Transcription
Hello and welcome back to another episode of the Collective Clicks podcast. This is your host, Brandon Baitman, and today I'm joined by Garett Craigan, our Director of PT Media. We're going to talk all about Facebook ad creative testing. What is our process for that right now? We've simplified everything to a two-step process that helps get both the most leads and then also the highest quality leads. And we'll talk about how that works. How are you doing today, Garrett?
Doing good. How are you?
Fantastic. You might notice my, my voice is deeper, more, more booming, more podcast-like than, uh, than usual. And, uh, it's because it's 95% gone. So I'm gonna make it through this podcast, but probably not much more.
I like it. A voice for radio.
That's right. Um, for the first time ever in my life. Hey, I like it better than face for radio.
Fair enough. Yeah. Well, I think there's a reason we get a lot more viewership on our podcast than we do on YouTube.
I can't. Fair enough. Uh, don't make me laugh because I can't actually laugh in this current state, so it just sounds awkward. But yeah, let's, let's jump into the topic for today. Um, maybe this is just the, the, the blessing that you need to actually get a some room to say more things during this episode versus me just finally blabbering the whole time. Um, I'm, I'm super excited about, uh, this because this is actually one of the, the biggest improvements that we made to the way, way that we run Facebook ads in the past, what, like six months or so? I don't know exactly how long we've been doing this. Um, and I think it's whenever something like makes a lot of sense and then also works. Those are the things that make me happy. Lots of things work, but they don't make sense, which is just kind of annoying. And then lots of things make sense and then they don't actually work, which is even worse. So this is awesome. And, and I'm super excited to dig into it. And this is our two-step ad creative testing process, formerly three-step. We've actually simplified it even more. Um, and we're going to talk about basically, how do you test creative assets in Facebook? Turns out not a lot of people focus a ton on this in this industry, but it has some massive implications. And then there's different advantages to different types of ads that you can run. Um, so without, without getting too deep into it, I'm just going to turn the time over to you, Garrett. If you could share, uh, like step one of the, of the ad creative testing process, or an overview of what we're talking about.
Yeah. Um, so our ad strategy is kind of based on two different ad types. Uh, one being dynamic ads, and then two being a static ad. Um, not to, you know, go too deep into an obvious thing, but, uh, a dynamic ad is basically where we feed Facebook, um, a bunch of different headlines, images, videos, descriptions, and it's going to run the best combinations that help you achieve the goal that you set at that ad set. So it's go, going to test a bunch of different assets and then find which ones give you the best, you know, cost per lead or lead volume or whatever the goal you give it. Um, and the benefit of that is it tests a lot of things quickly. Um, and then the other ad type that we run is a static ad. Um, and the benefit of of that is that we're able to identify exactly which part of, of the ad was shown, and we're able to trace that back to our CRM to see how that asset correlated with lead quality. Um, the, the challenge with dynamic is you don't know which asset was shown, um, within, uh, your like offline data to know how it impacted quality. So that's why we have our, our, um, campaigns in that two-step process because it helps us to balance the, the fast testing of dynamic ads with the, uh, quality, uh, implications and quality tracking that comes from static ads.
Yeah, totally understood, Garrett. And I'm going to add just a little bit of color here in case somebody's heard some different vocabulary to explain these things and also explain a little bit how they work. Um, so dynamic ads, you may have read about something that people call DCO or dynamic creative optimization in Facebook. I, I, just making the connection if you've been reading up on Facebook outside of this, like same thing, just different words for the same thing that we're talking about. Um, and I want to make it really clear exactly why dynamic ads are so, so cool. Like you talked about the benefit, but just think about it for a minute. What if you have three headlines that you want to test, and you have three ad copy options, and you have three options for videos or photos, and then you have three call to actions, right? Just, just picture like all these things. If you wanted to actually test that in what we call multivariate testing, where you're kind of seeing like each combination between those things, you'd have to make a ton of ads, right? Because you're multiplying three times three times three times three, which is completely impractical. You're not going to create, I haven't even done the math, whatever three to the fourth power is, you're not going to create that many ads in your account and test them against each other. It's going to just murder Facebook, Facebook's algorithm, you're never going to get good data from it. Um, and, and dynamic creative actually can do that exact thing. So it's, it's super powerful in that sense because you can test so much stuff at a time versus if we're testing static ads, just like, oh, let's just change this one thing on the ad and let's see how that does, which is is cool, but like, you don't make progress nearly as fast. So a lot of people don't realize how, how strong that is with dynamic creative optimization. Like you can make a lot, a lot of progress really, really fast, which is why we love it. Um, but, uh, so, so explain like specifically why instead of just using one or the other, do you use both? I know you touched on it a little bit, but let's talk about like specifically, or, or maybe let's walk through it sequentially, right? So, so we start out, right, step one is we have this dynamic creative optimization. We're feeding it a lot of different stuff and kind of seeing, um, in a way, this is a bad way to say it, it's kind of like throwing a bunch of stuff at a wall and sort of seeing what sticks because you have the ability to throw more stuff at the wall with dynamic creative optimization. Um, but why do we do that first? And, uh, and what can we learn from that? And then how do we take that to the next step?
Yeah. So the first thing is like, if, if, if an account has enough budget behind it, it could just have all of of this run through purely dynamic because you could, in theory, have the ad set optimized for a qualified lead action. Um, in general, the bulk of of our clients don't have the budget necessary to do that. So this is our way of like being able to optimize for quality without having to have a massive budget to have enough, uh, leads come through that are qualified to give, uh, Facebook enough of a sample size to optimize towards a, a lower funnel action. So, um, we first run, um, dynamic because it, um, helps us to escalate and like and, and, um, and speed up our, our testing process of a ton of different assets. And we, we are basically always adding in new copy, new images, new everything into that that ad, so that we're always, um, finding new best performers. Um, and then as we find assets, be it an image, a video, or text, that's that's doing well, that's then put into our static ad, um, um, pool, I guess you could call it, where then we're able to to get dialed in at the quality level, how each of those, um, those elements are, are, uh, um, are impacting quality. So we start with a wide range of assets and get that that dialed into like five or six, um, assets through our dynamic ad. Then we dump those, those best five or six into a couple of, of static ads, and then we dial it in to a quality level once we've already found which, um, which assets give us the best, uh, um, cost per lead or, uh, impressions per lead, any of, of those big metrics.
Yeah, totally agreed. I can explain this similar, similar way, different words, just in case somebody needs some help following Garrett's genius here. One of the, one of the things that's, uh, bad about static ads is if you make bad ones, it can hurt you because you could spend a lot of money on them. That's what's great about dynamic is it optimizes really fast. Um, so static ads, a lot of people think they don't work really well. I think what we found is they work really well if you only have really good ads there. So this is sort of a process of finding what the really good performing creative is, and then making static ads with that good performing creative, so we don't accidentally spend money on now poor performing creative. But the other drawback of dynamic ads is attaching a specific asset saying, the the person clicked on the, that had this image in it, and because it had that image, then we ended up with this lead quality. For example, is, as far as I'm aware, not really possible. Um, very, very difficult. I haven't found anybody who knows how to do that, right? So that's where like, we, we've noticed a lot of our competitors that are running Facebook ads, because dynamic ads are working better generally when you have less proven creative, they're using dynamic ads, but then they stop there. So then they're left with just whatever ads get them the best cost per lead, not necessarily the best lead quality. Whereas what you can do is you can then take those best performing assets and then you can track everything full circle, right? So we're taking those five or six, and with every one of those ads, we're tracking how many net leads come from them, how many opportunities, how many contracts, how many deals, what the revenue is associated with them. And we start to learn things like, if we use this image instead of that image, it increases the number of leads that we get that match this certain profile by this amount. And then we can dial more and more in on quality because Facebook's not going to understand the quality unless you're optimizing for quality, which most companies in this industry can't do based on budgets, and even if they had the budget, they don't have the market size to do it in many circumstances. So anyways, it's a, it's a super cool process. I hope you guys are like catching the, catching the vision a little bit talking about this, um, in the sense that you can narrow down the creative significantly with the dynamic portion, um, and then you can have only good quality static ads, and then you get dialed in in quality and you start to know which ones are working better from a quality standpoint, which ones aren't, and then you can get rid of the ones that aren't, even if they have a good cost per lead. This is why this is one of the secrets to having great lead quality on Facebook. Turns out what you attract people with, because Facebook, these these ads are seen by tons of people. I was looking the other day, we have, we had on on Facebook close to 118 million impressions on our ads last year. Some of those were Google, but a really small sliver. Facebook is a massive amount of that, despite being a smaller portion of the spend, because tons of people see the ad. That's really cheap, right? You pay like 40 cents to send a postcard, you might pay like one or two pennies to show a Facebook ad to somebody. So it's cheap, but the way that you qualify with the ad drastically affects what people you're going to get. Because the targeting is targeting all of them, most likely what the ad says and how it speaks to them is what gets the right people or the wrong people to reach out to you and has a massive, massive impact.
Yep. Um, one thing that I don't think we covered that is helpful to be aware of is how we, we can trace the, um, how the static ad, uh, impacted lead quality. And what we do is each ad combination has a, a, a unique, um, ad name that tells us which headline, which description, and which image is in that static ad. And then that whole ad name is pushed into our CRM. So that's how we're able to, um, tie back that attribution is, is, is from the name of the ad. Um, people don't always know that you actually can make ad name templates in Facebook. So you can have it like pull like the, the description and the image, um, and, and then make the ad name just, just like, just have those two elements with a, a dash between them, you know, that that kind of thing. So that's how you can get those attributes pulled back into your CRM is by making a custom name for each of your, uh, asset combinations. And then we have some sort of database where we basically have the names and then what the ad is basically, so we're able to cross-reference it and understand, yep, exactly how it affects the quality. Um, which was a little tricky to set up, but works like a charm once you got it set up.
Okay, that's awesome. Well, thank you for, thank you for joining me for this episode, Garrett. Short and sweet, talking about the biggest things. Um, I hope, I hope you guys are seeing the value in this, specifically this episode and the last one, if you're doing really well at, uh, tracking the right kinds of conversions and you do really good with qualifying properly with the ad creative, so you make good guesses, but then you're also able to track the quality to dial it in. This is the way that you get the number of leads per contract that you need on Facebook, which for the vast majority of people that I talk to in this industry that, uh, that don't like Facebook, it's usually because of the number of leads that it takes to get deals done. And these kinds of things could be a game changer. So, uh, if you, if you like this episode, if you saw value, definitely, um, leave us a review on Apple Podcasts or Spotify or wherever you listen to it. Share this with a friend that you think might be running Facebook ads. Uh, I'm, I'm serious about this. When you see those people in Facebook groups or something that are talking about how much they hate this marketing channel or something, share this and, and talk about, you know, did you try these things or did you not try these things? And I guarantee you, most of the people who are not having success, just are not checking the boxes that you need to check to have success with this channel. Um, so anyways, super grateful for your time, Garrett. And to everybody else, I'll see you next week. Thanks.