GSD 200: How Do Facebook Ads Work?

Welcome to today’s GSD Daily. Today we’re going to talk about how Facebook ads work.

Now, Facebook ads are notoriously finicky, especially… Well, they have been for the last two years at least, but more so in the last couple of months. Facebook is going through a lot of changes from an algorithm standpoint. The change from Facebook to Meta threw a bunch of shit off. There’s a lot of privacy concern, all kinds of weird shit around Facebook ads. But at the end of the day, they remain one of the best ways of advertising your offer online and getting leads.

Now, part of that is because they collect so much data about every single one of us that they can well whether something is going will us or not. A great example… So it’s Christmastime right now. If you’re watching this video in the future, it won’t be Christmastime, but it’s Christmastime right now. It is the 22nd of December, which means Christmas is in three days.

Now, I did all my Christmas shopping early, and then for whatever reason, about five or six days ago, the algorithm just started showing me cool shit. So I have been buying not a lot, but enough things through Facebook and Instagram, because they continue to show me ideal stuff, and I bought the first thing, and then I saw the second thing. I’m like, “Fuck, I got to have it.” Well, not for me, but for other special people. So I bought the second thing, and then the third thing showed up; I’m like, “Fuck, I…” So now I’m deep into this because they consistently show me things that they know I’m going to like because of the theme of the things. So I don’t want to spoil any surprises. Maybe I’ll do part two of this video. But Facebook is good at knowing what you’re going to buy and what you’re not going to buy.

Advertisers can key into these things, these buying habits, for their gain, and they did very much so on me in the last couple of days. But what I want to do is I want to show you how Facebook ads work and explain how Facebook ads work so that you know when it’s a good fit for your business and when it’s not a good fit for your business.

So the first thing is that Facebook ads are interruption-based. Somebody is scrolling through their phone looking at cat videos and funny pictures and whatever else. They’re not there because of you. They’re not there because of your offer, your product, or your lead magnet. They couldn’t give a shit. They are there so that they can consume content from their friends and family. That’s why they’re on Facebook.

But they do certain things that allow Facebook to key into them on their algorithm. So life events are one: getting married, getting divorced, separated. Facebook knows when you’re looking for a new house, all of those kinds of things. So if there’s a life event, a big momentous life event that is happening in your life, Facebook probably knows about it, just because of the actions that you’re taking.

Another thing is Facebook knows what you’re clicking, what you’re watching, and what you’re seeing, so it can key into those interests. Those things are all categoric, so if it sees that you are clicking on things that are Universal Orlando related, then it’s going to show you more Universal Orlando-related things. That’s just how the algorithm works. So again, as advertisers, you can key in on that stuff.

That is how, in essence, you set up your targeting for your ads. So you have your offer. You have a landing page, a shopping cart, or whatever, and you can say, “I want to show my ad to somebody who likes this, that, this, and this, but not this.” So you can shrink your audience down only to the people who are going to be relevant for you, and then you can further layer things on top of it, like a lookalike audience or a custom audience.

So a lookalike audience is, let’s say you have a thousand buyers in a basket. A thousand customer profiles. You’re going to hand it over to Facebook. Facebook is going to look at each of them and say, “Well, this person likes this, this, and that. This person likes this, this, and this. Okay, those are the same.”

So what it does is it finds the commonalities between your customers and then handpicks 2 million more people just like them and says, “Would you like to target these people too?” That’s called a lookalike audience. And your answer as a business owner is going to be, “Yes, I want a million people just like the thousand people who already bought from me.” So that’s what a lookalike audience is.

Now, a custom audience: a custom audience is that same thousand people. You can upload them as a custom audience and say, “I don’t ever want my ad shown to these people because they’re already customers.” You can upload them and say, “I want this new ad shown only to these people.” It might be an upsell or a special exclusive event or something.

So there are lots of ways that you can take lookalike audiences, custom audiences, and then those demographics, those interests, and mix them and match them in different ways. That’s what Facebook does, really, really well.

Now, what I want to do: we’re going to hop over to the laptop and I’m going to show you one of our campaigns. I’m just going to click through it. Understand that when you’re watching this video, it’s probably going to look different, because Facebook changes at least 34 fucking times a day. So it’s probably going to look a little bit different, but the overall structure is the same as what Facebook looks like now and what it’s looked like in the past. So it’s still campaigns, ad sets, ads, and when you can understand that, when you can break that down and you know where to look for certain things, then the rest of it all makes a lot more sense.

On the screen here, we’re seeing a campaign that we got running. 184 website leads. We’re paying $7 and 24 cents per lead, and we’ve spent $1,300 on this. It’s a solid campaign. We can see some of the age and gender distribution here. So most of our stuff appeals to men. By and large, 45 to 64 is our audience. Above that, not so much. Below that, not so much. So these demographics play out in a bunch of the stuff that we do. And then, if we look at platforms, our audience tends to be mostly on Facebook, and less on Instagram. And also, most of our stuff, you can see here, is desktop. So most of our folks are desktop-based.

Now, if we pull this down, this is the campaign. So this is the overall campaign performance review. You can see this top level here. This is a campaign. And then if we pull down to the ad set here, you can see it changed behind us. So we have campaigns, then we have ad sets, and then we have ads.

I’m just going to look at it this way. So we have two campaigns running right now for this particular campaign. One of them is getting us leads at $6 and 58 cents. The other one is getting us leads at $8 and 44 cents. There is a little bit of a split test going on here. So same audience, but this is just a different type of ad set. I’m just going to go into the ad set here so that we can look through how this is all positioned. Got our campaign view here. We’re going to look at the ad set. The name of this ad set: HubSpot BO Funnels 25 to 60 videos.

Then we have our conversion event, which is a website conversion event looking for a lead. What we’ve had to do lately is we’re looking for clicks or traffic as opposed to conversion events, just because in the holidays it’s been cheaper. This particular campaign wasn’t necessarily effective. We’re still getting cheap leads. So we started this campaign back in October, on October 31st, and then we’ve split-tested it to where you see it. Location-based. We’re targeting people 25 to 60. The language is English. People who are interested in lead generation, HubSpot, and marketing strategy. And then they also must include funnel, behaviors, and business-page admins.

So if we just look at how this is set up, we are targeting the United States, and we could target other countries. The big ones that we target a lot of times are the United States, Canada, the UK, Australia, and New Zealand. Those are the big ones. And then every once in a while, we’ll break the assets or the campaigns down by age. But this one is working well for us. We could just advertise to males and then do a different campaign for females, but again, the overall lead cost is pretty good.

When we break down our targeting, HubSpot has a giant audience. So the person who uses HubSpot is our ideal customer avatar. So somebody who likes HubSpot and also is interested in lead generation and also… I shouldn’t say “and also”, but they are one of these things. So they are interested in HubSpot, or they’re interested in lead generation, or they’re interested in marketing. Then if they’re interested in one of these things, they are put into this audience.

Then we narrow it down by people who are business-page admins, meaning they own a business page, or they are interested in funnels. Now, this one shouldn’t be here. So industrial materials and equipment. This guy should not be here, so I’m going to go ahead and pull this out. So basically, they’re interested in HubSpot or lead generation, and they are also a business-page admin. That puts them into this audience here. So I’m going to go through and update that because we don’t care about people who are looking for funnels. We don’t sell “funnel” funnels, do you know what I mean? It is updating in the background, but in the meantime, what we’re going to do is we’re going to talk about this particular setup.

So basically, as you saw, we start with a big pool of people, all genders between 25 and 60, and then we pull it down to people who are interested in lead generation or HubSpot. So let’s diagram this out on the iPad here.

All right, so we have a big audience of people who are 25 to 60, so that’s Audience One. And then we got male and female, and then of these people, only those who like HubSpot or lead generation, and then of those, only people who are business-page admins. So when you put all of this together, our audience is right around 500,000 people. So 500,000 people are seeing our ads.

Now, there are some ways that we could work with this. I mean, what we could do is we could still have all of this and then put them inside of a circle so that we’re still pulling them down, but that we’re using a lookalike audience. So this is 2 million people just like our customers, and then we go down and pull this same filter through it, so that of the 2 million people just like our customers, they exhibit these same behaviors. And this number might not necessarily be 500. It might be 300,000, but these are uber, uber, uber-qualified people. So that’s one way to look at the audience.

A lot of the time we like to segment it down or we like to expand it because we know Facebook’s algorithm is doing so well at targeting folk. At the end of the day, with $7 and 24 cents lead cost, we’re doing well for this lead magnet, so I’m not worried about the targeting. We know that these are good, solid leads coming through the system, so it’s solid as it is, but I wanted to walk you through that.

Now, the other piece of this is the ad itself. We didn’t necessarily talk about that, but if we’re looking at the ad, we set up what is called dynamic creative. So I recorded three different videos that we’re split-testing. One is 53 seconds, one is 38 seconds, and one is a minute 48, so different lengths. And we’re also testing different primary text, different headlines, and a description free download was the thing that won. And then we have the website URL.

So there are some different versions of this ad out there testing. And, for shits and giggles, what we’ll do is we’ll go over here and look at the breakdown, just so you can see how that stuff is super important. So I’m going to, from this page, break it down by image.

So I took a couple of clients through this process the other day and it was eye-opening to see how… So you can see right here, if we look at just this month, so the last 21 days, we see that the majority of our traffic is being spent on this video. So this is the mid-range one, the 48-cent one. The cost per result is $6 and 54 cents, but Facebook isn’t even giving the other ones a try, just because all in all, it’s zeroed in on this thing. We’re going to hit the maximum.

Oh, check this out. So it’s spent $11 on this one and realized it wasn’t worthy it. $25 on this one; realized it wasn’t worth it because the lead cost was too high, 11 bucks and 12 bucks, and then just standardized on the middle one, which is getting us $6 and 69 cent lead. So if we can pull this down also, let’s look at the text and see what variations are winning.

So the variation that is winning… This one and this one are neck and neck. So if your funnel isn’t converting, it’s time to adopt a better strategy and also download our free Funnel Factor ebook. Those two are neck and neck. The one that isn’t converting is updated. Our ebook shows how to convert more clicks to customers. So that one is one that we should probably remove from the split test.

Let’s see, what else? Headline. I think we had a couple of headlines. So this is the difference in headlines. Download your free Funnel Factor ebook is by far and away the winner. So we test these things quite a bit. We’re always optimizing the ads this way. So maybe what we’ll do is we’ll record another video in a couple of weeks, and show you the results of our future split test.

But all in all, this is how Facebook ads work. You want them to click the button; sign up for something. It’s pretty simple. We want them to engage with our sales process one way or the other, and this is a way of getting leads cheaply and effectively.

So if you have any questions at all, if you have problems with your Facebook ads or if you would like another set of eyes looking at your Facebook ads, or if you want to talk about scaling your inbound marketing strategies in your sales funnels, go to doneforyou.com/start and we’ll jump on a call together and we’ll go through everything. If you like this kind of stuff, you want more tools, techniques, and tips, go to doneforyou.com/gsd, and we’ll make sure to send you this stuff a little bit more often through email. And if you have any questions at all, answer them below whether you’re on YouTube or Facebook. And if you like this video, make sure to subscribe, like it, and comment if you have anything. I’ll talk to you soon. All right? Thanks. Bye.