Skip to content

How To Test Facebook Ads: A 5 Phase Strategy To Go From Failing To Profitable


Fear of failing is what stops most people from having a winning advert, because they want it to be profitable on day one.

And when it doesn’t work, they simply assume that ads ‘just don’t work for them’…

But the reality is that almost every ad starts at a loss. Unless you’re on a new platform with insanely cheap clicks, or an ad somehow goes viral, then yeah – they all start off losing money, and that’s ok.

The goal instead is to get initial data and then improve it to make it more profitable, either by throwing a huge budget at the AI, or by tweaking and testing for yourself first.

If your serious about creating profitable ads, then you need to embrace the failure, so you can find your winner. Otherwise you run the risk of ‘shiny object’ marketing: bouncing from one channel to the next, hoping for a silver bullet.

In this chapter I’m going to walk you through a step-by-step process, breaking down EXACTLY how to test your Facebook ads, so that by the end, you’ll have a profitable ad running to a cold audience.

Keep reading to learn how…

Bottom Up A/B Testing: How To Take Your Ad From Negative To Positive ROI

A few weeks back, we started running paid traffic to a new blog post

The issue?

It was losing money…

Almost every advert starts at a loss

Most people would have stopped here, but the thing is?

It wasn’t losing money…

…we were ‘buying data’

We needed to find what failed, to find what worked

With a/b testing you can find a winning ad, at a lower cost

This isn’t anything revolutionary

Its a method used in product design all the time…

Bottom up testing for facebook ad roi

You can know all the best practices but the reality is?

You can’t know what works, without testing

And so with each new test, the ad cost lowered, and the CTR improved

It went from losing money, to breaking even, to making a small profit

Until suddenly?

We were getting huge returns while promoting our content…

How to get a $22:1 roi with facebook ads

The best part?

You can run this same process for your own ads… even if you only have a small daily budget

The 5-Step Facebook Ad Testing Strategy

Just 5 simple steps:

Step #1-4: Run a sequence of test’s to improve the ads performance

Step #5: Broaden the audience and goal and unleash the AI

Let me break it down…

You’re going to run 4 A/B tests to your focus group audience

The 5 step- a/b testing process to a profitable facebook ad

Each test is designed to improve the click through rate of the advert-to make it so appealing that people can’t help but want to click on it

You test each element to find what fails, and what improves

Then you put those winning elements together, to create the most effective ad

Easy right?

By now the ad costs less to run, and it gets the most clicks


Then you simply scale it out to a broader audience, and let the AI do its thing

This lowers the cost even further, and it helps you scale up your audience…

How To Test Facebook Ads(Using Only A Simple ‘A/B’ Testing Framework )

Each test is ran in a specific order, based on highest impact:

  • Test #1: The Image
  • Test #2: The Headline
  • Test #3: The Subhead
  • Test #4: The CTA

We base this order, off of how someone consumes an advert…

How a reader consumes a facebook ad in their newsfeed


The image is the first thing that gets their attention- that’s why you improve this first

Then they check the headline, so you improve this 2nd

Then they read the sub head

Only then do they read the ad copy and take the CTA

By improving the most important elements first, you’ll see faster improvements in CTR

Make sense?

Once you find the winning variation, you have a ‘Control’…

How to use AI testing to improve your facebook ads

The control is the winning version…

Its the baseline result that you measure all future tests against

(i.e. Did it improve or lower performance etc)

With each test you improve the CTR and find a new winning element

This element then gets added to your new ‘Control’

(This way each new test always performs better than the last)

Now don’t worry

It might seem complex, but its actually super simple to do

When you break it down?

Each test takes about 5 minutes to set up

So lets walk you through it all…

Step #1: The Image Test

The image is the first thing that gets their attention

You’re going to test out different images, to find an image that gets the most clicks

If you’ve been following along, then you should have a few image variations already made from Chapter #5: Ad Design

How to design your facebook ad creative

I recommend you have at least 4 image variations for this test

Go ahead and open up your Facebook ads creator HERE

When it loads up, select the ‘Traffic’ option…

Set up a simple traffic campaign to get impressions on each image test

Go ahead and give the campaign a name, then click the blue continue button


Facebook has its own split test feature, but its expensive and doesn’t let you test everything that you want to

We have a work around for this, which I will share in just a second

Make sure you leave this split test option unpicked

Once you click continue, the ‘Ad Set’ will now load up…

Give it a name, and then select the ‘to website’ traffic option

Use your focus group audience to split test your images

Scroll down to the audience section

Click on the ‘Use A Saved Audience’ option

Choose the focus group audience that you made in the last chapter

Then scroll down to the ‘Placements’ options

Because you’re running News feed ads only, remove the other options…

Remove all targeting options and leave the newsfeed


You CAN run ads in those locations later on

The issue is that they have different requirements, such as image size or text available to use etc

This current advert is tailored for the News feed

By using it in the other locations, it would only show half an image or no text etc, and would simply waste your money

Make sure they are NOT ticked, then scroll down to the ‘Budget + Schedule’ section…

Set your facebook ads minimum budget of $5

Set the budget at just $5 per day, and have the schedule set to ‘Run my ad continuously’

Then select the ‘Optimize For Impressions’ button

This will then show your advert to your audience as often as possible

Why do this?

Because it can take 5-12 impressions before people pay attention to your adverts, you need them to see the ad as much as possible

By optimizing for impressions, they will see it more often, and you’ll see results quicker

Once you’ve done that, click the blue ‘Continue’ button

Now its time to set up the advert…

Only upload 1 image for now

Go ahead and give the image a name

Then select the ‘Single Image’ option

(This will allow you to use the large News Feed image you created earlier)

Then click the upload button


Only upload just 1 image for now

Why do this this?

Remember how I said we would show you how to get around the split testing, without using Facebook’s split test option?

This is how you’re going to do that

I’ll show you the other steps in just a second, but for now just make sure that you only upload the one image

When you’ve done that, scroll down the page to the ‘Links’ option

Go ahead and enter the URL of the article that you want to send them to…

Enter the URL of your article you want to send traffic to

Once you enter the URL, the ad will auto populate words from your article for the headline etc

You don’t want that

You want the ad copy that you wrote earlier

Go ahead and load up the advert ‘copy’ that you wrote back in Chapter #4: Ad Copy

How to write ad copy to a cold audience

Simply open up your ad copy document, then copy and paste it across

Easy right?

Be sure to copy the right text, into the right place…

Set up your ad copy for you test ad control

‘Headline’ is the headline

‘News Feed Description’ is the subhead

And ‘Text’ is where the rest of the content goes (Intro, body and CTA- the majority of the advert copy basically)

Once you copy it across, you’ll notice the ad preview will change

It should look similar to this

(With different ad copy ofc :P)…

What your long form advert should look like...

Easy so far right?

Once you’ve added your ad copy, scroll down a little further

Make sure that your tracking pixel is turned on, and then click the green ‘Continue’ button…

Make sure that your tracking pixel is all set up to record your traffic


You now have you ad running with just 1 image, to your focus audience

Now you’re probably thinking:

‘Aren’t we supposed to be testing different images to find a winner?’

You sure are, and you’ll add them in just a second

The only issue?

The tools that Facebook gives you are broken…

What Happens When You Upload Multiple Images

Facebook allows you to upload and run up to 6 images in an advert

The problem with running multiple images at once when testing

The only problem with this?

There’s a flaw in the algorithm…

Once one image starts to outperform others, it stops showing the other images as often

That’s what happened here:

How facebook favours stat sig far too soon

Facebook decided that the 1st image was better, and so it stopped showing the 2nd image

In fact?

It showed the 1st image almost 5x more!

Why care?

Because the ‘winner’ Facebook chooses, is not always the best performing advert

(It chooses far too early)

In my experience, you can have other variations that are slow to start- but then end up being the best performers

Make sense?

That’s why you don’t test multiple images in the same campaign…

The issue with multiple ads in a single ad set

You need an unbiased way of testing instead

So what can you do?…

Why Not Use The Split Test Tool?

Facebook does have a split test option

The issue with Facebooks split test tool...

You choose your variations, and it will split the audience equally to test

The thing is?

It usually costs more to test like this

(Facebook will usually charge more to show the test to people)

Not only that?

It only allows you to test Image or Audience variations…

The limitations of the facebook split test tool

Why care?

Because we know that tests for the headline, subhead and copy can make a HUGE difference to improving your CTR

(Its not all about the image you know!)

But this tool doesn’t allow you to test those elements

Fortunately there’s a work around

Its called ‘Duplication, Variation Testing’ or D.V.T

If you want to split test your ads easier and cheaper, then here’s what you need to do…

Duplication, Variation Testing:How To Get Accurate, Low Cost Test Results

So this is the best way I’ve found to test Facebook ads

Let me break it down…

All you do?

  • Make duplicate copies of the adset and advert that you just made
  • Change some minor targeting details so that the copies don’t ‘compete’ with each other
  • Change that one element that you want to test (In this case, the image)

Its as simple as that

Copy, edit, publish

Easy right?

How duplication, variation testing allows you to accurately split test your ads

This method allows you to get accurate tests of different elements of your ad- all at a lower cost

The best part?

It takes 2 minutes to set up

Lets walk you through it super quick…

Duplicate The Adset And Advert

Log back into your ads manager and find the ad that you just created

Tick the box next to the ad

When you tick it, the adset and image tabs will load up and allow you to edit them…

Duplicate the ad set

Go ahead and click on the tab that says ‘Ad Sets for 1 Campaign’

(If it says more than 1 campaign, then you must have multiple boxes ticked by accident)

Once you click on that tab, a new page will load up…

Duplicate and test image variations

Scroll your mouse over your advert to see some new options

Click on the ‘Duplicate’ text

A new pop up will load on the screen…

Create as many ad sets as variations that you want to test

Go ahead and select duplicate ‘Existing Campaign’

Then go ahead and enter how many copies you want to make


You need to make 1 new copy for every test variation

That means if you want to test 4 different images, then you need 3 more copies

(You already have 1 variation)

Does that make sense?

Go ahead and click the blue ‘Duplicate’ button

This will now make 3 more copies of both the adset AND the advert that you created

Now you need to edit each copy…

Organize the Adset

A new page will load up, showing the copies that you just made

First thing is to get organised

Organise the ad set tests

You’ll notice that all of the copies currently have the same name

Go ahead and click the ‘View + Edit’ text on the top right of the page

A new pop up will appear

Go ahead and rename each of the copies…

Give each unique ad set and image a name for testing...

Call them image test 2, image test 3 and so on

(Do this for each copy that you have)

Then click the blue ‘Save to Draft’ button

Now that you have each adset organized, you need to edit them slightly


If you have 4 adverts with the exact same audience + bid cap, then Facebook will make each variation compete with each other


You’ll be bidding against yourself, and spending more money

You can get around this by editing the audience slightly


That way Facebook see’s them as different, and so you won’t compete with yourself

All you need to do?

Simply change the age range for each copy, by 1 year

(This makes Facebook then see them as different audiences)

With each copy still ticked, scroll down to the location section on the right hand side…

Edit the age of each adset so you dont self compete

See the ‘Age’ option?

Go ahead and click the ‘View + Edit’ button next to it

This will load a new pop up

Change the age range of each copy by 1 year each…

Adjust the age range on the competing ad sets

Be sure to change the age range on EACH copy, so that none of them compete with the original ad

For Example:

  • Original 25-55
  • Copy 1 25-56
  • Copy 2 25-57 etc

When you’ve done that, click the ‘Save To Draft’ button

Now you just need to add the different images to test…

Add The Variation To Test (Different Image)

Scroll back up the screen

Click on the ‘Ads’ section, in the top right

Adjust the ad images

A new page will load up, with each copy of your ad

You need to edit them one at a time

Deselect each of the copies, by clicking the top tick box on the left…

Select the images one by one to edit them


A pop up may appear asking if you want to publish those changes

(If that happens, then just click ok)

Now you need to click just 1 copy at a time, and upload a different image…

Add the split test variable image

Scroll over the top image until you see the ‘Edit’ button

Give it a click

The ad manager editor will open on the right hand side…

Remove the image and upload the variation

Scroll down until you see the ‘Clear Images’ text

Click that and the image will disappear

Click on ‘Select Image’ then upload a new image variation

(Be sure to use a different images each time)

Now you just need repeat that for each copy:

  • Tick the ad you want to change
  • Remove the image
  • Upload a new one

Add the a/b test image variations

Once you’ve changed each image, click the ‘Save + Publish’ button

Easy right?

That’s your image test all set up

Side Note:

The first test takes the most effort

After this the tests are far easier and quicker to set up

All you need to do now?

Is make sure the ad is running!…

Check The Ads Are Working

A lot of people ask me how long it takes to test a Facebook ad

The reality?

They have to run for AT LEAST 3 days before you do anything


Because it takes around 3 days for Facebook to balance out an ads performance

(This means you can’t make any changes during that time)

But before you sit back and wait?

You need to check the ad is running!

(It would suck to wait 3 days and then come back and realize its not turned on right?)

It can take 15- 30 minutes for an ad to be approved

So simply wait for 15 minutes and then check back into your ads manager…

Check in that your split tests are all working

You will see a notification letting you know your ads have been approved

(Along with a green ‘Active’ notification next to each advert)

Now its up and running?

You wait…

4th Day Check In + Test

By now your ad has been running for 3 days

This gives Facebook enough time to balance out the ads performance

(Too soon and it might not be shown yet, or give random results)

Dont touch your ads during the learning phase!

Log back into your ads manager and find your advert

Click on the tick box next to it, and click on the ‘ads for 1 campaign tab’

This will load up the results from each image test…

Check in on your ads and look at the results

You’re interested in:

  • Impressions (How many times the ad has been seen) against
  • ‘Results’ (How many clicks its had)

This will then help us measure the CTR of each variation

It should look like this…

Facebook creative testing

(Notice how there are different results for each variation)

You need to measure those results

Go ahead and open up this A/B Testing calculator in a new tab

(Its totally free to use)

Put the impressions in the left column, and the results in the right…

test your ads until you have statistical significance

Do that for each image test

This will show you the current ‘winning’ ad

I say ‘current’ because you probably don’t have enough data yet, for an ‘accurate’ test

I won’t go into all the boring math, but you need to see a confidence rating of 95%

All you need to do?

Simply let it run longer…

This way you’ll know that:

  • The winner is accurate, and
  • The winner will continue to perform at that level as you scale up to a larger audience

Why care?

Because choosing too soon can mean choosing a weaker advert

(Which can lead to losing thousands of dollars as you scale up…)

Not great right?

If you don’t have a 95% confidence rating, just keep the ad running and test the results again in 24 hours…

How long to run your ad for? Keep running until you have a 95% confidence rating

That’s all there is to it

Now you know which image will get the most clicks from your audience

We call this winner the ‘Control Image’

Its the image that you’ll use for the rest of your tests

Even better?

Because your using that winning image for the next tests, it means they should see higher clicks right away

Pretty cool right?

So now its time to test out the other elements of your ad

Good news?

The rest are much quicker to set up…

Step #2: The Headline Test

Alright so these next tests are real quick

You simply

  • Copy the winning campaign,
  • Add the variations, and then
  • Test it for 3 days

Easy right?

Lets get into it…

Set Up The Winning Control

Time for a bit of organization

Go ahead and log into ads manager

Click the tick box next to your advert, and then go across to the ad set section

Organise the winning ad

The ad set tab will load

Find your winner, and rename it to CONTROL IMAGE

Rename the winner as the image control ad

Then go ahead and delete the other ad sets

(This will also remove the failing ads and images)

Simply tick the box next to the ones you want to delete, then click the trashcan

Delete the failed tests

Super Important:

Make sure your winning image isn’t highlighted also, otherwise you’ll delete it as well!

Now that you’ve tidied up the ad, its time to duplicate the control campaign…

Start New Test From Zero

You always start a new test from zero

This helps to make sure the ads perform fairly

Mouse over the winning ad, and then click on the duplicate text

Duplicate the control image, to start testing the headline

When you click that, a new pop up will appear

This time the duplication is a little different

You’re going to duplicate an entire NEW CAMPAIGN from this control…

Set up and split test the headline

Select ‘New Campaign’ and then call it ‘Headline Test’

Leave the bid options the same

Make just 1 copy, and then click duplicate

The ad will load up as before, with all the settings of the winning ad from the last test

(Same image, headline, sub head and copy)

Click the green ‘Publish’ button

Run your headline a/b split test

So now you have the new control ad up and running (from zero)

The last thing to do, is turn off the previous image test

Go back to your campaigns and find it…

Pause the previous control ad

Then simply click the blue button so it turns grey

So now you’re all organized, its time to set up the new variables…

Set Up The New Headline Test Variables

Same as before

You’re going to:

  • Dupe the control,
  • tweak the audience to not compete with yourself,
  • and then test different headlines

Click the box next to your Headline test, and then click on the ‘ad set’ tab

Duplicate the winning adsets for a new split test

Once the Ad Set loads up, mouse over the ad so you can see the duplicate option

Give it a click

Then select ‘existing campaign’ and choose the headline test you just created…

make as many adsets as tests that you want to run

Make 3 copies, and then click duplicate

The new copies will appear

Now you just need to tweak those audiences, so that you don’t self compete…

Tweak the Audience

Start off by editing each of the new copies names

Click the ‘View + Edit’ text to edit them all at once

Organise the new split test

Then edit their names …

Name each f your headline tests, so you can track the results

Adjust the audiences so they don’t compete with each other

Scroll down to the age section

Click on the ‘View + Edit’ text…

Edit the age range of each ad set so you dont over spend against yourself

This will load up each ad

Edit the age by 1 year for each

(This way you won’t compete with yourself)

Organise your headline ad set test

Easy right?

Now that those duplicates are all organized, its time to add your new variable to test…

Add The New Headlines

You should have a list of different headlines from the work you did in Chapter #4 ???(LINK

You should already have made multiple headlines to test

Go ahead and open up that document

Pick 3 headlines that you want to test

Then scroll to the top of the screen and click on the ‘Ads section’

Go to the ads section

Now this part isn’t complicated, its just frustrating


Facebook doesn’t allow you to simply edit the advert, as its a copy of a previous ad

That’s fine for the control…

…but you need to tweak the headlines right?

Keep the control as an exisiting post

(Notice all the text options are removed-gah!)

So what can you do?

Its a little long winded…

Make sure that you have each of the duplicates selected, and then click on the ‘Create Ad’ button

Copy and paste your ad copy and image for your new tests

This will then remove all the ad copy, and image’s from before…

Not great, BUT it does allow you to enter different headlines

(Sorry but its the only work around atm)

All we do?

Simply copy and paste again…

With each duplicate still selected, click on ‘select image’ and choose the winning image from the last test

This will upload that image to each copy

Keep the winning image control for all new split tests

Scroll down the page to the ‘text’ section

Still got your original Word document open with your ad?


Go ahead and paste the intro/body/cta section into the first box

(Just like you did in the first test)

Copy and paste the control elements into your advert

Then enter the URL for the article in the box below that

Scroll down to the ‘Headline’ box

This is a little different

Click the ‘View + Edit’ option below it…

Add the different headline test variations

A new pop up will appear

Go ahead and enter the new headlines that you want to test

(Just 1 variation per box)

Go ahead and write each headline variation


You don’t need to ad the original headline, as that’s being tested with the control ad

Once you’ve added the new headline variations, click the blue ‘Save to Draft’ option

Then scroll down to the ‘News Feed Link Description’ box

Paste the SubHead of your advert here…

Use the previous sub head

Then select the ‘Learn More’ button, and click the green publish option


That’s the 2nd test all set up

You’re now running the winning image with original headline, vs winning image with new headlines

Every test from now on, follows this same pattern

The only difference is which control you keep, and which new variable you test

Make sense?

Now you just need to measure the results…

Test And Measure

Just like before

  • Wait for 15 minutes and check the ads are running
  • Wait for 3 days and then test the results using the A/B Test tool
  • Keep running and testing each day, until you have 95% confidence rating

You now have a winning headline AND a winning image

At this point you should already see a lift in click through rate

Now its time to test that sub head…

Step #3: The Sub Head Test

Quick recap:

The audience is scrolling through their News Feed

The image gets their attention, and then they read the headline to see if its relevant to them

After that they check out the subhead

The subheads goal is to get them interested even more, so that they open up the ad copy and read it

Remember how an audience looks at your advert...

Now the test process is identical as before (Apart from the new variable)

Because of that, I’m going to simplify the steps below…

Create The New Control

  • Rename the winner of test #2 (Call it CONTROL IMAGE + HEADLINE)
  • Remove losers from test #2
  • Duplicate entire campaign to start at zero (Call it SUBHEAD Test)
  • Turn off the previous headline test…

Save the headline control for future tests

Add The New Variable (SubHead Testing)

  • Dupe the Ad Set x 3
  • Give each copy a name, and tweak audience age gaps
  • Go into ad section- make sure only duplicates are selected, and click ‘create ad’, so you can add the variable
  • Add the article URL
  • Add the winning image and headline from the previous tests
  • Copy the rest of the ad into the correct boxes
  • Change the SUBHEAD for each copy so you can test different variations
  • Publish the ad

Add each sub head variation into the advert builder

You should have written a few subheads already

Feel free to tweak the ones you have so that they work with your new control headline

(Remember the goal of the subhead is to tie into the headline, talks to the ‘after state’ goal and builds curiosity to read the intro)

Your sub head should always support the headline, while also building desire and interest

Once you have the variations up and running, you just need to measure them as usual…

Test And Measure

  • Wait for 15 minutes and check the ads are running
  • Wait for 3 days and then test the results using the A/B Test tool
  • Keep running and testing each day, until you have 95% confidence rating

Once you have the winning subhead, its time to test and improve the CTA section…

Step #4: The CTA Test

The final test

In theory, you could try out all new ad copy

But the biggest difference to your CTR?

It will be improving your call to action-the paragraph of text that gets them to go read the article

The process to set it up is still the same…

Create The Latest Control

  • Rename the winner of test #3 (CONTROL IMAGE + HEADLINE + SUBHEAD)
  • Remove losers from test #3
  • Duplicate entire campaign to start at zero (Call it CTA Test)
  • Turn off the previous Subhead test…

Save the sub head control winner

Add The Final Variable (CTA Variations)

  • Dupe the Ad Set x 3
  • Give each copy a name, and tweak audience age gaps
  • Go into ad section- make sure only duplicates are selected, and click ‘create ad’ so you can edit the text sections
  • Add the article URL
  • Add the winning image, headline and subhead from the previous tests
  • Change the CTA for each copy
  • Publish the ads

Set up your different call to action split tests

Just like with the subhead- you can click the ‘View + Edit’ button, to edit each version all at once

A Call To Action needs 3 elements:

  • Ask them to do something (Go read the article)
  • Tell them what will happen when they take the action
  • Connect to an emotional driver…

If you remember back to the ad writing chapter, we talked about the emotional reasons that people do anything

How to improve your call to action

They are:

  • Self Identity (to see ourselves in the offer)
  • Need for Certainty and to remove anxiety
  • Need for Uncertainty and to create excitement
  • Need to feel Significance from others
  • Need for Connection from others
  • Need for Growth and struggle
  • Need to Contribution and give

Every member of your audience wants what you have to offer

The only difference is the REASON behind WHY they want it

At the moment, your call to action appeals to one of those emotional factors- usually fear of loss/to remove anxiety

In this test, you’re going to tweak your call to action to appeal to a different emotional driver

Now its real easy:

You already know your audience, because you did your research

Think about your article, and what they could gain from reading it- but from different emotional appeals

Then go ahead and write 3 new CTA variations

By doing this, your different CTA’s will appeal to different segments of your audience

One of them will be more effective than the other- improving the CTR across to your article

Simple right?

Once you have them up and running, you just need to measure them as usual…

Final Test And Measure

  • Wait for 15 minutes and check the ads are running
  • Wait for 3 days and then test the results using the A/B Test tool
  • Keep running and testing each day, until you have 95% confidence rating

By this point your ad is the highest performing CTR of any variations

Great work!

This will make a huge difference to your ad cost, and its performance…

Step #5: Broaden And Launch

The last step

Remember that your end goal is to drive paid traffic to your article, at a profit

Now you know that your ad converts, its time to scale it out to a wider audience

During the test phase you focused on 2 things

  • Getting the audience to see the ad as often as possible (impressions)
  • Measuring their response (i.e clicks)

With all these tweaks, you might even have a profitable ad by now

Lets take it a little further…

You’re going to change your audience, and goal

Why care?

By broadening the audience, Facebook lowers the cost to show the advert

(Less specific targeting = less costly)

So why change the ad goal?


By ‘optimizing’ for a specific goal, Facebook starts to learn about the audience that converts on your article

It then uses that information to improve targeting, lower costs and get better results

Most people stick to their tight focus group, and so struggle to get a profitable ad

This is the secret you’ve been missing

The best part?

It takes around 3 minutes to set up…

Find The Final Winning Control

Go ahead and open up your ads manager

Find your winning control, rename it and delete the failures…

Save the winning CTR control

Now its time to create a new conversion campaign…

Create New Campaign

You can’t duplicate and change the campaign goal from the ad section

So go ahead and click on the ‘Campaign’ button in the top tab…

Duplicate the winning campaign and change the ad goal

This will load up your ad at the campaign level

Scroll your mouse over the winner and then click the ‘Duplicate’ button

Duplicate at the campaign level...

A pop up will appear

Select ‘New Campaign’ and give it a name

Then the ad editor will load on the right hand side of the page

Set up a conversion advert

Click the ‘Traffic’ button to load a drop down menu

Select ‘Conversions’

Then click the ‘Ad Set’ button on the top right

This loads up the ad set manager…

Use the custom conversion that you created earlier

Now you need to select your conversion goal that you created in Chapter #6: Ad Tracking

This is the goal you want Facebook to record results from

(It’ll then use it to measure ad performance, cost and learn about your audience)

Simply click on the box, and then select the correct conversion goal from the drop down menu

Then scroll down to the budget section

Set your daily budget at around 4 x your bid cap

I recommend starting your daily budget at around $20

(Too much too soon can skew the algorithm, so start low at first)

Now that the budget is set up, you just need to change the audience targeting…

Scroll down further to the audience section

You should see the focus group audience still targeted

Remove it, and then select the broad audience you made earlier…

Use your broader, lookalike (LAL) audience now

Easy right?

Before you hit publish, its always a good idea to check on the advert

Scroll up to the ad section and give it a click

Give your advert a quick look to make sure there are no errors

Make sure that your pixel tracking is turned on!

If its all ok?

Just hit the green publish button!…

Test And Track

Just like before

  • Give it 15 minutes and then check the ad is running
  • Then come back after 4 days and check on the results

(Any results before might not be accurate, so don’t touch the ad before then)

Now this time you’re not split testing or measuring results


You’re looking to see if the lead cost is within range…

You should see your conversion goal on your dashboard

Measure your ad ROI and ROAS

As long as the cost per optin is BELOW the max cost that you found in Chapter 2, then great news…

Your ad is now profitable!

You can happily keep promoting your article to cold audiences, knowing that you will see an ROI

Not bad right?…

What To Do Now:

So there you have it

Not just an entire step by step process of how to test Facebook ads, for your business

Even better?

You also have a system to run a profitable advert to a cold audience, and promote your content

Now more ‘ghost town’ content for you buddy!

With this method you can scale and compete with MUCH larger companies, without writing all the damn time!

Better still?

You can run this as low as $5 per day and KNOW that you will make money

(As well as grow your audience and leads…)

All that’s left to do now?

Go set it up for your own content!


You can download a PDF version of this article to follow along and take action

Simply click the image below, then enter your email to have it sent to you

Even better?

We’ll also send you 2 follow up articles via email, when they go live…

  • How to troubleshoot your ads performance (and further lower cost)
  • How to scale up to larger audiences (and stay profitable)

Simply click below to get this bonus AND those next articles!