Facebook Ad Cliff Notes - What Are They

[Vagabond 007]

What are some cliff notes on running ads on FB?

For example (not saying any of these are correct)...

*Use CPC
*Only use news feed ads on desktop
*Only one interest per ad


I know the ultimate answer is to test. But, to hit the ground running and start somewhere, what are some key pointers to keep in mind?

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[theag]

Test, test, test.

Other than that:

Image is most important. Stand out without using cheap red borders.
Start with website clicks cpc bidding, then test OCPM optimized for conversions.
Layer lookalike audiences based on conversions with precise interests.

So many things to consider..

Check this out, its tonight:

Facebook Marketing masterclass with Keith Krance
https://www.facebook.com/events/356561567844302/

 

[Silverhawk851]

Yep Test.

Test min 25 diff ads per campaign, laziness kills on FB.
Ideal targetting is 300,000 people to 600,000 people, big enough niche to make some solid $, small enough to stay lucrative.

Split each variable into it's own campaign, so you can track every little part.

ALWAYS place a pixel. Tracking is your Secret sauce.
Use your own tracker to track also.

And the golden key.

Your goal is NOT to make money off the bat. Your goal is to figure out what your COST PER ACTION is. It could be $100 in the start.
That's ALRIGHT. Most people know their Click through Rates and their Conversion rates, but do not know this number.

Just run some educated guesses on what may work, and figure the number out.
Then you chip away, by testing and testing, and testing, to lower that.

Keep testing until it is well below your LIFETIME EARNINGS PER CUSTOMER, which is how much profit you will make on each customer over the lifetime.

Formula: LIFETIME EARNINGS PER CUSTOMER - COST PER ACTION = PROFIT

Once your CPA is significantly lower than your LEPC, you can scale HARD!

If you need any help, feel free to PM bro

 

[Vagabond 007]

Test, test, test.

Check this out, its tonight:

Facebook Marketing masterclass with Keith Krance
https://www.facebook.com/events/356561567844302/

Thanks for the info. I signed up.

Split each variable into it's own campaign, so you can track every little part.

So make a new campaign for EVERY ad? So I'll have 25 different campaigns, each with one ad, each testing something different than the other 24?

What's the benefit to doing it that way rather than 25 ads in the same campaign?

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Silverhawk851]

So make a new campaign for EVERY ad? So I'll have 25 different campaigns, each with one ad, each testing something different than the other 24?

You don't need to just do 1 ad per campaign, but campaign per variable.

What my point was to be organized.
There are alot of variables, Age, interest, gender, platform, etc

You want to hold them all constant for each campaign, and change one variable.
E.g. same age, same gender, but maybe different interest for one campaign.
Then figure out which interest is working the best.

Then take the winning interest and gender constant, and try different age groups.

The mechanics can be different, but the main point is to be able to hold different variables constant, and test one at a time, instead of trying a new ad with a random mix interests, random age groups, gender, etc which will never let you pinpoint what is the main winning formula.

That's how you are able to track exactly what works, and create the big winners

 

[JasonR]

Honestly, it's really easy to "track" your progress. Are you making money?

So many people overthink this, but I don't care what my CPC, CTR is etc. if it's profitable. That's literally all I look at. I may look at other metrics to test and track things, but so many get overwhelmed by the small things.

You just need to dive in.

 

[theag]

What Jason said.

I dont even track on traffic source or even campaign level right now.

I just look at total ad cost for all traffic networks and total revenue per day. Making so much money so profitably that I dont have time to optimize ads but have to take care of other things to scale, haha.

Luxury problem I guess

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Gsuz]

In addition to what has already been said:

Give the campaign/the ads some time to spend money. Don't go into panic mode after 10 bucks spent and try to "optimize" by changing the color of your CTA. Acquire some data first and then tweak your approach accordingly. You already know that the traffic source is working, all you need to do is make it work for you.

 

[Vagabond 007]

Thanks guys.

I'm not worried about spending money to test things. I expect to lose money at first. I'm ok with that. I look at it as buying data.

For some reason I am having trouble getting my head around the proper way to set everything up. Kind of what Silverhawk was talking about. I get the mechanics behind ads and testing them...testing different variables and looking for the winner(s).

But what's the reasoning behind using multiple campaigns? Why not just use multiple ads in the same campaign? There's obviously a reason, but I'm missing it. Is it because it allows you to see the data easier? Campaigns share the same demographics therefore if you want to test different ones, you need different campaigns? Something else?

The way I understand it now is there are 3 levels...

Campaigns -Based on your objective
Ad Sets - All share the same targeting, budget, schedule
Ads - Self explanatory

Seems like I just need to create different ad sets, not necessarily different campaigns??


=======================================================================

Btw, my one ad was running for almost 24 hours and only reached 12 people. I assume it could be because of low budget, not enough time, something else?

$10 a day budget
$2 max per 1,000 impressions

UPDATE: Redid some things and got 73 impressions within minutes.



THANKS AGAIN. I appreciate the help!

 

[theag]

Did you attend the masterclass webinar? Heres the replay:

http://www.tribelearn.com/fbmarketing/online-training/replay

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[theag]

Seems like I just need to create different ad sets, not necessarily different campaigns??

Yes, you can.

What I do is this:

On campaign level I seperate by objective AND targeting. So for each objective/targeting combination I have one campaign. Then multiple ad groups with one ad per ad group where I test the ad copy. When I find a working ad I pause all the other ad groups and multiply the working ad group to scale. Then rinse and repeat.

I always set the ad group budget to 500-1000€/day each. I dont really care about budget, just CPA. So I control the budget with the CPC/CPA bids.

 

[theag]

Facebook did an update again to campaign structure. Targeting is now located at the ad group level.

https://www.facebook.com/business/news/facebook-campaign-structure-updates

 

[Silverhawk851]

Seems like I just need to create different ad sets, not necessarily different campaigns??

Yeah Ad Sets. Mind you, I run display predominantly, so yeah, I'd call that a campaign. Lol
Set up can be whatever you like, goal is to seperate variables to empirically understand what works

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Vagabond 007]

Guys, again...Thank You. I really appreciate you taking the time to help me out. Especially since they were probably dumb questions.

Did some more research last night, read some things and watched some videos from a FB ad course. Things are starting to click now.

I also just joined a monthly FB mastermind group. $69 a month and it looks very promising. Here is a link (not affiliate) to info about it if you're interested...

http://www.tribelearn.com/fbmarketing/special/index-1

 

[theag]

I also just joined a monthly FB mastermind group. $69 a month and it looks very promising. Here is a link (not affiliate) to info about it if you're interested...

Are you still a member?

Just canceled my membership there after 3 payments and not watching a single video lol. Too busy. Checked out some stuff and it is very beginner level in my opinion, but I'm already pretty advanced in PPC so maybe thats just me.. The FB group is also pretty inactive, only people posting are mindvalley support staff.

 

[Vagabond 007]

Are you still a member?

Just canceled my membership there after 3 payments and not watching a single video lol. Too busy. Checked out some stuff and it is very beginner level in my opinion, but I'm already pretty advanced in PPC so maybe thats just me.. The FB group is also pretty inactive, only people posting are mindvalley support staff.

I recently canceled.

But, mainly for my own reasons, not because of anything they did, or didn't do. Currently, I'm not doing any FB advertising so I don't see the need to be a part of it. When I ramp things back up I may join again.

I felt if I had any questions they would get answered. Which is obviously good. But yes, like you said, the only people posting were the staff. Wish it had more interaction.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[sitemaster]

Facebook Relevance Score
(rolls out globally starting this week)

The facebook relevance score is calculated based on the positive and negative feedback that facebook expects an ad to receive from its target audience.

It helps advertisers in a few key ways:

1. It can lower the cost of reaching people.
2. It can help advertisers test ad creative options before running a campaign.
3. It can help optimize campaigns already in progress.

Use relevance scores as a way to reach your audiences at lower cost.
To check your ads' scores, head to Ads Manager and add the relevance score tab to your ads reports.

For more details see:
https://www.facebook.com/business/news/relevance-score

 

[Y.B.]

I heard a lot of different things as far as optimal targeting. For example some say that going after a few million people is what you should do.

The only way to truly know is by testing each item but that would mean you could have hundreds if not thousands of ads to split test which isn't idea.

For example you want to test:
Audience, Demographic (age, sex), Image, Title, Description, etc.. There are so many different combinations. That's the biggest struggle.

Right now I'm running a campaign with only 1 ad set to Female, 25-55, US only, several audiences in the millions and it's doing pretty well. Sure it could probably do better if I split test 1000 different metrics to see where it can be improved.

 

[eliquid]

Im going to be presenting almost 100% of this at the FLF in March. Hopefully, those of you going wont get bored watching it here, then hearing the same at the meetup in 3 weeks.

http://affplaybook.com/blog/podcast/apb-podcast-jason-brown-from-serpwoo-2/

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Alan at SE]

What are your thoughts on scraping and targeting by UID custom audience?

It's a little grey hat but I've done this personally and have had great success in laser targeting my offer/ad/demo. It's becoming a bit intensive to scrape and constantly refill audiences so I've been relying on it less.

Any kind of remarketing has done very well for me. I did a split test between site remarketing, FB page fans, and email list subscribers, which are all in theory similar in nature and by far the remarketing converted the best.

 

[theag]

What are your thoughts on scraping and targeting by UID custom audience?

I did that in the beginning of my business and FB efforts. Worked well for me too.

Not doing it anymore because I want to run my ads as clean as possible to have a good relationship with FB, get a rep in the future, get monthly billing credit terms, etc...

Would be pretty nice to scrape a few group though. ROI must be amazing when I would apply everything I learned to those laser targeted audiences now.

I think it has become even harder to use scraped uids since they make you specify the app where you got those ids from during the import. Pretty sure theyll do some validity checks on that..

 

[Vagabond 007]

I would NOT scrape ID's anymore. It's against their rules and they may ban you. They seem to be cracking down on it.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[eliquid]

Scraping UID is harder and will become harder in the future.

It's an almost dead tactic that won't get any better in the future.

 

[wilddog]

Are you guys split testing landers as well?
Seems to me like optimizing your LP is as important, if not more important, than you creative?
The creative grabs attention, but the LP drives your sell.

 

[Y.B.]

Are you guys split testing landers as well?
Seems to me like optimizing your LP is as important, if not more important, than you creative?
The creative grabs attention, but the LP drives your sell

Yes they are all important. Your ad has to drive visitors to your lander, and your lander has to get the users to take the desired action. You have to test the ad to see what the response is on the landing page. If your ad is deceptive for example, it will get the clicks to the landing page, but won't get you the conversions. It's all part of 1 big process.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[theag]

I dont split test my LP currently. Reason is that the first rule of conversion optimization is, that you have to keep your traffic steady to be able to evaluate tests on your LP. Since I'm testing a lot of stuff with my traffic right now, I just have 1 single LP.

Its definitely important though.

 

[blackz_11]

Been looking into running Facebook adverts for affiliate marketing now. Is anyone successful using Clickbank etc and linking directly to the product or are you guys using your own landing pages?

Also is anyone targeting any emerging countries EG India, Brazil where there CPC can by very low?

 

[Y.B.]

Been looking into running Facebook adverts for affiliate marketing now. Is anyone successful using Clickbank etc and linking directly to the product or are you guys using your own landing pages?

I'm about to get started with this. I recommend using your own landing page to get emails first...because otherwise if they don't convert into a sale for you, you have nothing to show for the $ you spent on ads. Whereas if you could at least get some emails, you may be able to break even or profit with an email sequence/offers.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Journey2Million$]

What Jason said.

I dont even track on traffic source or even campaign level right now.

I just look at total ad cost for all traffic networks and total revenue per day. Making so much money so profitably that I dont have time to optimize ads but have to take care of other things to scale, haha.

Luxury problem I guess

What do you sell?

 

[theag]

What do you sell?

Underwater basket-weaving starter-kits.