run for a full month and believe it or not, a best time started to emerge.
Well really it was two best times. In the graph above the blue line is the CPM, or the cost of getting our ads in front of users. The grey line is the CTR, how frequently users clicked to our site. The CPM moved fairly cyclically throughout the day, with a big peak around 4 AM, while the CTR stayed pretty much consistent throughout the day (except for a weird anomaly around 5 AM). Because of this the CPC, the cost to actually get one more visitor to FranchiseHelp, peaked at 4 AM and then dipped to a low in the mid morning and had a second low in the evening. These low points are pretty interesting. They’re right when people are getting up and getting ready to go to work in the morning, and then again right after people have gotten home from work. In between it’s not nearly as cheap to acquire more traffic, maybe because people aren’t browsing Facebook as much at work as we think they are.
Looking at this data got us thinking, was this just a one time result because of this test or are there actually better times of day to advertise on Facebook? So we decided to take a look at every single ad we ran on Facebook this year, hour by hour, to see if there was a difference.
We didn’t see exactly the same pattern. The hour by hour spikes went away as we looked at a much broader set of adspend data. But there was still a major decrease in costs in the early evening.
So what does this all mean? We’re not going to suddenly stop advertising throughout the middle of the day and only try to find new users in the evening. But we might change up some of our ads earlier in the day to see if there are ways to attract more of the users who are on Facebook then. If you have any questions about how you can use this data to spend your marketing budget more efficiently, or have any questions about Franchise Lead Generation in general feel free to shoot us an email.
If you’re interested in learning more about FranchiseHelp’s lead generation programs, click here to learn more.
Browser Evolution 2014-2016: A Silent Killer
Over the course of the last few months, we’ve started to pick up our coverage of website design and actually capturing contact information from people visiting your site. You can check out the two most recent articles here:
My best email subject line this year (sort of)
It is nearly impossible to over-state the importance of an email's subject line. In my days as a franchise marketer, I've written well into the thousands of subject lines.
Call Verification: Here’s exactly how we do it (for now)
In case you’re already scratching your head a bit, I want to speak a little bit about “call verification” as a concept.