It sucks when your ads don’t get results. It feels horrible.
It can be accompanied by panic about how much money you spent to find out it didn’t work.
You may not know why it didn’t work. You may not have set things up so that you learned anything from what didn’t work. You may not know what to do next.
I wish every Facebook ad was a success. I wish every Facebook ad campaign worked brilliantly the first time. But that’s simply not realistic.
Even though we’ve had a ton of successes for clients (some of them are here– and there are about ten more super cool case studies I haven’t had time yet to document)…
There are many failures, and that’s just part of all digital marketing.
And life, really…
Michael Jordan said, “I failed over and over and over and that’s why I succeeded.”
So your Facebook ad MINDSET is critical…
Because even after you avoid the list of mistakes I’m going to give you below, sometimes you still need to…
- Gird your loins
- Steel your nerves
- Grit your teeth
- Seize victory from the jaws of defeat
So let’s look at the biggest reasons for Facebook ad failure, starting with the simple and technical and graduating to the biggest and most strategic problems.
Before we get into the list, I have a special freebie for you…
Ok so you opted into that webinar above? Great, let’s get to the list. 🙂
Hey not a big reader? Don’t want to read through this epic blog post? I mean it is 5,800 words long… Watch this video and I’ll talk you through it:
Or just want the short funny summary version? Here’s the ROI vs ROY video (ROI is Return on Investment. ROY is a redneck.):
#1 Insufficient ad testing budget
If you don’t spend enough, you have enough money to test enough ads to find the ads that work super-well, especially if your goals are lead generation or ecommerce sales.
If you really go cheap on your ad budget, you might not find anything that works.
You might run out of money before you get to your goal.
You need enough budget to test at least 50 to 100 ads. At least.
So if we had only run 76 ads, maybe we wouldn’t have had a single sale!
But one of those ads got 11,800% ROI. Every dollar we spent on it made them $11,800.
Another got 4,000% ROI.
Yes, 150 ads is a lot of work… but it’s worth it.
Sometimes when I look at someone’s Facebook ad account after they’ve said, “Facebook ads didn’t work.”
I see their account only has 10 ads in it and I say, “No, YOU didn’t work.”
You need to do more than that.
What’s the bare minimum you need to spend on Facebook ads?
Most agency folks and consults I’ve spoken to agree that a good start on Facebook ads requires at least $1,000 of spend.
How many things do we need to test? We typically need to run 50-100 ads per product to dial in what is going to work. We need to find the targeting criteria, images and copy that will sell. Some won’t work at all. Some will.
For ecommerce, we can look at your margin and cost per sale and estimate how much we want to spend on each ad before saying it isn’t going to work and turning it off.
- The more proven the product is (already selling online) the lower that number will be.
- The newer and less proven it is (hasn’t sold online or hasn’t ever sold anywhere), the higher it will be.
- If you have a lot products, the more similar they are, the less ads we need to run.
- The more different your products are, the more ads we need to run, because people will respond differently to them.
Some of your products may be more popular than others. Some may require slightly different targeting. Some may be a good first buy and the others might be better sold as follow-ups via email. None of that is clear at first. It becomes clear over time.
We won’t know what a reasonable cost per sale is until we get an ad that sells. It might be $5, $10 or $20, or more. As we run the ads, we’ll find a number that convert all at different costs-per-sale.
For lead gen, it might be reasonable to go for a $2-5 lead in the business-to-consumer world.
In business-to-business, it could be $10-50 or even higher. And that’s fine when you’re selling things that cost from $1,000 to $100,000 or more.
Running Facebook ads to see what works is kind of like day trading, but you only get info about stocks you trade, and there are no mutuals- you can only buy single companies.
As we discover more ads that are converting, we can revise that cost per conversion target number. If we think $20 per conversion is reasonable, and we need to run a 100 ads, that’s $2,000, just to give you an idea.
The more ads we run, the more we learn and the more ads we find that convert.
This process is a function of the spend more than the time. So if you spend $3,000 in a month or $5,000 in a month, you’ll get there faster.
At the Brian Carter Group, we don’t increase our fees until a client spends $1,000 a day or more (it becomes a lot more work because the ads burn out faster, because the audiences are finite and they’ll all see the ads and tire of them, and stop responding, and the cost goes up, so we have to refresh the creative).
As you might be able to conclude by now, it’s impossible to set an initial ad spend without some degree of guessing. And your budget- the amount of money (hopefully profits from other marketing channels) you have to invest in growing the Facebook channel- plays a part too.
Knowing you need to spend between $1-5k a month gives you an idea what it costs to run a bare minimum professional Facebook ad campaign.
#2 Creating the wrong ad type
I’ve written and spoken extensively and I’ve taught elsewhere about how Facebook has ten different ad types and every ad type is goal-oriented.
You could go for engagement or video views or website traffic or event attendance or website conversions, but you have to know what goal you want from these ads and choose the right type.
- If you boost a post, you’re going to get likes, comments, and shares, but it’s going to be really expensive to get traffic or conversions from that boosted post. If that post has a video in it and you boost it, you’re going to get engagement on that post but not as many video views as you could if you ran a video view ad.
- If you want to get website traffic you could run the website traffic ad, but a lot of people have run that type of ad and said, “Wow. I got a thousand people to my site but no conversions, no leads, no sales.”
- That’s because there’s another type of ad for conversions. It’s called the website conversion ad. If you run that one, you have to install the conversion tracking, which is mistake number five. If you don’t have conversion tracking running properly, then the website conversion ad will function like a website traffic ad and only get you traffic.
The reason that that’s so important is that, no matter who you target on Facebook, Facebook will first show your ads to the people within your target group who do the most of whatever your goal is.
Let’s say you want to target moms who make more than $50,000 a year. If you boost a post, or promote post engagement, then…
Well, there are a lot of moms who make $50,000 a year (17 million in the U.S.). They can’t show your ad to all of 17 million of them at once, so first they’re going to show it to the ones that they know do a lot of liking, commenting, and sharing. Because you chose the post promoting type of ad. That’s the goal you chose.
And those heavy engagers are not the heavy link-clickers or heavy converters… if you want the latter, you need to choose a different ad type.
If you want to get ecommerce sales from those $50k+ moms, you might run a website conversion ad and send them to an ecommerce site. Then they’re going to show it to the moms who make $50,000 who’ve converted on external websites before.
Do you get the point? They’re going to show it to that subgroup of your target audience that’s done the goal that you’re telling them to do with the ad type you chose.
The dumbest thing you can do is only boost posts and expect to get more than likes, comments, and shares…
…because that means you’re just not aware there’s nine other types of ads and you haven’t gotten into the Ad Manager or Power Editor and you haven’t created those other types of ads.
You have to get in there and create the right ad type for your goal.
#3 Not creating enough ads (not testing enough targeting or creative)
Our best case studies where we’ve had 2,200% ROI or we’ve gotten incredible results for low-cost leads, and all those kinds of things, and where we prevail with the most difficult circumstances and where we drive the most incredible results are when we’ve created the most ideas and put them in front of customers and we see which ones customers respond the most to.
If you only create one idea, it’s a real crap shoot. The chances that you created that home-run ad is very low.
If you create a hundred ads, there’s a much better chance that you found the right combination of targeting and image and headline that people will go crazy for. Then you’re going to get amazing results.
You’re looking for an outlier…
Statistically speaking, WHEN WEIRD IS AWESOME and awesome is weird.
The way I talk about it is I say look at any sport, like the NBA, where there are some amazing athletes…. people like Michael Jordan, Kevin Durant, Steph Curry and Lebron James.
These guys are freaks of nature and have drive and practice and they’re very exceptional people, but they’re only four guys out of millions that have ever picked up a basketball and out of thousands of guys that have been in the NBA.
If the NBA had only ever had four players total, the chances that we’d have those four guys is very low.
When you create an ad campaign and you have an ad account, the more ads you run in there, the more ideas you create and test, the better chance you’re going to have of having something really awesome happen.
The more ideas you put in front of your customers, the better chance you have of them going crazy for one of them.
It doesn’t have to happen all at once.
- You may start by testing your targeting, find the right way to target people.
- You might test some images, find the best image.
- Then you test the headlines, find the best headline.
It might happen over time, one thing at a time. Over the course of time, you’re going to test fifty, a hundred, hundred and fifty, two hundred, maybe a thousand ads at some point.
You get to stop at the point where you decide, “The results are so amazing that I don’t want/need any better.” However, it may burn out at some point, so you should always be testing.
Don’t stop with laziness in the beginning or you’ll lose.
- You’ll cost your company a bunch of money.
- Your customers won’t be excited.
- They won’t care.
- You’re just not going to do as well.
You have to test a lot of things. You’ll get results while you’re doing it, and you’ll get better and better results as you go.
#4 Putting too many ads into one ad set
The way that Facebook works, if you put ten ads into one ad set, it’s going to figure out which one is performing the best and it might only show that one. It’s going to decide on its own how much to show each ad. It doesn’t really give you any control over how much each ad in the ad set is shown to people.
Like we talked about earlier, there are different types of ads. If it’s a post promotion ad, it’s going to find the ad that’s performing the best for getting engagement and it’s going to show that one the most.
Sometimes it will show the one that doesn’t have the lowest cost per engagement, and you have to do your own optimization and pause the expensive one. That’s part of your job with the Facebook ads.
If you put ten ads into the ad set, some of them may never get enough reach and be put in front of enough people to really be tested. You have to only test two, three, four ads at a time, or they just won’t get enough reach for you to know that they were tested well.
What that means is that, if your ad set is to a certain target and you have all your different ad sets set up by different targeting, sometimes if you want to test a new ad you have to pause a good ad.
You might want to change the name of the ad. You’ve got the name of the ad itself, and then you just append to that, “Restart later.” Pause it and run some new ones and see if they do as well as the best one.
Another approach is to create a new ad set that’s got the same targeting and apportion some specific budget to the new testing.
I’ve got some accounts where we have
- BEST PERFORMING AD AD SETS
- TESTING AD SETS
We’ve got a specific amount of money budgeted for testing and a specific amount of money running into the best ads.
That’s a smart thing to do too, because then you’re optimizing your budget. If you were investing, wouldn’t you put most of your money in the stocks with the biggest return?
You could have 80% of your budget going to the best-performing ads and 20% of your budget going to testing new ads.
You should always be testing new ads, because eventually your ads are going to stop working. If you’re showing the same ad to a finite audience, and all audiences are finite, eventually everyone that’s going to like or be influenced to click on that ad or do something with it or watch that video, eventually they’re all going to see it. Some of the people will never respond to it, so the performance is going to go down. It’s going to burn out. The costs will go up.
If you don’t have other ads that you’ve been testing over time that are on deck ready to go into your best-performing ads group, you will have to start over from zero and you’ll have a temporary dip in performance.
It’s not a bad idea to have best-performing ad set and testing ad sets.
#5 Not installing conversion tracking properly and testing it ahead of time
If you’re going to run website conversion ads, you need to have conversion tracking set up with either standard events or custom conversions.
That’s a little technical. I’m not going to explain how to set that up here. Facebook has some great help screens.
I will tell you that we found custom conversions to be more reliable overall than standard events.
You do need to get that set up so that you know which ads are working and which ones are not.
It’s not good enough to track Facebook ads in Google Analytics. You need the conversion tracking from Facebook in your website or landing pages to:
- Tell Facebook ads which ads are working,
- So that you can optimize at that level and only run the best-performing ads and
- Stop the ads that aren’t performing well.
Like I said earlier, if you’re trying to run website conversion ads and you don’t have conversion tracking set up they will run like website traffic ads. Often you will get traffic with no conversions, which is a waste of money.
When you set up the conversion tracking, then you need to test it before running the ads. You don’t want to waste any money, so you need to get the pixel in. You need to check the pixel dashboard and make sure that it’s firing. You need to make sure it’s firing on the URLs that you have it on.
You need to, if you do custom conversions, to find that custom conversion by the thank-you page, or whatever confirmation page shows up after the conversion is complete.
If they complete a lead and then they go to a thank-you page, or they buy something and go to a confirmation page, the URL of that thank-you page or confirmation page is your custom conversion URL. You need to set that up, define it, and then make sure that you’ve gone to it again, and then that the custom conversions dashboard says that it’s active.
Once you’ve done that, then you can create and run ads that are website conversion ads that work. Make sure you’ve got all that stuff set up and verified ahead of time.
#6 Pushing on what you want rather than following the customer’s lead
This is a mistake I see a lot of businesses make on Facebook, is…
- they post something and nobody responds,
- the response is very low, and
- their reaction is, “Well, we need to advertise that more, because this is an important thing and we want to make sure we hammer this into people.”
That’s the exact opposite response of how you should look at these things.
Facebook is a customer laboratory for you, where you can put things in front of customers and see what they like and see what they’ll respond to and what they won’t. Whether it’s engagement or videos or leads or sales, you can find out what works and what doesn’t.
You’ll do better if you go in the direction of what customers love and what they respond to. If you do that…
- Your costs are going to decrease
- Your profits are going to go up
- Your customers are going to love you more
If you go the opposite direction, where you say, “No. This is our initiative. This is the thing we’ve decided is important. We need to hammer this into customers even though they’re not responding to it.”
- It’s going to be more expensive
- You’ll get less results
- Customers won’t like you as much
- You’ll seem out of touch because you’re not following their lead
The second one is like having a conversation with someone and not listening. You say something to someone, they don’t like it, and you say it again louder? That’s not how relationships work.
Follow the customer’s lead.
This is a really cool time to be marketing, because for years and years and years we’ve wanted to know, as businesses, what customers like and don’t like. There have been a lot of ways to try to figure that out- through surveys and focus groups and all that kind of stuff.
Every one of those customer discovery methods has flaws.
One of the biggest flaws they have is the customer knows that they’re being asked these things. That’s very different.
How a customer or a person acts when they don’t think they’re being watched, what they will buy, what they will do, is different. It’s more true and it’s more accurate than when they’re in a focus group trying to impress the other panelists or the person asking the question. Or when they’re taking a survey talking about who they wished they were instead of who they actually are.
You want to know how they actually behave, not how they wish or intend to behave.
When we put things in front of customers and they don’t realize it, they’re not really thinking about how we’re monitoring whether they did or didn’t respond to the post or the ad, but we are.
It’s like a top-secret survey they don’t even know we’re doing on them.
While we’re getting Facebook results, we’re also constantly learning from them without them realizing that we’re surveying them.
That’s what Facebook posts and ads do for us, is that they are a huge source of customer intelligence and they’re much more accurate than some of the other methods we’ve had for a long, long time.
#7 Bad copy
That means copywriting. Copywriting is a fundamental marketing discipline.
It’s very important to understand that different phrases and different words affect people differently. There’s been work in this area for over eighty, almost ninety years. People in marketing have been trying to write things that get bigger and better results from customers.
You need to understand the fundamentals of copywriting. There are many books out there, many courses out there, about copywriting.
Beyond that, there’s even psychological research about what words people respond to the most, what words are positive for people, negative, arousing, stimulating, which ones men like, which words women like. There’s a lot of good research out there too.
There are places like BuzzSumo that analyze blog posts that work and don’t.
There’s a lot of data out there, and then you can create your own data. You absolutely should because your customer group is going to be somewhat unique and is going to respond uniquely to your offer and brand, so ask yourself:
- Which subject lines get them to open your emails?
- Which blog posts get the most attention?
- Which ads get the best results?
- Which posts get the most engagement?
If you’re smart about it and you’ve learned the basics of copy writing, you have thought about what are the benefits of my product or service, what is my unique selling proposition… there are a whole bunch of fundamental copy things you need to know about your business. If you’ve figured those out, you can test them with Facebook ads and find out what works the best for your customers.
You may start with bad or mediocre copy, but you don’t have to stay there.
Facebook is a customer laboratory, and testing is the process that saves us from ourselves and our bad ideas and our office politics, and it helps us get better results.
#8 Not split-testing landing pages
Fundamental thing to understand about getting results online is that a landing page is anywhere you send somebody. It could be your homepage. It could be a leadpage. It could be a click funnel’s entry page. It could be any webpage. A landing page is the first page they go to. You want them to do something when they get there. Maybe you’re trying to get a lead. Maybe you’re trying to get them to sign up for your email. Maybe you’re trying to get them to register for an event. Maybe you’re trying to get them to buy something. Whatever it is, you’re trying to get them to do something.
You can split test that landing page to see which page gets a better result. What’s the conversion rate? What percentage of people who went there did the thing you wanted them to do?
It’s not as smart to do sequential testing, which means we’re going to have our website look like this for a week, and then we’re going to change it and have it look like something else for a week. Because who knows what happened this week versus last week? Maybe there was a holiday. Maybe there was a national crisis. Maybe the economy went up or down. Things can change in time.
Split-testing allows us to see how people responded to different things at the same time, which eliminates a lot of variables that could screw up our results.
Split-testing is really important. There are a lot of different platforms that allow you to do that. There’s lead pages, click funnels, Unbounce, Optimizely. There are WordPress themes. There’s a WordPress plugin that lets you split test your blog post titles.
Split-testing is a fundamental part of digital marketing. It’s really very similar to testing multiple Facebook ads. You need to have this understanding that the cost per conversion, whether that’s a lead or a sale online, comes from a couple different things: the cost per click, how much did that traffic cost to get there; and then how many people did you have to send there per lead or sale, that’s the conversion rate.
If you can increase the conversion rate or lower the cost per click or both, you lower your cost per sale cost per lead. You increase your profits. These are your two biggest levers.
We’re back to copy writing, images, all those things you put on a landing page, the format, the layout, whether there’s a video or not, what kind of video it is.
Let me add one thing here. When it comes to testing ads, testing copy, testing landing pages, the research says it doesn’t matter how long you’ve been a marketer, you don’t get better at guessing. Your guesses do not get more accurate. You can get better at writing copy. You can get more disciplined at the testing process. You can get more creative. You can get better at understanding customers, but you can’t get more right. You still have to test a lot of different things.
Testing is liberating, too, because not only does it increase our chances of getting higher profits, it’s a profitable activity, but it liberates us from the personal opinion, feeling, emotional, messy side of creativity, where somebody’s attached to the image they like or the headline they wrote, or whatever. If you can get everybody on board with science and say, “Look, these are all great ideas. Let’s see what the customer likes. Using a scientific process, we’re going to find out.”
Then everybody on the team is free to have good ideas and bad ideas. No one has to say, “I’m a good person or a bad person because I had a good idea or a bad idea.” That’s not what it’s about. It’s about a process, trying to understand the customer, and create things, and test things, and get better and better results. It’s just a conversation with the customer where we serve them better and better. Landing pages are a big part of that.
#9 Too many steps before conversion
One thing to understand about digital marketing is the more steps people have to take to get anywhere, the fewer people will get there.
The longer the journey is, the more people die on the way. That’s a horrible analogy.
If you think about email marketing, for example, to get people to buy something you’d have to get them to open the email, that’s step one. Before that you have to get them to sign up for your email list. You have to get them to go to your website to sign up for the email list…
- Go to the website,
- Sign up for the email list,
- Open the email. That’s three steps already.
- Click on the link. That’s four.
- Then however many steps it takes to buy the thing on your site. Maybe five, six, seven steps.
That’s email marketing.
How many people open the email? 20%? 30%? That means 70-80% didn’t open the email.
How many people that opened the email click through? Maybe it’s 20%, 30%. Again, you lose another 70-80%.
When they get to the site, how many purchase? A good ecommerce conversion rate is 1% or 2%.
98% or 99% of the people that clicked on your email, went to your site, didn’t buy.
You lose most of your people at every step you make them take.
With Facebook, if you have to get a fan, get them to see your post, we know Facebook reach is a problem, get them to click the link in the post, that’s a lot of steps to get them to your site.
You can cut out some of the steps just by having a Facebook ad that sends them directly to the site. You cut out two steps. You didn’t need to get them to be a fan. You didn’t need to get them to see the post. Just running the ad got them to see it.
We know website conversion ads, get better people there to convert. Your conversion rate’s going to be higher. That’s even better.
That’s why fan marketing is broken. It’s partly you might have the wrong people. They might not be buyers. It’s also extra steps.
Having too many steps in your funnel
- Reduces your conversion rate,
- Increases your cost per sale,
- Lowers your profits.
Simplify the customer journey and you increase your profits.
#10 Custom programming and generally reinventing the wheel
Sometimes I see entrepreneurs hire a programmer and have the programmer create things that are not as good as stuff that’s out there. There are a lot of software-as-a-service apps out there. Landing pages are a big one. Have a programmer create a way to buy something on your site instead of just using lead pages or click funnels and plug it into PayPal, or something. You’ve got two obvious SaaS’s that are bulletproof and they always work, as opposed to having a programmer create something that might be buggy.
The problem with programmers is you never know if they’re going to be really good or they’re going to do things on time or they’re going to hold your code hostage or what. There’s a lot of problems there. You don’t know if they’re going to provide good customer service. I heard so many nightmare stories with programmers.
It’s better to find an existing service out there that does something you want them to do than to custom program it.
You create a lot of problems for your self by reinventing the wheel and hiring programmers to do custom things that are already out there. Don’t do that. Check and see what’s already out there before you hire somebody.
Uh oh, there’s one more!
THE BIGGEST mistake that will cripple your Facebook Ad Campaigns…
OFFERING SOMETHING NO ONE WANTS
We have sometimes have clients who are entrepreneurs who have a new idea.
It sounds great, it looks great, but we don’t realize until we put it out there nobody wants to buy it.
If it’s a product that no one’s ever bought before, you can test it with Facebook.
Facebook is very affordable- more affordable than Google ads. If it’s a totally new thing, nobody’s searching for it, so AdWords doesn’t make sense.
It might be a new category of things. Nobody’s searching for it, so Facebook makes sense.
Reach the right people, tell them about it.
People go, they check it out.
IF… no matter how you explain it, no matter what you do, you get influencers, you create awesome videos… No matter what you do with it, nobody wants it….
Sometimes you have a business or a product idea and it’s just a DUD.
The good news is that you can pretty affordably test your new product or business or service idea with Facebook ads.
You can even use Kickstarter to test a product idea without putting out money to create the product… because if no one wants to fund it and buy it ahead of time probably nobody would buy it.
Facebook ads and Kickstarter are really good ways to test an idea ahead of time and not commit a lot of time and money and emotion to something that no one’s going to want to buy.
It’s a tough thing, because
- You could totally love the idea.
- You could be super-passionate.
- You can be convinced your logic is sound that people should want it…
But there’s still a chance that they won’t buy it.
I’ve seen numerous situations where it just totally made sense people should want this thing, but nobody wanted to put money out for it.
I’ve even seen products created from customer surveys where people said they wanted it, but when it came time to purchase, nobody bought.
What people say they’ll do and what they’ll actually do are very different.
Until the cash register rings, you don’t really know for sure.
Save yourself some heartache and some money and use Facebook ads to test the idea.
Just seeing if people will click through at a high rate on the ads for it. If they won’t click through a newsfeed ad, a 1% or a right-hand column ad at 0.1% for it and it’s the right target audience, there’s just not enough interest.
You could do a beta, do a website that’s like, “Check this out. Sign up. Put your email in to hear about it when it’s available,” or, “we’ll give you 20% off when it’s ready.”
If you’re not getting a lot of interest from that, then you got a good indicator there’s something wrong. Something’s not right.
It’s good to know that as soon as possible. You don’t want to invest a lot of time, money, heartache into something that no one’s going to buy.
It can be challenging, because there are a lot of mistakes people make. And I see so many businesses make the exact same mistakes over and over. That’s why I write posts like this.
It’s frustrating, because I think a lot of people get excited about things.
They say, “Oh, Facebook ads are going to be great! Snapchat’s going to be great! Live video is going to be great!”
They get very excited about the opportunities, but they don’t see all the traps and pitfalls and mistakes ahead.
I’ve seen so many people make the exact same mistakes and fall into the exact same traps.
My passion is to just tell you guys, as difficult as it might sound, “Here’s some problems ahead of you that you need to avoid.”
I want you guys to succeed. I want you guys to profit. I want you guys not to waste your money.
Please don’t make these mistakes. That is my message to you.
But at the same time…
- There’s never been more opportunity in business.
- We’ve never had better data about customers.
- It’s never been more affordable to try new ideas and get new customers.
So if you’re smart about it, you can succeed with digital and social marketing.