Episode 9: Facebook Ad Split Testing
Facebook ad split testing just got much easier with Facebook's new tool for testing elements of your ad right from within Facebook's first party tools.
00:02 Stewart Anderson: Welcome back to the "Spend $10K a Day" Podcast, I'm Stewart Anderson. With me, as always, my good friend, Steve Weiss today. We're gonna be talking to you a little bit about a new tool out in your Facebook ads manager or business manager, the Facebook ad split testing tool. Steve, tell me a little bit about what this tool does, how people can use it.
00:23 Steve Weiss: So, first let me just talk about how split testing works. So, obviously you wanna A/B test creatives to one audience on Facebook. You wanna know which specific creative works to which audience. And back in the day, before this split testing tool came out, whenever you had multiple ads in one ad set, AKA, you can only target one audience in one ad set, Facebook would just send all the traffic to the one ad that's actually converting and getting the most engagement. So you actually never really got traction if you have multiple ads in one ad set.
Now, with Facebook's new split testing tool, you could actually do weighted tests. So you could actually say that, "I want 25% of traffic, I want... " You can't actually... You can't go that granular, but you could say you want specific amounts of traffic to each creative in the ad set, which is game changing because if you wanna test out messages and creatives, and you wanna actually give each ad a shot, this is how you do it. Facebook is finally working with marketers to actually give them the ability to do split testing inside of ad sets.
01:29 SA: Yeah. This is gonna save people a lot of time. There's been entire third party tools that people have used to try and streamline this process, make it easier, but the fact that Facebook is finally doing this in their first party tools and allowing marketers to have access to that, it's a real game changer in terms of work flow. It's gonna give you better data, too, it makes things easier.
01:48 SW: Yeah. So just something to be aware of, you can only use this on three different conversion objectives. We don't have a lot of data back on it yet, we literally just got access to it a couple weeks ago or a week ago, and you can only use it on website conversions, lead ads, and mobile app installs...
02:03 SA: That makes sense.
02:04 SW: So you can't actually use this tool for page post engagements, video views, or anything top-of-funnel, so just remember, this is purely a conversion tool that most likely will be used on mobile newsfeed, as well as mobile audience network, but highly recommend that you start... If you haven't already, just start playing around with it. Now you just take an ad set and just upload maybe like 10 ads in there for a specific audience you wanna test, really start trying to find those winner ads to specific audiences.
Never before can you find winners so easily now on Facebook. I highly recommend you not be complacent on this platform 'cause if you're complacent, you're not aggressive, you're not gonna win.
02:48 SA: Yeah. So I think this brings up an interesting question, Steve, which is, do you think this tool, short term, long term, is gonna be more valuable, and make a bigger impact for people who are newer to Facebook ads or people that have been using the platform a lot? Is this gonna be a tool that the experienced Facebook marketer has been yearning for, it's gonna save them a lotta time? Is that gonna have a bigger impact? Or is it going to be just a game changing tool for people who aren't as experienced and need to figure things out more easily?
03:19 SW: That's a really good question. I think that it's gonna be used from both parties. I think that people who just start off on the Facebook platform, they're gonna wanna see... They're gonna wanna try and understand, at least, which ads are resonating with which audience, and if they have any background in research and they wanna understand, they're gonna spend the time and effort to really understand how split testing works.
03:39 SA: Yeah.
03:40 SW: Split testing is a marketing buzzword for doing weighted variable tests for different types of advertising creatives.
03:48 SA: Yeah.
03:49 SW: It's very simple. I hope you'd wanna use the word "split testing" as a very buzzy word because it's really simplistic in nature. As a marketer, you're always experimenting with different types of messaging, different types of images, different types of videos to find out what's gonna be a winner. And I think with this new split testing tool, experienced marketers are gonna look at it and say, "You know what? I'm gonna start creating research and development ad sets where I'm gonna just throw in 10 or 15 ads and see how each one of 'em resonates to one audience."
And they're gonna have 15 different ad sets running, just figuring out which ads are working to which audience, then turn off all the dud ads. And I think this is gonna allow Facebook to actually make more money from advertisers.
04:36 SA: Yeah. Absolutely.
04:37 SW: And really allow people to spend more money efficiently. I think it's win-win for both the advertiser as well as Facebook to keep serving the highest, most engaging, most interesting content on the newsfeed.
04:49 SA: Yeah. I completely agree. I think this is the first step... Or maybe not the first step, but it's definitely a big step in making this... Making the backend for Facebook ads more user-friendly, and easier for people to get better results. We just talked a little bit about who you think might get a bigger impact for it, but I think... Feel free to disagree with this, I think one of the things that I'm thinking about, as I think about who can use this tool really well, is I think if you're a newer business, and you're doing things like app installs, and lead gen, this tool's gonna be really, really great for you 'cause you're gonna be using those campaign objectives anyway. If you're doing a lead ads campaign, you'd do a lead ads campaign; if you're a mobile app, mobile app installs really the main campaign that you're gonna be running.
I think when you talk about website conversions, I think that aspect of it, I think it's gonna be more useful obviously for businesses that have probably a higher span and a bigger audience, obviously 'cause you wanna be able to feed that pixel more and more data, you're probably gonna be able to utilize that aspect of it a little bit more.
05:52 SW: Yeah. I totally agree. Let's talk about how to actually leverage this tool. In practicality I think that when people do split testing, they wanna go a little crazy with it. They're like, "I'm gonna test this, I'm gonna test this, I'm gonna test this, I'm gonna test this."
06:04 SA: Yeah, over-test.
06:06 SW: They're over-testing, so they're just going crazy, and what usually happens is you don't actually compile any real data that's statistically relevant, so you don't really have any learnings, you just have little bits and pieces of, "Alright, well this worked at $20, this worked at $40 in spend. So I think Stu, you'd probably agree with me here that, highly recommend that you don't over-test. Only do two or three tests, two or three ads per ad set, with weighted variables, and then start thinking about spending enough money in that ad set to get data that's statistically relevant.
If you have the budget, try and get that data as soon as possible. Maybe you put $500 a day for two days, just to get statistically relevant data, instead of putting $40 a day for three weeks. I highly encourage people after the holiday season to really, after all this craziness is over, really use the split testing tool and try and get data back that you could actually use...
07:10 SA: I completely agree. I think the inclination, and I've seen a lot of, my past consulting clients and businesses I've worked with. I go in and I look in tools like AdEspresso that have had this baked in as their third party offering. And you see people that create campaigns and they wanna test 32 different variations 'cause they wanna do four different creatives, and four different headlines, and then two different versions of the copy, let's say, or something like that. That's a lot of variation, and then you're like, "Yeah, we're gonna do a $50 budget." It's like, "Cool, you're gonna spend like a dollar and a half per ad."
07:48 SA: You're not gonna get data on that. It's not gonna tell you anything, so...
07:52 SW: One other interesting point that I don't think anyone's covered. Whenever you launch tools, they make a lot of the preferred marketing developers, like AdEspresso, and Navigants, and all these other tools that people use. Facebook is really trying. I feel like they're trying to compete with a lot of these tools which is weird because AdEspresso, those guys are awesome, and they have the ability to split test multiple variations. Now with Facebook really iterating the split testing tool, they're really trying to compete in a way, with AdEspresso. They're saying, "You're a great partner of ours, but we're gonna take some of your thoughts, some of your tools, and add them on our platform."
08:35 SA: Yeah, that's the thing is whenever you have an ecosystem and you have these third-party vendors, they'd come in and they do the ad management spot in this case in a better way than is available on a first-party platform, inevitably, that first-party platform, in this case, Facebook, is always gonna look at that and say, "Well, those are the next features I'm gonna build in, those are the things we're going to adapt and offer."
Because if Facebook wants people to spend more money, which is obviously what they want, the best way for them to do that is to build in features that now allow all these advertisers to not have to spend hundreds or thousands of dollars on other tools, they can put that in their ad budget.
09:17 SW: Well, they get direct people using Power Editor and Ads Manager, but it's really interesting how the relationship I'd say between these third-party... You did a whole podcast on just the relationship between these third-party developers and Facebook. I could imagine if I were a third-party developer on top of Facebook's algorithm, I'd be very worried and scared right now knowing that if you build out really cool features and you're innovative, they're watching you. They are gonna build out those features too because it's a battle for market share right now.
09:56 SA: Definitely, definitely. So again, great new tool coming out from Facebook. Split testing allows you to do a lot of really creative things, and test ads in a way that really wasn't available in an easy way within Facebook's first-party offerings, Business Manager, Ads Manager, anything. Power Editor. If you have any questions about it, feel free to email us, firstname.lastname@example.org, you can email Steve, or myself, as well. Our personal email addresses. We're gonna be taking a really, really close look at this tool and figuring out the best ways to make use of it. We'll create some more content for you guys, but other than that, we'll see you next time on the podcast.
10:36 SW: Yup, take it easy, have an awesome New Year.
10:39 SA: Happy Holidays.