Here’s how we spent 3.5 million dollars on Facebook ads in 56 days
and 7 simple tips that can help you scale your ads profitably…
Facebook Ads can be widely profitable if used correctly…
Today I wanted to create this quick (lengthy) case study on how I was able to scale this campaign from $50 per day to over 3.5 Million, profitably. I’ll speak on why I would like to spend more but the one thing that’s holding me back.
What I will also cover is the 7 crucial learning lessons from this campaign and how they can help you scale any Facebook campaign with ease. If you guys aren’t at this level yet don’t worry, I just want to give you a heads up that just a couple years ago I was losing my ass on Facebook trying to figure out paid advertising so what I’m about to show you can easily be duplicated even on a smaller scale.
Lesson #1: Doesn’t matter how much you spend if your offer doesn’t convert (KNOW YOUR NUMBERS)
When I first started running this campaign our allowable CPA was $150 that meant that I could spend that amount per customer and it would be a success. Well, the first initial ads that I launched, after 24 hours sales started coming in at around $383.20 per lead. Obviously, you can’t run a business losing money so we went to the drawing board to evaluate the data.
We started this offer with this setup: Facebook Ad > Advertorial > VSL (Offer) > Order Form > Thank You Page
What we saw was that the CTR hovered between 8% – 12% on the ads. This high CTR made the CPCs low, which in our case we need the CPCs to be low in order to scale so this was a good sign. From the people that actually clicked the Facebook ads around 38% of them read the advertorial and clicked over to the offer. These numbers were good but where things got iffy was when they actually got to the offer only 2.84% of the people actually clicked over to the order form and from that number, only 15.77% actually purchased which left us with 0.45% conversion rate. Without going into much detail which would make this post super long, what we ended up doing was switching from a VSL to a full-blown HTML sales letter and switched the order form in order to be more in line with the offer. After doing this we were able to get the overall conversion up from 0.45% to 1.26% and got the order form CTR up to 5.03% from 2.84%.
In the past, I would have targeted a new audience in hopes that the CPA dropped and just raised the budget on it. NOPE!! That’s not how it works, you need to know your funnel front to back and know that if you send 100 people through, X amount will convert and from there you can spend endless amounts of money and know with almost exact certainty what will come out on the other side.
From here it was off to the races, I no longer had a daily budget. Now I was able to do some magic and start bringing in the sales.
Lesson #2: Your initial targeting is the lifeline of your campaign for one reason
Now, this step is a little tricky and most would probably put it before knowing your numbers which is justifiable. Here’s where you need to make sure you know your audience and make sure your ads are getting eyes on it from the right people. If I run a finance offer to people in the dog niche, of course, my CPA will be high because I’m not locked in on my customer avatar unless I have some super elite data that says my potential customer has X amount of dogs. This particular offer was in the finance niche so I started off by doing some research in Facebooks Audience Insights of a big competitor. From there that was all I needed, I knew what sites they visited, what TV shows they watched, how old they were amongst other useful information that could be used to create a starting audience. I took this information and created 5 target groups that I wanted to go after and made sure that the audience sizes were over 9 million.
Why 9 million you might ask?
Well because remember before I said that we need the CPCs to be low, from my past experience I knew that the only way to get this was to go after targeted, large audiences. Also, getting a 9 million audience size in this niche it super easy but if your niche didn’t have this many people I’d say just group interest or make your audience broader.
Now the reason why this part is important is that of Lookalike Audiences. What I’ve learned in the past is that the more targeted I am with this initial audience, the actual converters (the people that purchase) will give Facebook better data on these people for when it’s time to create the actual lookalike audience. To get more advanced with this, I only ran this campaign to mobile users on Facebook (No desktop, No Instagram, No Instant Articles, No Right Side Ads) what I’ve noticed in the past is that my Lookalike audiences will also be all mobile users which allows me to scale to the moon.
#Lesson #3: The one change that increased Ad CTR by 120%
As I mentioned earlier, we wanted to get low CPCs which meant that we needed to have ads that not only captured users attention but also made them click. What I started to do after multiple tests was removed the actual “Learn More” button from all of my link post ads and just didn’t use a button. I can’t begin to tell you the improvements that we saw with this campaign after this. In my opinion, I think it makes the actual ad look more like a news article instead of an ad which builds trust with people rather than making them feel like their being pitched.
#Lesson #4: Stop Anything That’s not in line with your CPA and Scale Whatever is making you money.
Once you know your numbers, you should be able to stop any ads or ad sets that aren’t in line with your CPA. Some people will say that you should spend 3X whatever your product price is, or whatever your potential profit is etc. That applies just to the testing phase, I don’t necessarily care once I know we have a working funnel, once I know the funnel works I stay under my allowable CPA and look at the ads on a 7-day basis to go back and see if the CPA is under my allowable. I rarely cut off an ad based on yesterdays stats or anything shorter than 7 days because with Facebook, you never know what the algorithm is doing and your CPA could fluctuate on a daily basis.
Far as scaling, I’ve heard people in the past say that you shouldn’t scale more than 50% of your current budget and I’ve also heard that you shouldn’t scale at certain times in the day. With this campaign, I’d see a particular ad set with a $50 CPA, in that case if it was prior to 5pm I had no problem increasing the budget to whatever I felt comfortable with. In some cases, I’d increase the budget from $1000 to $3000, watch to see if the CPA holds up for a couple of hours and then raise it to $10,000 right after that.
I think we’re able to do that in this case because we know the funnels working and the prior history of that ad set supports the drastic increase. I’m basing all of this off testing so I have data behind this method. I will say that when I’m working with clients that have lower CPAs let’s say around $50 then I notice that I can’t increase the budget so frequently and at large increments.
Lesson #5: Creating LAA audiences that remain the driving force moving forward.
I made sure that Facebook pixels were placed on the Offer Page, Order Form & Thank You Page. What this allows me to do is capture users at specific parts of the funnel with each step of the way being more valuable than the other. What I’m then able to do is create various different percentage LAA off each action taken. I will first create 1%, 5%, and 20% audiences. From there I run all converting ad images along with converting copy to these audiences which range from 2 – 41 million in each segment. I also take the emails of people that actually purchased and create audiences following the same rules. I have no problem running new ads to the same audiences as I’m sure there’s overlap but I never ran into an issue with increased CPA so I just run them.
At this point when I find a lookalike audience that’s working I will duplicate it in the same campaign and change the bidding from automatic to manual with a manual bid of 3x my CPA with a budget of 1k until I get traffic then just adjust it from there. Manual bidding is really tricky, I have no specific technique for this it’s just a matter of playing around with it to see what works and doesn’t work.
Lesson #6: The trick that allows us to keep frequency down and lowers CPA 66%
When you have large audiences of this size you would rarely run into any issues with frequency if your budget is really low but when you’re spending 25k + a day on one ad set you could have an issue. That’s why here I like to create an audience of individuals that just land on our advertorial page (clicked on our ad) this way I can exclude them from the audience. I like to think that this excludes people from seeing my ad that were just curious or took the specific action that we wanted them to take.
Building this type of audience can get large pretty quickly, at the moment I believe the audience is at around 1.2 million people. I can get pretty creative with this audience which I will in the coming days as I will start to split it out and run ads speaking to them in a different tone based on where they landed in the funnel. As of right now, I’m running a retargeting campaign for this audience and I’m seeing $50 CPAs with around 100-200 conversions a day from people that have seen the offer more than once. Not bad when you think about it, that’s 1/3 of the cost of a cold prospect.
Lesson #7: You’re really at the mercy of Facebook, don’t let this be your only means.
Believe it or not, I could have scaled this campaign even more on Facebook (we’ve spent around the same amount on content networks and google) but was limited by budget. When I started this campaign it was a brand new Facebook account with $0 spend so I slowly had to build up the account so that Facebook would lift the cap. It started from 5k to 10k and up from there. Right now it’s capped at around 90k per day and I’d like to spend more than that but Facebook won’t budge so we constantly ping chat support every day in order to raise the budget.
One way around this that we are getting ready to test is running this same campaign in multiple ad accounts. What this would mean is that if we have a 90k a day spend cap on both accounts then we should be able to spend 180k a day across the board.
I say this because, if you have a campaign that’s really working don’t just rely on one traffic source. If Facebook were to disable our account today we would still bring in 500-1000 sales per day from the other traffic sources. I know that this really hurt us in the beginning when the offer was fresh, we were getting CPAs so low and spending so fast that we would spend 60k before 2 pm and go the entire day without any more spending. Imaging what would have been possible then!!
Whew! Ok, that took a little while to put together haha, I hope you guys can grab some nuggets from here. Hopefully, in the coming weeks, I’ll get some more data from this campaign as well as some others that I’m running to share some more.
If you guys have any ideas on how to scale this even more on FB comment below!
CPA: Cost Per Acquisition
CPC: Cost Per Click
CTR: Ad click-through rate
VSL: Video Sales Letter
★ This post first appeared in ClickFunnels (Official) Facebook Group – It has since been deleted and taken down.