Detailed Facebook ads guide from Amadi15. The previous part is available at the link: https://cpa.rip/en/facebook/experience-with-facebook/
Trying to scale, it’s been really hard, but I’m gaining more traction with some interesting changes. Spending between $700-$1000 profitably on FB a day just so nobody is misled. This is like all I do and based on my own tests and spend on my businesses.
It’s annoying when I hear advice for ads, but they leave out super important details and you have no idea who to do after a certain point, so in this guide I tried to cover everything. Please let me know if I missed anything.
CAMPAIGN LEVEL
Ad set budget conversion campaign
Structure: 3 campaigns in total. 1 TOF, 1 Retargeting/MOF/BOF, 1 DPA Catalog Sales.
TOF = Top of Funnel. BOF = Bottom of Funnel. DPA = Dynamic Product Ads meaning the product that this person lands on will be delivered to them if its in the catalog you have connected to your DPA ad.
ADSET LEVEL
Pixel: Set up properly using the FB app if you’re using Shopify. Configure your events from most important to least important: Purchase > Add Payment Info > Initiate Checkout > View Content > Search.
You might be wondering how I have 6 events there? Well I disabled the “value optimization” option so I get 4 more slots and ever since, things have been doing much better. I also convinced my buddy to do the same, and things also got better for him. This will make it so you can’t use value optimized adsets just as a heads up. Do this at your own risk, but I am only seeing better results. I would test it no matter what though, because you can switch it back after 72 hours. That’s my opinion.
Conversion event: Main pixel + Purchase
Budget: $20
Start time: Next day at 2 AM
Exclude: Customer list from Klayviyo, 30 day PUR, 30 day site visitors, 30 day all page engagers
Locations: USA / Canada first, then scale to Europe/AU after you see results
Age: I do 18-54, before FB took the ability to see data based on specific age groups, I noticed 55+ audience was much more expensive and I didn’t see great results over a lot of spend. I used to just do 18+ but after seeing that I don’t anymore. The older age group might be great for you though.
Gender: Whatever works best for your product. I do both for unisex products, but you can single them out if you like, I think more options for FB is better.
Targeting: I have been selecting detailed targeting expansion as my one ad account has gotten an update that makes it so you can’t remove it. I don’t want to have a bunch of successful ads then FB switches it and performance dies. I am selecting very broad interests for the most part that are related to the product. Some of the interests are not related at all though. Think outside the box and just try stuff, once I targeted nurses with a product that had nothing to do with them and it was my best adset, I have countless examples, so always test. I don’t even pay attention to audience size as everything is expanded and hitting 100m+ people anyways.
Placement: Auto
Optimization for Ad Delivery: Conversions
Attribution setting: 7 days after clicking or 1 day after viewing, this is a new change for me that I did when I changed over the value optimization. I always did a 1 day click as my products are typically purchased within one day. I know it doesn’t make sense, but what does on FB? Lol. Test it and see if they perform better, I’m seeing mine do better.
ADS LEVEL
Tip: For the ads I will share a little thing I do. I have a campaign called “make ads here” where I have one adset 65-65 age incase you do age sensitive products and forget, with all my ad templates in that one adset. So I changed the image and the link in the ad section and it’s done in 2 seconds! Copy the ad ID from your Page Posts section under the proper page and use that under existing ads when making new ads. The reason you do this is to maintain social proof and ease of use.
How many ads per adset? I’m doing 1-3 as all my budgets are at $20 or if you’re going to spend more per adset you can go higher, you just don’t want 10 ads in an adset with a $20 budget then it gets split 10 ways and you’ll never have enough data. To add to this I’m seeing 1 ad per adset work well, collection ads do great here. What I mean by collection is that I do a single image with many different products, then send them to a collection page showing those products first, with many under. More options can help increase conversion rate, especially if you got DPA ads going, then they will get targeted with the specific things they looked at in that collection.
I will make another post about how I make my ads, that’s a loaded topic, same with retargeting.
SCALING AND OTHER COMMON QUESTIONS
Scaling: What I’ve been doing is taking my winning ads and creating 10 $20 ad sets and giving them all different new interests. I’ve gotten to $1000 a day profitably doing this so far and I’m still going, it’s looking good! Next is to scale those into different countries, take the exact same ones in the USA that are working and hit Canada, then I normally target specific places in Europe, but recently I’ve just been going with Europe and I’m seeing good results. After that hit other countries that might work for you, I like AUS. If your wondering about LLAs they haven’t been working to well for me, the best one was a 10% LLA based on all page engagers. I will keep testing them here and there, but it seems like interests are doing better now and it makes sense.
I have been very hesitant to do this on winning ad sets, but I think I am going to start soon, I haven’t yet, but it’s been recommended to me many times by reliable sources. lol. Increase everything that’s working every 3 days by 20%. I know, I know, 20% of $20 is $4, brutally slow scaling. But you know what’s worse than slow scaling? No scaling and losing money every time you launch anything over a dumb $20 budget. Haha. Slow and steady, and if you have like 30 $20 adsets $120 a day doesn’t seem so bad.
How much do I spend before I kill an ad? I do $30 – $40 per ad if the stats don’t look good at all, I’ll do a max of $60 without a sale if stats are looking good. This will depend on your product cost / AOV as well. Mine cost $21 – $40 with an AOV of $65 to $70. The reason why you should wait is that FB will report sales up to 72 hours later. On top of that, they won’t attribute it to that day, but the day when FB decides to show the data. So you can launch something on Monday, kill it Tuesday and see a conversion Wednesday, madness I know. Also while we are on the topic, FB will also only report back the highest value event that happened, what I mean by this is this? If someone comes to your store’s home page, goes to a blog post, goes to a product page, adds to cart, checks out and adds payment info then leaves, the only event that will be sent back to the ads manager is the “add payment info” event. That’s why FB stats make no sense half the time right now, and why you need multiple tracking solutions, in my opinion anyways.
How I kill adsets: So when I have 3 ads in each adset, I won’t kill the adset if it’s failing, I look at the ads and kill only the ads that are doing bad and let everything else keep running. This has been working really well and I’m keeping my spend higher by not killing as much and the other adsets typically make up for the failures. If all 3 ads are looking really bad, I might kill it early though. For single ad adsets, I kill them how I mentioned above.
I’m pretty sure this covers it all, or I’m drawing a blank, but please let me know if you want some clarification on anything, or a better explanation, I will do my best to answer you! Thanks for getting this far and I hope I provided you with some value! You may have seen my last post, but that’s how fast things are changing all the time. Always testing presents better ad buying opportunities so I’m going to take them and in this case share them.
