You probably heard often that Facebook is one of the best traffic sources, but as usual, people don’t explain why… they just say its the best. So let me help you understand:
-Very little bot traffic
-Highly accurate targeting options
-HUGE chance for scale
Even though Facebook makes it easy to target your perfect audience, you can STILL blow thousands of dollars $$$ and end up with 0 conversions if you don’t launch a campaign with the right ‘elements of success’.
Let me cut to the chase, here are The top 7 tips on how to setup your FB campaign for success in no particular order:
1. Make sure you use unpublished PAGE POSTS for all newsfeed ads. DO NOT create ads from the power editor/ad manager. This has two benefits, a) it will help ensure that when people like/share/comment on your ad and b) it will make it way easier to manage, especially when you have to moderate and/or respond to comments.
2. Create one adset per element being tested. So if you have one angle, and one image – do not throw multiple angles/images into the same adset. The algo works in strange ways, and unlike Google you don’t have the option to rotate ads evenly.
3. Set the tone for your ads – a lot of the time, the attack of the negative people is imminent. What does that mean? You publish an ad, and right away you will have the ‘good simaratins’ throwing words like it sucks, it doesn’t work, you can buy it on ebay for 1/3rd the cost, etc etc. That’s why it’s important to be proactive. Once you create an ad, go to it, like the post, and ask friends to reply, like, and post some positive comments.
4. Build highly passionate audiences to target, always start with a broad audience (you will know this by typing the most broad related keyword to your product/service/offer and it will have the most people in it) then click NARROW audience, and start adding in relevant interests, ie magazine names, club names, tv shows, tv channels, organizations. How this works is, if a person likes lets say Body Building, and they like FLEX magazine, they are more likely to engage with your ad, vs someone who just is in the broad bodybuilding interest group. The reason is they might have liked a post or video related to body building but thats it; vs the guy who liked a post and also liked stuff from flex magazine.
5. Make sure your audience size is at least 1 million; if it’s too small that means you will burn out your ad too fast, which can cause all sorts of problems (audience burn out, spam complaints, CPA rising rapidly)
6. Include a call to action link in your ad text – use a link shortener like bit.ly or goo.gl or something similar. People are not the smartest, and unless you show/tell them where to click/tap they will post comments like HOW DO I GET THIS? Even though the ad might have a button to learn more, or buy now, or whatever. This is definately an ROI Boosting practice.
7. Be the first to reply and comment on your ad with a link to your offer. This goes hand in hand with what I said above, people aren’t the smartest…
Read more FB related blogs at http://iamattila.com/facebook
Me too, I need a bit more explaination 
Thank you!
Thanks for the post Attilia!
Question: When scaling your campaign/ad across multiple ad accounts, is it 'safe' to use a page post associated with a separate ad account/page/person across multiple accounts?
I.e., if Facebook deems the post and subsequent ads associated with it a 'Policy Violation', would all ads running that post be shut down as well?
Thanks Attila, some great pointers there.
With regards to point 6:
>> Include a call to action link in your ad text – use a link shortener like bit.ly or goo.gl or something similar.
So do you mean that literally in the ad text you would write the call to action and then followed by a Bit.ly link: so it would look
something like:
"Apply Now Bit.ly/dhjfdjd"
Or is there another way that I am missing that can make this look a bit cleaner?
And re point 2.
>>Create one adset per element being tested. So if you have one angle, and one image – do not throw multiple angles/images into the same adset.
So, if I have several adsets in a campaign that I have divided to test different age ranges and want to test 3 different images/angles per adset.
I would simply duplicate these adsets to have 3 copies of the same age range adset and then
have 1 image/angle in each of these?
Would this then ensure that Facebook shows them all evenly?
My split tests told me time after time that the Google shortener gets more clicks than bit.ly.
Okay thank you
And re point 2.
>>Create one adset per element being tested. So if you have one angle, and one image – do not throw multiple angles/images into the same adset.
So, if I have several adsets in a campaign that I have divided to test different age ranges and want to test 3 different images/angles per adset.
I would simply duplicate these adsets to have 3 copies of the same age range adset and then
have 1 image/angle in each of these?
Would this then ensure that Facebook shows them all evenly?
Great tips , few questions
"1. Make sure you use unpublished PAGE POSTS for all newsfeed ads. DO NOT create ads from the power editor/ad manager. This has two benefits, a) it will help ensure that when people like/share/comment on your ad and b) it will make it way easier to manage, especially when you have to moderate and/or respond to comments."
You mean create a fresh ad from the ad manager?
Also not sure I understand this sentence:
it will help ensure that when people like/share/comment on your ad -what suppose to happen than?
He means create the ad content in Page Posts section within BM (copy the Post's ID) and then go to Power Editor, paste the Post ID in the Ad level and launch this way.
When you duplicate the ads making sure all have the same Page Post ID - all the engagement on your ads will be counted for that particular post you're running, therefore your post's social proof will lower your cost prices.
How would you effectively track using 1 single page post however? I also do something similar for a number of reasons, but it seems the URL variables are not passed to the LP with page posts.
Are you using different tracking links per adset then?