Want to Get Profitable on Facebook? You should probably read this post. (23)

Here is the thing. Facebook hands down has some of the best quality traffic on the planet. It is also one of the few traffic sources that allow advertisers and affiliates to reach HUGE scale (think $100,000 a day spend). That being said, it’s by far one of the most sensitive traffic sources to work with as an affiliate.

Even if your campaigns are 100% compliant running e-commerce your ad account can still get banned, your ROI can go from a juicy 400% to -200% as soon as you increase budgets.

As annoying as it is, when you hit a profitable campaign on Facebook the feeling is unreal, you can literally put your foot down on the pedal and Facebook spends like there is no tomorrow.

Our team has spent the better part of a year and half mastering Facebook and here is what we have learned about slaying the Facebook beast.

Now I will start by saying this is what works for us in-house, doesn’t mean it is the right strategies for everybody. Every affiliate has a unique approach to running on FB.

Ok lets dive in!

Leverage Data:

Most affiliates run clicks to website campaigns using targeted interests only. This is usually because they are doing something naughty and can’t exactly integrate their Facebook pixel without accidently revealing their dirty page. While there is nothing wrong with this, you are putting yourself at a HUGE disadvantage. You are using the bare minimum targeting that Facebook offers. Think about it this way, you have the keys to a Ferrari 458 Italia but you are basically driving it at 25miles/hr.

If you want to really get the BEST performance out of Facebook, feed it as much data as possible. They have advanced machine learning algorithms that find patterns that get relayed to their ad serving algorithm. Bottom line Facebook algo’s are much smarter than your manual page interests targeting could ever be.

Don’t have data to use? Ask your affiliate network for some, broker a deal with an offer owner, collect e-mails yourself, partner with affiliates who have audience data.

1 Adset – 1 Variable at a Time:

As sophisticated as Facebook’s ad serving algorithm is at times it’s too damn smart for its own good. This is the case when you group too many ads into one ad set. If you put 10 ads into one ad set you will find Facebook only serves a few of them and than heavily weights one particular ad. For this reason, when you set up your campaigns it’s always a good idea to just target one interest at a time (so you can see what converted) and place one or at max 2 ads per ad set. This way you get best chance of getting accurate data back.

Duplicate Ads for Best Chance:

Facebook algorithm will serve your ad to 500 to 1000 people and quickly determine metrics like: User’s likeliness to click on the ad, the actual CTR, users likely to hide the ad, etc.

The problem arises in how Facebook randomly samples the demographic your targeting. It may pick 500 people that respond really well to your ad or it may pick 500 that really hate your ad from the same demographic.

This is why you will see the exact same ad have a huge CTR in one campaign and then complete flop in another. To prevent this, duplicate the exact same ad a couple times in your ad set. This means the same ad will be randomly served to different segments of users within the same demographic. The hope here is that one of your 3 identical ads hits a favourable audience and has a good CTR. Once you identify which one of your triplicates hit scale that bad boy up!


Scaling:

Facebook’s algorithm will absolutely punish you for cranking budgets to aggressively. Typically, we have found in our testing that you shouldn’t increase your budget more than 50% every three days. If you go more aggressive than this, you will likely lose your ROI.

One trick that has worked for us is taking an ad that you know is performing well and break it off into it’s own ad set with a high daily start budget. So you might start a proven ad at a $500/DB. This is not a smart strategy if you don’t have a good bank roll. But say your running BH campaign and need to get spend ramping up fast – this is the perfect medium risk solution.

We hope these tips help you crush it on your Facebook campaigns. If your doing big damage on Facebook WH OR BH (over $2k a day spend) we would love to connect with likeminded FB buyers. PM to chat!

Originally Posted by synapse

Here is the thing. Facebook hands down has some of the best quality traffic on the planet. It is also one of the few traffic sources that allow advertisers and affiliates to reach HUGE scale (think $100,000 a day spend). That being said, it’s by far one of the most sensitive traffic sources to work with as an affiliate.

Even if your campaigns are 100% compliant running e-commerce your ad account can still get banned, your ROI can go from a juicy 400% to -200% as soon as you increase budgets.

As annoying as it is, when you hit a profitable campaign on Facebook the feeling is unreal, you can literally put your foot down on the pedal and Facebook spends like there is no tomorrow.

Our team has spent the better part of a year and half mastering Facebook and here is what we have learned about slaying the Facebook beast.

Now I will start by saying this is what works for us in-house, doesn’t mean it is the right strategies for everybody. Every affiliate has a unique approach to running on FB.

Ok lets dive in!

Leverage Data:

Most affiliates run clicks to website campaigns using targeted interests only. This is usually because they are doing something naughty and can’t exactly integrate their Facebook pixel without accidently revealing their dirty page. While there is nothing wrong with this, you are putting yourself at a HUGE disadvantage. You are using the bare minimum targeting that Facebook offers. Think about it this way, you have the keys to a Ferrari 458 Italia but you are basically driving it at 25miles/hr.

If you want to really get the BEST performance out of Facebook, feed it as much data as possible. They have advanced machine learning algorithms that find patterns that get relayed to their ad serving algorithm. Bottom line Facebook algo’s are much smarter than your manual page interests targeting could ever be.

Don’t have data to use? Ask your affiliate network for some, broker a deal with an offer owner, collect e-mails yourself, partner with affiliates who have audience data.

1 Adset – 1 Variable at a Time:

As sophisticated as Facebook’s ad serving algorithm is at times it’s too damn smart for its own good. This is the case when you group too many ads into one ad set. If you put 10 ads into one ad set you will find Facebook only serves a few of them and than heavily weights one particular ad. For this reason, when you set up your campaigns it’s always a good idea to just target one interest at a time (so you can see what converted) and place one or at max 2 ads per ad set. This way you get best chance of getting accurate data back.

Duplicate Ads for Best Chance:

Facebook algorithm will serve your ad to 500 to 1000 people and quickly determine metrics like: User’s likeliness to click on the ad, the actual CTR, users likely to hide the ad, etc.

The problem arises in how Facebook randomly samples the demographic your targeting. It may pick 500 people that respond really well to your ad or it may pick 500 that really hate your ad from the same demographic.

This is why you will see the exact same ad have a huge CTR in one campaign and then complete flop in another. To prevent this, duplicate the exact same ad a couple times in your ad set. This means the same ad will be randomly served to different segments of users within the same demographic. The hope here is that one of your 3 identical ads hits a favourable audience and has a good CTR. Once you identify which one of your triplicates hit scale that bad boy up!


Scaling:

Facebook’s algorithm will absolutely punish you for cranking budgets to aggressively. Typically, we have found in our testing that you shouldn’t increase your budget more than 50% every three days. If you go more aggressive than this, you will likely lose your ROI.

One trick that has worked for us is taking an ad that you know is performing well and break it off into it’s own ad set with a high daily start budget. So you might start a proven ad at a $500/DB. This is not a smart strategy if you don’t have a good bank roll. But say your running BH campaign and need to get spend ramping up fast – this is the perfect medium risk solution.

We hope these tips help you crush it on your Facebook campaigns. If your doing big damage on Facebook WH OR BH (over $2k a day spend) we would love to connect with likeminded FB buyers. PM to chat!

Solid post Synapse! Nice tips in there!

All on point, good stuff Synapse!

Great post!

I have a question then about targeting only 1 interest per adset - do you go for very broad targeting or precise targeting?

Great post !!! So know the reason why my ROI decrease >50% after scale up my daily budget to >200% in 2 day .. And best way to keep ROI on my experience is "never pause your campaign" if you still got profit, because cpc will always stabile

@xkjonz

"I have a question then about targeting only 1 interest per adset - do you go for very broad targeting or precise targeting?"


Not sure exactly what you mean by this. But ill take a swing at answering it.

Let's say you are targeting Tiger Woods, Callaway, and Nike Golf. Regardless of how big the audience is for these interests I would still seperate them into different ad sets. As far as targeting - You can run somewhat broad at the start and then look at your demographic info in reporting to see which age groups / genders are performing the best and break them out into a more targeted ad sets.

Hopefully that answered your question

Hey Synapse,

Do you also have the problem that your ads are in competition with themselves? Always when I duplicate adsets on the same target interest, it will influence the original ad or just not get any reach..

Cheers

Good share!

If I may add about Scaling, this "duplicating adsets and using 3 of the same images in one adset" hack is something that's getting patched (if not already) very soon. It's kind of stupid and FB knows it.

Using the new conversion pixel and letting FB's algo optimise internally will be the way forward, sure you may lose ROI initially but as it learns more about what converts for you, your ROI consistency will be constant regardless of how much you ramp up your budget as long as the market size supports it.

Strong post synapse.. Thanks!
This Duplicate Ads for Best Chance is very interesting. I will put this to test right away and see how it improves my campaigns.

Have you test FB pixels and see any difference in running them or not?
What i am worried about , is that with it, they can track the end product served after the lp ( as they can see the changes on the lp pages etc)....

#CoughCloakCoughTheCoughPixelCough#

Originally Posted by synapse

Leverage Data:

Most affiliates run clicks to website campaigns using targeted interests only. This is usually because they are doing something naughty and can’t exactly integrate their Facebook pixel without accidently revealing their dirty page. While there is nothing wrong with this, you are putting yourself at a HUGE disadvantage. You are using the bare minimum targeting that Facebook offers. Think about it this way, you have the keys to a Ferrari 458 Italia but you are basically driving it at 25miles/hr.

If you want to really get the BEST performance out of Facebook, feed it as much data as possible. They have advanced machine learning algorithms that find patterns that get relayed to their ad serving algorithm. Bottom line Facebook algo’s are much smarter than your manual page interests targeting could ever be.

Great post!

Originally Posted by omrikos

#CoughCloakCoughTheCoughPixelCough#

WOW! Great post

Good tips, it is always desired and welcomed to learn any experience of affiliates regarding Facebook since running campaigns there is really a tricky business. Many thanks. Worth reading.

Awesome post! this forum is the best of the best!

Very good post. Love this forum.

Great post much appreciated Ill hit you up when Im at 2k per day!

Great post! but what about negative feedback?

that seems to be the biggest hitter right now, any thoughts guys?

"Duplicate Ads for Best Chance:

Facebook algorithm will serve your ad to 500 to 1000 people and quickly determine metrics like: User’s likeliness to click on the ad, the actual CTR, users likely to hide the ad, etc.

The problem arises in how Facebook randomly samples the demographic your targeting. It may pick 500 people that respond really well to your ad or it may pick 500 that really hate your ad from the same demographic.

This is why you will see the exact same ad have a huge CTR in one campaign and then complete flop in another. To prevent this, duplicate the exact same ad a couple times in your ad set. This means the same ad will be randomly served to different segments of users within the same demographic. The hope here is that one of your 3 identical ads hits a favourable audience and has a good CTR. Once you identify which one of your triplicates hit scale that bad boy up! "


Any update on this point? Is this strategy still valid?

Originally Posted by nirvana

"Duplicate Ads for Best Chance:

Facebook algorithm will serve your ad to 500 to 1000 people and quickly determine metrics like: User’s likeliness to click on the ad, the actual CTR, users likely to hide the ad, etc.

The problem arises in how Facebook randomly samples the demographic your targeting. It may pick 500 people that respond really well to your ad or it may pick 500 that really hate your ad from the same demographic.

This is why you will see the exact same ad have a huge CTR in one campaign and then complete flop in another. To prevent this, duplicate the exact same ad a couple times in your ad set. This means the same ad will be randomly served to different segments of users within the same demographic. The hope here is that one of your 3 identical ads hits a favourable audience and has a good CTR. Once you identify which one of your triplicates hit scale that bad boy up! "


Any update on this point? Is this strategy still valid?