What is the Difference Between Paid Advertising on Facebook and AdWords?


What is the difference between paid advertising on Facebook and AdWords? Effective Facebook advertising. Tips on how a company can use Facebook for advertising their product or service.

Katie Beckam of International Marketing Institute and Tim Barrie at Holistic Web Presence in discussion.

T: The question is, Katie, is when you do Facebook campaigns, what’s your approach? I would think Facebook needs to be approached in quite a different way to how you might construct an AdWords campaign?

K: Right, it is different but it is actually very similar if you’re looking at from a structure perspective. The same way that with Goggle search, or grouping keywords, and then with our ad groups, we’re making them really small tightly focused ad groups, and in one account I might have a hundred ad groups. You really want to make everything down into as specific as possible with still having enough volume of course.

We take the same structure approach over to Facebook. Only the targeting is totally different. And so, instead of grouping people and testing them by these long groups of keywords, we’re breaking people up by demographic. And so, Facebook is pretty prolific, it’s amazing how much detail we can get now and sometimes that’s a little bit counter intuitive. I’ll give you an example.

Facebook – Breaking People up by Demographic

So, for a shoe company, there was the certain line of shoes, that we were selling, that client were selling. And so in advertising to them, it’s intuitives, you’ll start with things with people that like fashion. And then layer that, with people who purchased online in the past whatever days or people who shopped at the actual stores that sell the shoes and so on all the very intuitive things.

And we will group it up. And from there, we might segment it down to okay… they shop the Target, and they like fashion, and they’re 25 years old. And then we’ll test that against, they shop the Target, they like fashion, and they’re 26 to 34 and so on. So we’ll do different age groups because we want to see how are the different demographics perform against each other.  But we don’t want to go too small because we’re not going to have enough meaningful data to know anything.

That’s one way to group it. So we can cross layer these demographics to make hyper targeted audiences. We can learn behavior, plus their demographic, plus their interests, plus their associations, they call just keep going until Facebook like, "too small!" or if I run it and I’m getting hardly any people, then I’ll back out a little bit.

Anyway, there’s also very counter intuitive ones that a lot of advertisers don’t think of. So this would be akin to those long tale keywords that we go after in search.

One of the things that we found out about this line of shoes was that people who liked Taylor Swift, specifically, and people who liked Florence and the Machine, had really high conversion rates for this shoes.

Associating Music with Shoes

What does music have to do with shoes? Nothing, on the surface, but if you can target people by their interests, even though it’s not that product is not something directly related to that product, but you know that it segments them into this super defined group of people that are going to love your product, then that’s what you go for.

T: So, that’s the targeting. What about the nature of the ad itself? My understanding of Facebook is you need to find some kind of indirect way of capturing their attention.

Targeting the ad and Hot Buttons

K: Yeah. So this is going to depend on whatever the audience itself is. I always go after hot buttons. Hot buttons are like little triggers that get people to click on the ad, click through to the web page. And of then course we want to be looking at what they do once they get to the web page. So I always break things up. I’m looking for what resonates with people and what has the most prevalence among those people.

I’ll give you an example of what I mean by those two different things. I never go in depth this far. I have in my own spreadsheets, I calculate what I call a prevalence score, and I also I calculate a resonance score. I those are just my own made up terms.

For example, with Marriage Helper since you know a little about them already. For Christian woman in the household income brackets, between $75,000 and $100,000 annual. They have hot buttons that are… with women, the husband is addicted. Here’s how to talk to him. Another one could be, ‘if you feel isolated in your marriage, here’s what to do‘. Another one could be, ‘if you feel controlled and berated by the man you married, here’s what to do‘. So we have these pain points, so we always want to address some type of pain point or trigger that’s going to get their attention and then we can test those against each other.

T: So, it sounds like what you’re doing, is providing an intelligent solution to a defined pain point that relates to a specific demographic.

K: You can put it like that…

Example Exercise – Applying Hot Buttons to Pain Points

T: With the new client, their… some of the types of clients are people that are in distress to sell their property quickly or it’s just for a for sale by owner type of individual that feel that they don’t want to have to pay a realtor for something that they should be able to do themselves. Those are the types of clients and the sell, is one of getting them to work with Simple House because Simple House will provide a cash offer there and then, making the process of selling the house very simple and straightforward. As against the client worrying that they’re going to be able to sell, how is it going to take, and just go through a very complicated process to sell their property. So given that it’s that type of market, would you be thinking in terms of providing them some kind of content, that helps them understand how to do something, or would you take them straight in a conversion page. What would be your approach on that?

K: As you know, I am all about the survey funnel, whatever the pain point is, if they’re frustrated because they’re stressed and let’s say they’ve contacted a couple of realtor’s and they’re so confused and frustrated. And the ad reference is that, if you don’t have a page that speaks directly to that, then we’re going to need one, of course, and from that page, we want to make it really simple on what to do next. And what we’re going to do on that page is speak to that frustration and simplify the process. Get them to take a deep breath. If you’re a client gets to your page and goes “ahhh! I feel like in this case.” That’s probably what we’d be going for. If a good client for them is someone that’s stressed.

T: To make it work in Facebook, what does the ad need look like in order to grab somebody’s attention, given that they’re not using Facebook to go buy something. You’re interrupting them, right?

K: Uhummm..That’s why the ads always want to speak to the pain point or the trigger and want to blend in. Here’s an example. For a USAopoly, the game company, they have a new game in November and it’s an adult drinking game basically, and the game comes with spill proof cards. And so, the actual ad copy that’s doing best, it’s a video. But this ad copy works on static image ads as well, and it’s in the person’s newsfeed, and the ad copy says something long lines of, “the spill-proof cards in this deck means that we don’t want to worry about, ‘that guy’. Haha! Who’s usually me. Haha!”  Sounds like one of their friends is posting about themselves and you think, "Oh, that’s not going to work in an ad”, but it does.

So it really depends on your market, and how formal should the messaging be based on who you are targeting. In targeting in a newsfeed, we always want to be as authentic as possible. We don’t want to sound like we want to trick them, that are company. We also want to talk to them the way that we know, they connect with. And if it’s an adult drinking crowd, then we’re going to use text type speech.

T: So, do you have any idea how you’d go about identifying a demographic for a distressed homeowner?

What Demographic?

K: I will start with the client. They should have a good understanding of their starting point. And if they say: “I don’t know,” then I would interview them.

And so, for a bigger company that has a customer service department, what I’ve done is I’ve called customer service and I interview one of their top reps. So for Griswold Homecare. They had a customer service person who’s been there forever and gets really good feedback from clients and stuff. So I called her.

Okay so what are some of the people that you feel like you hear from is it men or women? And they answer, "oh it’s mostly women".  Okay are they more like in their 20s or 30s, and she’ll answer me that. And then I’ll say, “are they feeling stressed financially when they call? And then she’ll say, "yeah well some are, and blah..blah… and some aren’t… and here’s with their feeling.

So when they’re feeling like pressure that they’re spending time with their mom who’s aging, and why can’t they care for their mom themselves? And she answers, “well, because their time spent together is now more like a burden.

Okay, so is this person is feeling kind of guilty maybe?” So it’s really just a conversation with the client where we’re going to pull that out.

If they really have no idea they can’t even answer those questions to start that conversation, then I would not want to work with them, ever. It’s not my job to define their market. They should have an audience.

T: Right.

K: They might not be able to define them formally or have it written out in some business plan, that’s fine. As long as they know who their audience is and they’re providing value to that audience. They might know they know. That’s why you got to…you know same way like you’re interviewing people on content … they don’t know how much they know.

For example, with the game company, we started with a certain demographic for the scheme Tapple, it was posed as a family game, fun for the whole family.

Their tag line was, ‘fast word fun for everyone’. And then as I started out the demographics, I narrowed in on moms because I began noticing that women with kids outperformed any other demographic.

So when I started narrowing it down on women with kids and doing different messaging,  I’m identifying, whether it fun, it’s educational. And so, you keep refining the messaging and testing the different messaging, against each other. So for example, on that one after a few months, I came down to this phrase that was: “Kids have so much fun they don’t know how much they’re learning.” And then  I could not beat that phrase, or months after that, and I tried all different things

Another tag line that was really getting an audience was “fast word fun for the whole family". And so they actually changed the tag line on that box because they realized that their audience was even more specific than they knew. And they realized that it was a lot, mostly moms who are buying it, because it was helping their kids learn. And it’s fun for them too. The client often doesn’t know how the specifics that we’re going to get. But that’s fine. We do need to have somewhere to start.

T: In a nutshell, the way you would define the difference between doing AdWords and doing Facebook advertising is the way your targeting the demographics in a very, very tight way. Is that right? Or would you say it differently?

In a Nutshell – Demographics and Optimizing

K: That’s absolutely right. That’s the main difference is the targeting. In both cases, we’re constantly testing or optimizing. I recently stopped using word testing because John, the president of USAopoly, was like “Why are we still testing things?

T: Oh Right, "why haven’t you got it figured out yet"?

K: Yeah, it’s not that we’re still testing, it’s that we should be constantly optimizing because the market changes. So we never want to just, “okay this ad is getting these click through rate, it’s pretty good so we’ll just leave it”. We always want to have at least two ads to see which one is better, or if you’re still testing the demographics, and we have an ad that we know works pretty well, then we’ll have the same ad, but then we test different demographics with that ad. So that we can get good data about those demographics.

And often, no not often, always, we’re going to get insights on that about the demographics that’s more specific and they started it out with.  And that’s part of the cool thing, that’s why I love what I do because you learn so much about people. Because you’re constantly putting this stimulus in, and then you’ll get a response. Stimulus in – response and it’s like immediate feedback.

T: So, with AdWords, your targeting keywords were the assumption is people are actively searching for something. With Facebook, you’re showing up to a particular demographic. Doesn’t that mean that the nature of what you’re saying has to be very very different?

People are Always Searching

K: Yes and no. People that are actively searching, it’s like often because we’re in the right place at the right time. And they’re revealing to us with their search, what they really want. So we always say, keywords are windows to our customers’ souls. And it is.

Humans are always searching, everybody is searching for something. And at any one moment, we can give them what they’re searching for that’s awesome! People that are on Facebook, deep down they’re still searching for something but they might not be actively searching for it, which is why we have to get so targeted with the demographics.

There’s nobody on Facebook that, if like how do you target people who are struggling with affairs in their marriage on Facebook, you can’t. Like there’s no interest category on like the husband having an affair.

T: Right.

K: But, if someone’s husband is having an affair, they are hurting the same way that’s someone who’s actively searching, "what to do if their husband is having an affair".

T: Right. Now, I understand the term you’re using. What did you call it, hot buttons? Which is you’re finding that question that resonates with the individual that has that need.

It’s in the Way we Propose the Solution

K: The way that we pose that solution, to that thing that they’re searching for, is different depending on the platform. Because on AdWords, we can very easily say: “If your husband is having an affair, if you want a divorce, here’s what to do next.” If we just throw that out there on Facebook, they might be offended. They’re like, "what do you mean? I’m not having problem with my marriage. This is like privacy violation."

T: Right.

K: So we have to kind of approach it a bit differently on Facebook. And that’s why we have different… it’s a totally different ad format. You can have pictures, videos. You can have comments, articles.

T: What would be an example of how you would approach it? So you gave me an example of what you would say if you need a divorce, here’s what to do next. That’s your AdWords approach because you are driven by the keyword. But on Facebook, you would not say that. Instead, you might say something like?

K: So the same person who’s actively searching or not searching, when they’re actively searching, you can answer their query with "When your husband has an affair, do this next.” On Facebook, it would say: “If there’s been an affair in a marriage, it doesn’t have to mean the end. Here’s why."

TB: Right. Very good, very good. You nailed it I think. Yeah, I’ve learned something there. Just in terms of how to think about it right, good one! So do you think people that do AdWords necessarily are going to be good at Facebook.

If a Person is Good at AdWords is it safe to assume they will be Good at Facebook Advertising?

K: No

T: Or do you sometimes it’s better to have a separate specialist for Facebook because the thought process is sufficiently different, or not? What’s your thoughts on that?

K: Well, take me for example, I can do both very well. That’s because I understand the difference. It depends on the person. If you have someone who is approaching Facebook in exact the same way that they approach AdWords, it’s not going to be effective the way it should be. That might sound high and mighty but I’ve been in this industry long enough to know that people say that they do both, but they don’t really.

T: Right. That’s my experience.

K: They are like, “Oh yeah it’s all PPC, it’s all pay per click" – eh! Okay so just because you pay for the ad the same way, then you’ll approach it the same way? – eh…

T: So what do you charge for doing Facebook ?

K: It’s in the pricing sheet that I sent you, but it’s basically the same way that I charge for AdWords. It’s 15% or the minimum.

Facebook Advertising – Cost Per Acquisition

T: Do you have any thoughts on what the cost per acquisition within the Facebook environment? I mean for lead, are you able to break it down to that level the way you can say, you should be able to get a lead for ‘x’ amount or something like that?

K: Yeah. You can track it and so you can break it down and then you should be constantly trying to lower that by doing multiple things, targeting more efficiently, messaging more efficiently and having better landing pages that make it easy for them to pick the next step.

With Marriage Helper when we first started we didn’t have that survey funnel we just had their basic landing pages. We were getting leads, I don’t remember, I can go back and look but you know, in some of the ad groups, it’s like $75 per lead, that type of thing. And so, we did a lot of work and now in some of our little ad sets, which is basically the same thing as an ad group and AdWords. In some of our ad sets, those little clusters of demographics, were getting leads for like $2.50.

T: Well, that’s huge, huge! A ridiculously huge difference!

The Devil is In the Details

K: Yeah. I think the way you play it, is the devils in the details…

T: Right.

K: You can spend $2000 on Facebook and get hardly anything. You can also spend $200 on Facebook and get a lot. If everything is very refined. So that’s always the goal, is to get to that point. It needs money to get there because you have to test and get data and stuff.

T: Doesn’t Facebook also provide some kind of ad performance scoring or something where it tells the way you can fit in

Facebook Relevancy Score

K: They have the relevancy score. The relevant score is kind of similar to Google’s quality score.

T: Right

K: So basically, Facebook is telling you: “Hey, here’s how relevant we think this is to the audience that you’re trying to target with it. But there’s a few ad sets, with USAopoly, where the relevancy score is like a two but we have really good…so those are video ads and we really have good metrics on those. So it’s like whatever, Facebook just doesn’t like us…

But, what we’ll notice is a lot of times as the relevant score goes up, we’re targeting more efficiently. So we’re going to see our price go down. And the higher relevant score is, typically, the more Facebook does favor us, and will show our ad over a competitor in the newsfeed. So it’s like a win on top of the win. You know how in AdWords…do you know how an auction is run in AdWords, right? You know about the quality score and the way you win.

T: yes.

K: Can I just kind of speak to it for a minute?

T: Sure, yeah.

AdWords Quality Score

K: So, in AdWords, every time somebody searches for something, there’s a real time immediate auction. So, all these advertisers who want to show up for that word, are bidding. But it’s not a straight price bid like a regular auction. It’s a… you win based on your score, and you get score by multiplying your bid, times your quality score. There’s like probably a hundred things that go in your quality score, but one of the biggest foundations is your click through rate because it tells Google that your ad is very relevant, so they like it and they want to show more.

So let’s say we have a quality score of 8 and we’re bidding $1. And then our competitor over here has a quality score of 3 and they’re bidding $2. So, we have 8×1, get an ad score of 8. And in that same auction, they do 2×3 so they going to have an ad score of 6. So, we beat them. We’re going to show up higher and we’re going to pay less.

So, there’s an inverse relationship between the quality score and the amount you pay. So that’s why there’s huge incentives to getting your quality score up, but that will take time and skill, and eventually, hopefully, in the near future, you’re going to actually be saving money by spending that time, because you’re going to be spending so much less on dollars.

I really enjoy working with clients that provide a solution to a pain. That’s why I like Marriage Helper so much because people…their marriage is in trouble and are hurting. And Marriage Helper actually does help them.

Save