Podcast: Play in new window | Download
If you find the process of Facebook Ads to be overwhelming, you’re certainly not alone. But besides having great content, an audience, and an ad budget, Facebook Ads expert Amanda Robinson can’t stress enough the importance of building a solid foundation for your ads, and of course testing the heck out of them in a way that will definitely tell you which ads are working and which are not. Because no one wants to waste their ad dollars.
Rich: My next guest is widely recognized in the digital marketing industry for her knowledge, training, skills and coaching as a Facebook Ads expert with a flair for chatbots, analytics, and algorithms. She is an international speaker regularly featured as a guest on digital marketing podcasts, has been published by Social Media Examiner, and advises for their Membership of over 2,000 entrepreneurs and business owners. Amanda Robinson, welcome to the show.
Amanda: Thank you so much for having me, Rich, I really appreciate it.
Rich: I’m very excited to get you on here. We’re talking about something that we can always get better at, Facebook ads. So how did you actually get started in Facebook ads?
Amanda: That goes way back to when Facebook ads were first a thing. When Facebook first rolled out advertising on their platform – back in the day when you could take $20 and put it on a post without any complicated options and get huge reach and huge feedback from it – that’s what got me addicted in the first place.
So I started out growing with the platform over the years and learning all the new features as they’ve been rolling out, and learning all of the glitches and problems and troubleshooting through years and years of just growing with it. So it’s been a long journey. I look back now and think wow, if I didn’t know what I know and tried to get into the game today, it would be an uphill battle.
Rich: Do you find that the analytics and the recording tools that Facebook has have also become more sophisticated over the years?
Amanda: Absolutely they have, they’ve evolved. And if it’s been several years since you’ve been in the back end of your Facebook ads looking at your analytics in Facebook, then you’re in for a little bit of a good, positive shock and surprise at how much data they have in there and how much data is available to you.
Rich: And this is a really good conversation for us to have because I certainly can talk a little bit about Facebook ads, but I’m not the guy in there every day doing it, so I’m really interested in what we’re going to be talking about today.
Obviously there’s a lot written and spoken about running Facebook ads. There’s less, it feels to me, about how to analyze them. So how do we analyze them for better ROI, and what kind of information should we be looking for?
Amanda: Great question. So to take a look at your Facebook ads and see how they’re performing, there are so many different areas that we can look into. So you generally need to have a good handle on what’s happening at every stage of the game. And when I say every stage of the game, I’m talking from your campaign setup, to your audiences and your targeting, to your ad placement, the actual creative that you’re using. You need to look at all these individual pieces and dive really deep into those results at all those different stages, rather than just globally looking at all the results and just saying, “Well, I ran an ad and spent x dollars and got x results. Let’s do it again.”
There are so many different areas that you have the opportunity to peel back the layers, dive in deep, and kind of tune it up. Imagine it being an engine, and if that engine isn’t operating at peak performance, then there’s lots of opportunity for you to make tweaks and changes to optimize it so that it’s performing so much better and cleaner for you and using your marketing budget as efficiently as it possibly can be.
So I’d love to be able to chat and walk us through some of the stages on how you can audit your own Facebook ads from all those different stages.
Rich: Absolutely. And I’m speaking as a guy that spends a little more time with SEO than he does with Facebook ads. So I know that if I do things on the front end of my website, I don’t see the results on the front end of my website, I go to a different program, usually Google Analytics or Google Data Studio. This is a really dumb question, is there a different analytics tool for Facebook or am I seeing all the answers in the same place I logged in, to look at my ads?
Amanda: That is a great question. Facebook Ads Manager is where we’re going to be spending the majority of the time. So Facebooks Ads Manager is where you are going to actually create your ads. You’re going beyond the “boost” button if you’re not just using that lovely button, and if you are it’s time to start weaning off of that and getting into Ads Manager to create those ads. But that’s also where you analyze the results.
You have several different areas where you can analyze those results, either right from Ads Manager, there’s also Facebook Ads Reporting where you can make your own custom reports, and there’s also Facebook Analytics.
And Facebook Analytics is a little bit different. It’s a different tool than your Ads Manager, it’s a little more advanced, but there is such a deep wealth of knowledge hiding in there. You can analyze the analytics for your page, and you can also analyze the analytics for your pixel so you can look at the data a couple of different ways. And it’s quite a robust system. It’s not quite as good as Google Analytics is, I use Google Analytics as what I would call the “holy grail of data”, that’s the main data that I trust. But looking at Facebook ads and all the activity that’s happening in there, if you really want to dive in deep and look at how your funnels are performing, you can use Facebook’s analytics tool to really build those reports out and get a really deep look at what’s happening.
But for the main purpose we don’t need to geek out that hard. We can take a step back and let’s just focus right within Facebook Ads Manager and all the different areas we should be aware of and paying attention to when we’re analyzing how our ads have been performing.
Rich: Ok, cool. And you and I chatted a little bit before and we kind of identified some areas that I, as a business owner and as a marketer, should be paying attention to. So let’s start with objectives. Give me a little walk through of what objectives are. I know we need to decide if we’re using the right one, and tell me a little bit about why this is important to choose the correct objective.
Amanda: Right. So when you’re choosing an objective for your Facebook ads, let’s look at it this way, it’s you telling Facebook what your goal is. What are you partnering with Facebook on to make happen? What is that finish line that you need to cross? You’re establishing that finish line.
Now depending on what objective you choose it will influence the cost of your ads. And that finish line might be a little bit different in the sense of, say you have a finish line where you just needed to run a 100meter dash and you have a finish line. Excellent, you can do that. Versus the finish line being climbing an entire mountain. Well the amount of energy, effort, and spend that would have to go into climbing a mountain would be so much larger than just doing a 100meter dash.
So choosing your objectives have different intensities of what you’re trying to accomplish. So if you are doing a reach or brand awareness objective, those are what I like to call the “cheap and cheerful” objectives. So they cost less but they really don’t do much for you in the end. They don’t build an audience of people you can remarket to.
Whereas if you are doing a conversion objective or a lead objective, those are more costly but they have the ability to produce better results for you. But you also have to have the audiences in place that are primed and ready to receive ads with those objectives. So they’re a little more costly.
So when it comes to looking at how your Facebook ads have been performing, one of the first things I do is I look at the objectives that you’ve been using. Are you using the exact same objective every single time? Are you trying to use a conversion objective to say “that’s the finish line I want, Facebook. I want everybody to go on my site and make a purchase”? Well great, cool. But do you have enough data to give Facebook for Facebook to optimize the ads to actually achieve that goal together with you. Or is it performing as good as it could? You need to take a couple steps back and maybe run a traffic objective, which is not quite as expensive and can drive more traffic to the site to get you more data so that the next time you’re going to run a Facebook ad, then you can maybe focus on a conversion objective.
So not to get too geeky with all the objectives, but taking a look at all the objectives that you’re using is going to heavily influence the cost of your Facebook ads and determine whether or not you’re using your budget as efficiently as you possibly can be, depending on all the factors that you have to work with. Such as, as you advertising to a hot, cold, warm audience, and are you testing assets that you know will perform, etc.
So the objectives is sort of the first choice that you have to make and it’s establishing together with Facebook what is our collective goal at the finish line that we are going to cross and how are we going to get there.
Rich: Ok, I want to stop you there because I have a question about that. You mentioned a couple different objectives and I usually didn’t care about the cheap and cheery ones, as you called them.
Amanda: Cheap and cheerful.
Rich: Cheap and cheerful. I want to get people to my website and ultimately I want to get them to convert. I’ve always assumed that when I see that “do you want traffic or do you want conversions”, that it really was not important. And it sounds like from what you’re saying… it was not important in terms of objectives. Not that it’s not important, those are two very different goals. But are you suggesting that Facebook would show my ad to a different type of person if I said my objective is conversions versus my objective is just traffic?
Rich: So how would they know that?
Amanda: So let’s put it this way, I’ll give you an example. I had somebody who was trying to get signups for a webinar and they had a global audience worldwide. They had ads going out for a website signup, they had a conversion objective, they wanted people to sign up for their webinar. Facebook was optimizing those ads and delivering that finish line, that goal for them. Ok, so you want conversions, we’re giving you conversions.
But what was happening with those conversions is that a lot of those conversions were starting to happen in South Africa. And if you look at their sales and where they actually do business, they weren’t getting any people actually converting off of those webinars into sales from South Africa. And the places where they did need the business to happen, they needed that website registration as that finish line goal that they’re trying to achieve, it just wasn’t happening. They weren’t getting enough data coming through from places where they did need the business to happen and that website registration as the finish line goal that they’re trying to achieve. It just wasn’t happening, they weren’t getting enough data coming through from the local areas where they needed those conversions coming through. They just didn’t have enough data, it wasn’t optimizing.
So they needed to take a step back and change their objective and do a traffic objective to try and get more eyeballs onto the site by targeting the right areas and getting their people to activate and move and actually take action and arrive at the site. So it’s one of those things where depending on which one you choose, Facebook is just going to go up and say, “Ok, you want more people to your website, sure, we’ll find more people to your website.”
If you want more people to purchase, well the same people they can get going to your website aren’t necessarily the people who are actually motivated to purchase. Facebook has way more data than we have access to, and Facebook can optimize quite well when we point Facebook in the right direction and we give Facebook the right finish line that we need to cross, Facebook really can optimize for it. You need to know how to look at the data and understand did this ad that we ran work, is it getting to the right people, are we getting the right people in our funnel, and are we attracting the right people into our audience so that when we remarket to them are we seating them with the right people so that we have a healthy funnel all the way around.
Rich: Would you suggest that if we’re just getting started with Facebook ads that maybe we should be more focused on traffic over conversions to give Facebook a little bit more of a sample size so they understand?
Rich: Oh, ok.
Amanda: And that’s one of the pitfalls that I see quite often is that people who are new into advertising on Facebook, they just want to build the funnel the way all the pros do. They want to have everything operating the way that they think they should have it set up for conversions, but they don’t have enough data already built up to actually give Facebook enough to work with to properly optimize everything. So in the end you’re actually wasting your marketing dollars and your ads are not performing as best as they could.
You need to walk before you run, so you need to start by having some data that Facebook can work with. How do you do that? Let’s start with some of the lower touchpoints, not quite the high cost of engagement. We need to earn people’s attention span to start, and we need to make sure we need to test it as we’re going to make sure that the people who are being influenced by our content are the right people for our business. We don’t want to have you just deliver content out and have the wrong people engaging with it, and then remarket to the wrong people, and just flood our funnel with what I call “dirty data”.
Rich: Ok. So let’s move on to audiences. What should we be looking for when it comes to audiences? So when it comes to audiences, the audiences really are the secret sauces to your Facebook ads. The analogy I give is, you can have the right ad – the absolute pristine most perfect ad on the entire planet – and it you serve it to the wrong people, you’ll get no sales. You can have the worst ad in the world, a terrible ad, but if you serve that to the right people who are ready for it, you’re still going to get some sales out of it.
So your audience is the real key. And building these audiences over time and fostering these audiences, bringing them through your journey, continuing to serve high quality content to the people who are engaging with you and rewarding people for that engagement, that’s the real magic of Facebook advertising. So it’s not just flipping the switch on and start advertising and get high quality audiences. It’s something that you actually need to grow over time. So when you first start advertising, the expectation needs to be set that your first few thousand dollars you’re spending through Facebook ads is to grow higher quality audiences by getting your content in front of them.
So when it comes to those audiences, you have tons of choices. I just did an hour and a half long training yesterday just on audiences alone. Going through the different types of audiences, all the different choices, how to choose the demographics, how to choose the interests, the behaviors, all different choices that go into that. So you can build out your audiences in so many different ways, but because they are so critical to the success of your ads, this is where you need to be spending a large portion of your time. When I say “large portion”, when you’re first setting up your Facebook ads you should be doing hours and hours of research on your audiences, making sure that you’re setting up different types of audiences to try testing with. And then over time continue to revisit those audiences.
So something that happened recently, Facebook has recently started to kind of do a little bit of housekeeping in the backend with their interests, and they’ve been stripping out a lot of what they would consider to be underutilized interests that are insignificant that no longer really need to be targeted. So that’s how Facebook has positioned it to us. However, with all of the clients and ad accounts that I had access to, what I’m seeing is that some of these very high quality, very niche local audience interests have been stripped out.
Amanda: So what happens in these audiences people have built up and been using over the years and they kind of set it and forget and never really revisit their audiences. A lot of these targeting options have been stripped out. And when you go to look at your Facebook ads or create a Facebook Ad, sometimes you have little warnings along the top of your aAs Manager, and they’re just in really tiny type, it’s kind of annoying and they don’t really make sense to you, and you just click the little x and ignore it and move on. Well that’s an important part of Facebook trying to say “hey, by the way, we’re removing an interest that you’ve been targeting for a while, and you might want to go in and update your audiences”.
Most people are ignoring that, and all of a sudden, the performance of your ads is tanking. It’s going downhill because that little niche interest that you’ve been targeting for years happens to be one of the higher performing ones for you. Now it’s been stripped out, and Facebook is making you use more broad options to target. It’s definitely impacting the performance of ads right now, and a lot of people are just unaware of it. They just think Facebook ads are getting more expensive, but they’re not taking the time to look back at what they’ve been doing, and take time to do the research, and take time to go and build up those audiences and keep putting energy into it.
Another one, for example, if you’re retargeting video viewers, anyone who’s ever watched a video on your Facebook page. If you’re retargeting those folks, that’s one of the audiences that doesn’t automatically update every time you add a new video, you need to go back in and update that audience and keep adding to it. So there’s a little bit of work that needs to happen in there, and I think a lot of advertisers just have the go-set-it-and-forget-it blinders on without paying attention to what’s actually happening under the hood with the audiences that you’re targeting and setting up.
Rich: How carefully or how closely do you align your objectives and your audiences? Are you kind of playing matchmaker between those two groups?
Amanda: Yes.Yeah, I mean, you wouldn’t wanna send a conversion objective for a purchase, a purchase conversion, to a cold audience, that doesn’t make a lot of sense. When we talk about making first impressions, think about going to a party. I’ve been hearing a lot of people use this party analogy and I’ve been using it for a while. You walk into a party and somebody introduces you to somebody you’ve never met before, and you go to shake hands and say “Hi, my name is Amanda, great to meet you. I run The Digital Gal”. And that’s enough, that’s a first impression. When you’re walking into a party, you don’ t walk up and shake their hand and say “Hi, I’m Amanda. I do Facebook Advertising. I have all these different programs, here’s the price points, here’s what I think you need right now, do you wanna buy? Are you sure you don’t wanna buy? Are you sure you don’t wanna buy?”
That’s the first impression that people are making with Facebook Ads when they’re going and grabbing a cold audience, and having a purchase objective, conversion objective and serving ads out that way. Sure, some of you are going to make some sales, but it’s maybe not the most efficient use of your marketing dollars, compared to taking a step back, serving ads that make a really good first impression to your cold audience, serving additional content that warms up your audience and rewards them for their attention span, and then gets traffic onto your website so that you can then remarket to your website visitors with the actual ask, or asking for the sale.
Rich: Right. Plus, I heard this phrase today and I love it, plus, imagine all the innocent bystanders you’re hurting with your hardcore marketing message who really just wanna be left alone as well. So we talked objectives audiences, let’s talk about placements. Do you just let Facebook decide this? How do audience networks fit in, and what kinds of things should we be looking at to make a decision if our placements are on point?
Amanda: So for your ad placements, this is another area that when we start getting a little geeky with our Facebook Ads, when you first start getting into Ads Manager and creating your ads, you’re generally going to leave your placements on automatic placements and let Facebook decide where it’s going to place your ads. Now Facebook has a lot of inventory resources on where your ads can appear, where your ads can show up. Placements can include things like Instagram stories, Facebook stories, Facebook news feed, instant articles. So if you read an article on Facebook that is an instant article that opens instantly without taking you really off the Facebook experience, there’s a whole bunch of ad inventory placements all through that article.They break up the article with little ads.
And you also have something that’s called the audience network. The audience network is- think of it this way, if you’re on an app and you’re playing a game, let’s say Scrabble, and in order to get to the next stage you have to watch a 15 second video and then you get to move on to the next stage, those are ad placements, or banners. All sorts of things that happen inside the audience network, those are all apps, so it’s inventory placement inside apps.
When you allow Facebook to optimize and choose where your ads are going to be shown, then Facebook, for the most part, does a pretty good job, it does alright. But depending on what type of ad you have, what your end goal is, what your objective is, you may not want your placements being automatically decided by Facebook. Because again, how Facebook works with Facebook’s algorithm and how it optimizes is that once you start getting success in one area, Facebook puts the blinders on and just focuses on that one area and heavy delivers the ads to that one group of people or that one area that is getting you the success.
I have an example of when I was working with a client. They had automatic placements on, and their objective was traffic and it was optimizing for clicks. Not for landing page views, just for clicks. So what they’re doing is they’re telling Facebook “here’s an ad, go serve it wherever you want but the end goal is I need to get a click, ok? Alright, Facebook, go for it”. So Facebook goes out, serves a whole bunch of ads out, and is getting you a ton of clicks. But when we actually look at the data, all those clicks were happening in the audience network. So all those clicks on those ads were happening right within the audience network so they’re happening in apps. But when we compare the data on, well all those clicks are happening but did those clicks actually make it to the website, when we compare it to Google Analytics? No, it’s resulting in a whole bunch of abandoned clicks. And what was happening is because it was heavy delivering to the audience network, it’s heavy delivering to all these apps, and people are accidently clicking the ads and going like, “I didn’t wanna do that’, and then backing out. Well Facebook is giving you the goal that you are asking for. You wanted clicks, we got you clicks.
Thinking about it this way, so what we did to change that campaign is we went back in, we edited our placements, we removed out the audience network, and we changed the optimization of the traffic objective. We changed that to a landing page view. So now what we’re asking Facebook for is to say now go and place this ad in the areas that we know that our people are hanging out where they’re ready to consume this content, so in the news feed of Facebook and the news feed of Instagram for example. And we’re saying the end goal that we have here, is I need somebody to actually hit my website and have the page load and the pixel fire, so that I know they actually knocked on the door and made it into the website.
By making that one little tweak and one little change, they saw the performance of their ads increase, they saw sales coming through. It was an achievement that, had they not taken the time to go back and analyze what was actually happening with their ads, they would fall into the category that I’ve seen so many times where it’s just people that throw their hands up in the air and say “You know what, I’ve tried Facebook Ads. They didn’t work for us”. Ugh, I hate that!
Rich: Ok. Another section that we might want take a look at is demographics. How do we know who is interacting with our ads, and how can we improve the performance there?
Amanda: You have the ability to filter. When you look at your results, you have this breakdown filter that you can apply to your ads. So go into Ads Manager, and it’s a little drop down menu on the right hand side: breakdowns. And you can go in and analyze your ads based on all sorts of different views. You can break it down by country, you can break it down by different regions, you can break it down by age and gender. You can look at your ads in so many different lenses when you’re analyzing the results. I really do encourage you to take the time and go and do that.
Another example I have with running ads is now we’re matching up the different placements, versus what are the demographics that are interacting with those placements. So I have content, for example, that I put on Instagram and I put on Facebook. I notice that the content I was putting on Instagram was starting to heavy deliver to a male audience. When I look at the people I do business with, it’s closer to a 50/50 split male/female, or in some cases skews a little bit more toward female. So if I have ads running wild on Instagram, and the only people who are engaging with those ads on Instagram are men, then those people are now entering into my audiences of people who have engaged with my profile. Now they’re going to get remarketed to, they’re getting into my funnel, they’re getting into my system. But if I just set it and forget it and didn’t take the time to look at how those placements are forming, then I’d be feeding my audiences with dirty data and I didn’t want to do that.
By looking at that, I was able to go back in, edit those placements and say “you know what, Facebook, if we’re on Instagram, start delivering these ads to females”, and eliminate men out of that bucket. I’m steering it back on track to where I want it to go, but I’m paying attention to the numbers and I’m paying attention to the results, and where you can do that is by really leveraging those breakdowns when analyzing the results of your ads.
So like that one example where all those ads were heavy delivering to South Africa – and there’s nothing wrong with the ads delivering to South Africa they still have occasional business that comes out of South Africa – but the majority of their business was happening in the US and the UK. So by analyzing those breakdowns and looking at the demographics behind where those ads are delivering and the actual results of where Facebook has been serving those ads, it helps you be able to make some adjustments and changes to match it up to your actual business goals.
Rich: Are there other types of demographics in there? Like you mentioned male/female, age, location. I’m just thinking about my other brand, my other podcast Fast Forward Maine, where we’re trying to get in front of Maine business owners. Are there job descriptions in there, or is it really just those basic demographics that Facebook lets us use as a filter?
Amanda: It’s a little more on the basic demographic side, and again that has a lot to do with privacy, so Facebook can’t give you too much data on specifics on who they’re serving your ads to. It definitely has a little more of the broad strokes feel to it, but it’s very valuable data that you should be looking at on a regular basis.
Rich: Alright. The last one’s creatives, there’s a few things that I’m curious about. One is why, of all the things, did you list creative as the last thing? Coming from a creative agency, we always think of those as being first. Secondly, talk to me a little bit about A/B split testing, and also explain what dynamic ads are and whether I should be using them.
Amanda: Great questions. Let’s start with why I put that one last. It goes back to that analogy I gave, you can serve the worst ad in the world to the right people, and they will still buy. You can serve the best ad in the world to the wrong people, and nobody cares. You could be a creative agency and you could have the winning ad combination that you just know is gonna perform so well, but if you don’t have all the rest of these technical pieces aligned and working in your favor, then it doesn’t matter how darn good your ad was, nobody cares. So we need to make sure that we’re serving those ads to the people that care. That’s why Ad Creative kind of falls to the bottom of the list as far as this technical setup, but it’s incredibly important.
Why is it incredibly important? Another small little stumbling block I see a lot of advertisers do is they have one ad, with one headline, and one call to action, and they’re testing it to a whole bunch of their different audiences, they’re putting a whole bunch of budget behind it, trying it with different objectives. They’re really trying hard in all these other areas, but they’re narrow-focused in on just one image, one headline, and one call to action.
Now, human psychology, we might think something is gonna work a certain way, but people will engage and interact with it differently. Sometimes you just need to tweak and position it a little bit differently, people respond to different imagery in a different way. Your imagery really does need to reflect the people you’re serving it to. So if you’re serving an ad to an area where the people in the ad don’t look like the people who are actually purchasing in that area, there’s a disconnect, so there’s less likelihood that people are going to engage with it.
So from that perspective, you really want to be testing. You want to test different ad creative, you want to test different ad copy, you want to test different call to actions. You can’t do that when you just have one single ad that you’re putting all your eggs in one basket on, you need to change it up a little bit.
Rich: Anecdotally, when you’re doing your testing, do you see the biggest change in performance based on headline, on image, on descriptive text, or is it really just random?
Amanda: It depends on the industry, to be honest. In some industries, it kind of doesn’t matter too much what you’re serving them. The creative doesn’t influence as heavily as the rest of the pieces of this puzzle, but in some industries, yes. If you’re a coach and you’re selling a course online, the way that you’re positioning your ad copy is huge. So there are a lot of factors to it and it varies depending on what business you’re in, to be honest, on what areas you need to heavily focus in. In all honesty, you won’t know that until you test it for yourself.
One thing I will say is that I don’t want any advertisers comparing yourself to somebody else, even if it’s a competitor, somebody who’s very much aligned in your industry, you can’t compare your ad performance to their ad performance. There are too many factors involved, so your job is to set your own baseline, and then improve upon all of your ads from your own baseline. How well did your campaigns perform last month? Did we improve upon that this month? Yes, great, keep moving forward looking at your own results.
So that moves us into testing and creating different ad variations so that you can test what your audiences are responding to. Your audiences at different stages are going to respond differently, so a cold audience is going to respond to certain ad creative or headlines very differently than a warm audience. People who are primed and ready to buy from you, you’ve got a little more leeway on what you’re saying, whereas making that first impression is very important to that cold audience.
Keeping that in mind, there’s different ways that you can test your creative. You can do an A/B split test with Facebook where you can test different creative, and Facebook will split ad budget among the two and declare a winner in the end and that gives you more direction for ads moving forward. Or you can do what’s called dynamic creative. Now I just want to caution you, before we get into what is dynamic creative and should I do this, This is… Dynamic Creative is really cool if you have the budget to support it. If you don’t have the budget to invest into it, all you’re doing is you’re spreading your ad budget way too thin on way too many ads and you can’t really arrive at any conclusion on did it work, did it not work, because there’s not enough statistical significance within the volume of numbers for Facebook to declare this made a lot of sense and use these results going forward.
So what is Dynamic Creative? It’s basically a function Facebook has when you’re setting up your ads, you set it the ad set level – a little check box that you check off – and then when you’re actually setting up your ad creative you can choose more than one image, you can create more than one headline. I think you can create up to five headlines. You can choose up to five different call to actions. What Facebook will do is just mix and match everything kind of like a spin the wheel. It’ll match your image with a different headline, with a different call to action, and it’ll just mix and match all these ads and serve all these different ad variations out to your audience.
Based on the ad variations that are getting results, Facebook will start restricting the budget off of the ads that aren’t performing, and increasing the budget into the ads that are performing. And that gives you a little bit of insight as to what your audience is actually responding to. So where that caution comes in is that you need a minimum of 500 impressions per ad for Facebook to really declare if the ad is working or not.
And if you’re spreading that budget way too thin on way too many variations, like if you’re spending $50 on an ad that has 27 different variations, you’re not going to be able to hit that 500 impressions per ad to really understand did this work or did this not work.
Rich: Now I’ve heard that Facebook is sometimes too quick to pull the trigger on what’s the most successful ad. And if I understand correctly, you can choose to just have Facebook choose a winner, or you can kind of have it run 50/50 for a while and then you can decide who the winner is. Is that true, or how might you use that knowledge?
Amanda: I’m going to agree with you and say that Facebook is absolutely way too quick to pull the trigger on a winning ad and heavy deliver ad budget to the one that one. So you do have a little bit of control if you’re using A/B testing on what percentage you want to split, or whoever Facebook decides the winner.
It’s different than using Dynamic Creative. Dynamic Creative you really are setting up a whole bunch of variations and then you’re handing it off and Facebook is running it the whole time.
So my suggestion to use an advertiser is, if you really want to use a thing that you really want to A/B test directly, don’t use all of Facebook’s tools, run separate campaigns. Create a completely separate campaign and run a different ad creative in one than in the other, and split your budget evenly among the two.
That brings us into a change that is coming, which is called Campaign Budget Optimization. So Facebook is going to be changing where we set our budgets. Currently you can set your budget at an ad set level. And stay with me here if this is seeming a little overwhelming if you’re not as familiar with Facebook ads. But what this means is that Facebook was letting us set our budget at the same level where you choose your targeting, so you choose who these ads are going to.
So you can have a campaign and you can have multiple ad sets in it where you’re targeting multiple different audiences. So you can be targeting your website visitors with one, video viewers with another, email marketing list with another, and a cold audience with another. So you might have 4 different ad sets living inside a campaign.
Traditionally, Facebook lets us set the budget at the ad set level. So I can say I want to give $50 to each of those 4 ad sets, so spending $200 and they’re all evenly going to spend $50. But what’s changing is now Facebook was going to be mandating that the budgets were moving to the campaign level. And what that means is that Facebook decides, of all those ad sets you have living in there, Facebook decides who gets what portion of that budget. So Facebook is going to roll that budget down into those ad sets and give more budget to the ad sets that are performing better and restrict budget for the ones that are not performing as well.
So if you’re trying to do an A/B test and you have your budget set at the campaign level and you have two different ad sets, and then inside those ad sets you have two different ad creative – for example – Facebook is still going to take control of that and it’s not going to be an even split and an even test. So in that case you might want to break it out and create two different campaigns and have one campaign spending your $50, and have another campaign spending another $50, and run them alongside each other to try to compare results.
Rich: And then choose a winner yourself, and maybe reassess those funds.
Amanda: Exactly. So you have all these different options and I really do encourage you that you test all these different functions, keep testing over time. The way I do it with a client is I reserve about 20% of their ad budget for experimenting and trying things that we are not doing. Or I might have a client where I know traffic ads perform extremely well, so I’ll put 80% of the budget into traffic ads, but then I’ll take 20% of that budget and try something we haven’t done before, or try something we did in the past that hasn’t traditionally worked but it’s been a year or two so let’s try it again.
Things like using Instant Experience ads, which were previously called Canvass ads. Those are like a full immersive experience, they’re quite tedious to build, which is why a lot of advertisers are not using them. But they sometimes can really perform well. But it depends, it’s testing, so I’m not going to build out an entire ad campaign based on let’s try this new thing and let’s put 100% of our budget into it and see if it works. And then it doesn’t work and we’re disappointed, we don’t want to do that.
So we should be testing all these different functions. Try some A/B testing, try some dynamic creative if you have enough budget to. Don’t go hog wild on creating a whole new bunch of variations. Maybe only try a few different headlines with one image, and then create your own separate campaigns and do your own tests. And try changing up what objectives you’re using. And definitely, definitely, definitely try lots of different types of ads, try images, try video. Don’t get hyper focused on just one type of ad and hope for it to work.
And if you don’t have enough budget to support this immediately, that’s fine. Just spread this out over time and as you’re doing your ads just try new things over 6 months, then over a year. If you’re paying attention to the data and analytics that are happening within your account, then you really do start to get a handle on what’s working for you and what’s not, and you keep improving on your own results, your own baseline, as time goes on.
Rich: Amanda, one final question I have for you is, for the non-agency owner – for somebody who’s doing this for their own business – how many days a week do you recommend somebody goes into their ad planner, and how much time do you think you should be dedicating a week to running the most successful ads you can?
Amanda: You should be dedicating a few hours to setting up those audiences that you’re trying to use. If you’re just breezing through and setting up an ad in 20 minutes and throwing a dart at the board to try and pick your targeting and trying to run that, that’s the one area that needs the lion’s share of the attention at the beginning.
But once you have ads up and running, you should be checking in on them depending on if you’re running ongoing ads that have a daily spend, no start date and no end date so they’re just ongoing, say you’re spending $20 a day on an ad. You should be checking in on that periodically throughout the week. I’d say check in on that three times a week.
If you’re running something that has a short time span, say if you’re in real estate and you’re running a real estate ad that’s going to run for 7 days, you should probably be peeking in on that daily. You should be in their daily seeing what’s actually happening.
One of the areas you can look at is you can look at your charts from your ad set level and you’ll actually see over time it will show you your cost per result and your reach. So at the beginning you’re reaching lots of people with a very low cost per result. But as time goes on with that ad, there’s a decay that happens where you’re starting to reach fewer people in that audience, your ads are getting more expensive, the cost per result is skyrocketing. So you start to see that happen over time.
So by paying attention either daily or periodically throughout the week, you can actually see when these things are happening and you can learn when a good time is for you when you’re ready to pull out and turn those ads off and try something different.
You do also have the option of setting up rules if you want. So I peek in on my ads frequently, especially if I’m running to a hot or warm audience of website visitors or people on my email list. There’s a fairly small audience for me, so it’s very easy for me to exhaust that audience quite quickly with my ads. And when I say “exhaust that audience”, what I mean is there’s too much budget to too few people, and those people are starting to get very high impressions where they see that ad over and over in their newsfeed and they get tired of it. They’re not purchasing off of it, they don’t want to see it anymore, and they’re inclined to click “hide ad” or “hide all ads”. And that is a huge problem. As an advertiser that’s a huge knock against you and Facebook takes that into consideration when choosing who to show your ads to.
So those impressions, if you’re getting creeped with those impressions over time, you can actually set up rules on your campaigns to say, “once this ad set hits 3 impressions per person, shut the ad off automatically.” So you can set up different rules at different stages of the game depending on what you know are thresholds that you want to contain and manage. But if you’re not looking at these things periodically throughout the week, you’re not able to react or see these things on time.
Rich: It sounds like there’s a whole other rabbit hole we can go down, but this has been an awesome, great start. Amanda, where can we find you online?
Amanda: You can find me everywhere online at @thedigitalgal. The best places to find me are on Facebook, you can also find me on my website, thedigitalgal.com.
Rich: Awesome. Amanda, thank you so much for coming by today.
Amanda: Rich, thank you so much. I really appreciate it.
Amanda Robinson loves helping entrepreneurs and small businesses create effective Facebook ads by busting myths and sharing valuable insights that are proven to engage their audience and fuel their marketing strategy. Follow her on Twitter and Facebook for up to date tips and advice to help you drive results.
Rich Brooks is the President of flyte new media, a web design & digital marketing agency in Portland, Maine, and founder of the Agents of Change. He’s passionate about helping small businesses grow online and has put his 20+ years of experience into the book, The Lead Machine: The Small Business Guide to Digital Marketing.
Podcast: Play in new window | Download