As you might have heard (and in case you haven’t yet) I wrote an EPIC step by step guide on how to run the exclusive crypto offers from Grabads on Native traffic sources – you can read it here. Sadly the FB gods didn’t allow me to simply promote this guide to all of ya’ll aff marketers out there…
So, I had to come up with a work around because the pricks at Facebook wouldn’t allow me to promote/boost post the case study guide I created simply from the page saying it violates policy about cryptography lalala… even tho it’s an educational piece on how to run a native campaign.
I first tried various ways like leaving out the trigger words in the ads, and using semi-related imagery – but since I had to link it everytime to my actual guide, it kept getting disapproved because Facebook reviews whats on your landing page whilst doing approvals.
So, I signed up for lead pages and created whats called an opt in page, (you can see the actual page here) – then when people opted in, they receive the link to the guide on the thank you page which also allows me to gain more followers on social media, because I ask them to follow me if they don’t yet, and It also enables me to plug my design company Banners&Landers as well.
The GOOD NEWS is this method works, at the time of this writing already spent $312.11 – What the BAD NEWS is though, that the cost to acquire 1 opt in to the list to grab the guide is super high.
As you can see, I have a Website Conversions campaign, and one is being optimized for the VIEW OPT IN PAGE pixel event, the other is optimized for the THANK YOU PAGE LEAD/VIEW event (this fires when someone actually types their email and presses the button and they get taken to TY page).
My lowest price right now is $3.34 per person that signs up to the list, which is astronomical. My goal would be maximum $1 per person.
From my experience I know the following methods to lower CPA.
- Try new ads – video ad & controversial ad that engages the users. – That means I gotta test 2 new ads.
- Run a PPE campaign to my best performing ad in current adset – I was running a dynamic ad, so I gotta swift through and pick my best performing ad that got me the most conversions and run PPE just to that one!
- Generate a lookalike audience from my viewed the opt in page and then run an adset using the best converting creative I am running PPE too.
- Try new audience
So this is what we’re going to be setting up today and then run for a day or two and see what happens.
I picked the best ad using the Dynamic -> Combinations option from the dropdown menu FB offers. Now, I am a bit skeptical if this is really the best ad to focus on, because the ammount of reach it has is only around ~500 yet the other images/headlines have 2000-3000. But, I am going to take FB’s suggestion and run with it to see what happens..
SIDENOTE: Dynamic creative is good, but one of the cons about it, is it won’t evenly rotate the ads which you can manually setup if you do 1 adset for each ad. Which takes a ton of time but the data can be more consistent.
Sadly when I went to Business Manager > Page Posts and looked for the above ad there, I couldn’t find it. And when I clicked view post with comments using the Preview button that ad didn’t show. So I went with another ad, and grabbed the PostID from the actual address bar of Chrome Browser and created the PPE to run to this ad ID.
Now I am very curious if our CPA will decrease thanks to the PPE ad overall if we look at lifetime for our Opt In objective WC campaign using dynamic ads. My thoughts are dynamic ads suck and it’s better to do everything separate, but since time is of the essence and I needed to get this up fast – I didn’t break down the ads into separate adset combinations, etc.
36 hours later update: Ended up setting up #2 and #4 from above. The results are far from spectacular but the new audiences did better, PPE didn’t help it seems the original adset because if we look at lifetime, the value actually went up by 10 cents for each conversion and if we look at yesterday’s stats only then it jumped by almost a 70 cents.
So it seems the best way is to test new audiences, running PPE doesn’t help lower the overall CPA quickly (36 hours time).
I am going to end this case study here, because as always I confirmed yet again that doing dynamic ads when cost really really matters isn’t the way to go as you cannot control things as good as if you split things up into one ad per adset and separate adsets.
In my experience, dynamic ads are only good if you are working with bigger cpa goals and a few bucks here and there don’t matter; but when every dollar counts it’s way better to run separate adsets with the same ad in the adset (you can duplicate the same ad multiple times since fb’s algo is random it will give better CTR that way) .