How To A/B Test Your Meta Ads Creatives (+ Free Cheat Sheet)

Summary

In this informative video, SED from Vertex Marketing Agency shares a step-by-step guide on AB testing Facebook ads in Ads Manager. He discusses the agency's process for testing creatives, common questions like test duration and handling winners and losers, and emphasizes the importance of generating and refining ad ideas. The video provides insights on ad account setup, the benefits of using broad audiences, and a detailed manual AB testing method, including campaign structure, budget optimization, and monitoring strategies to identify and scale winning ads.

Takeaways

• 😀 AB testing is crucial for generating new ideas and optimizing Facebook ad performance.
• 📊 The speaker prefers manual AB testing over Facebook's built-in feature due to flexibility in managing test duration and momentum.
• 🎯 For effective testing, use a broad audience targeting to ensure a wide reach and easy scalability.
• 📝 Maintain consistency with the main campaign's objective when setting up a testing campaign to facilitate easy duplication and comparison.
• 🔢 Utilize ad set budget optimization to control spending on individual tests rather than campaign budget optimization.
• 📈 Start with a minimum of two and a maximum of six ad variations per ad set to ensure sufficient testing without overwhelming the analysis.
• 🔑 Test variations should focus on a single variable at a time, such as different hooks or headlines, to identify clear winners.
• 🚫 Avoid mixing different types of creatives in the same ad set to prevent skewed results.
• 🕒 Monitor new ads for at least 5 days to gather enough data for an informed decision on ad performance.
• 🏆 Identify and duplicate the best-performing ads from the testing campaign into the main campaign to capitalize on their success.
• 🔄 Continuously create new ad sets with fresh creatives to keep the testing process dynamic and iterative.

Q & A

What is the primary focus of the video?

The primary focus of the video is to demonstrate how to AB test Facebook ads in Ads Manager, covering the process used by the agency to test creatives and answer common questions about the testing process.

What does the speaker suggest is their favorite part of managing Facebook ads for clients?

The speaker suggests that their favorite part of managing Facebook ads for clients is the AB testing aspect, as it involves generating new ideas and testing them to see if they are effective.

What is the significance of using a broad audience in the AB testing process?

Using a broad audience in the AB testing process is significant because it allows for targeting a specific geographical area without narrowing down the audience too much, making it easier to scale and test different creatives effectively.

Why does the speaker recommend against using the built-in AB testing feature in Facebook Ads Manager?

The speaker recommends against using the built-in AB testing feature because it requires setting a start and end date, which can limit the ability to continue running successful tests and can disrupt the momentum of performing ads when moved to a main campaign.

What is the recommended number of ad variations per ad set during AB testing?

The recommended number of ad variations per ad set during AB testing is between two and six, to ensure enough testing without overwhelming the analysis with too many variables.

Why is it important to keep the engagement from a performing ad when moving it to the main campaign?

Keeping the engagement from a performing ad is important because it retains social proof in the form of likes, comments, and shares, which can boost the ad's performance and provide valuable insights from the testing phase.

What is the recommended approach to scaling a successful ad set?

The recommended approach to scaling a successful ad set is to gradually increase the budget every two days by about 20% as long as the ad continues to perform well, based on the cost per result or other relevant metrics.

What should be done with the ads that are not performing well in the testing ad set?

Ads that are not performing well in the testing ad set should be paused after a sufficient testing period and enough data has been gathered to make an informed decision.

How long should an ad be monitored before making a decision about its performance?

An ad should be monitored for at least 5 days, depending on the sales cycle, to gather enough data to make an informed decision about its performance.

What is the main reason for avoiding mixing different types of creatives in the same ad set during AB testing?

Mixing different types of creatives in the same ad set can lead to unequal comparisons and make it difficult to accurately determine which elements of the creative are contributing to its performance.

What is the purpose of the provided cheat sheet mentioned in the video?

The purpose of the cheat sheet is to provide a quick reference guide for AB testing creatives, ensuring that viewers follow all the necessary steps and can easily reference the process during their own testing.

Outlines

00:00

😀 Introduction to AB Testing on Facebook Ads

The speaker, SED from Vertex Marketing, introduces the topic of AB testing for Facebook ads within the Ads Manager. The video aims to demonstrate the agency's process for testing creatives and answering common questions such as test duration, handling winning and losing ads, and the importance of generating and testing new ideas. The speaker emphasizes the significance of AB testing in ad management and shares a personal preference for testing, suggesting it's a key part of their job that involves creating, testing, and scaling effective ad strategies.

05:03

📈 Setting Up the Ad Account for AB Testing

The video script explains the typical ad account setup used by the agency, which includes a main campaign for the majority of the budget and broad ad sets for targeting specific countries or cities without additional audience restrictions. The speaker discusses the evolution of targeting strategies, moving from lookalike audiences to broader ones due to advancements in Facebook's ability to understand ad content and customer demographics. The script also covers the benefits of using a broad audience for easier scaling and testing, and provides a walkthrough of the ad account structure, including the creation of testing campaigns and the manual approach to AB testing as opposed to using Facebook's built-in AB testing feature.

10:04

🛠 Detailed Steps for Creating a Testing Campaign

The script outlines the process of setting up a new campaign for AB testing, emphasizing the importance of aligning the campaign objective with the main campaign to facilitate the transfer of successful ads. It discusses ad set budget optimization versus campaign budget optimization, advocating for the former to maintain control over budget allocation among different ad sets. The speaker also explains the selection of a broad audience for testing to ensure a large enough pool for accurate results and to avoid overlap with retargeting audiences in the main campaign. The paragraph concludes with advice on naming the ad set and considerations for the number of variations to include in each ad set.

15:06

🎯 Crafting Ad Variations and Launching the Test

This section of the script focuses on creating multiple ad variations within an ad set, recommending a range of two to six ads to test different hooks, headlines, or video lengths while avoiding mixing different types of creatives. The speaker advises against combining various ad formats in the same test to maintain a fair comparison. The script then moves on to the launch and monitoring phase, suggesting a 5-day monitoring period post-launch to gather sufficient data before making any decisions. It also touches on the importance of understanding the sales cycle and adjusting the monitoring period accordingly.

20:07

🏆 Analyzing Results and Scaling Winning Ads

The final paragraph of the script details the process of analyzing the results of the AB test, identifying the top-performing ads, and deciding whether to incorporate them into the main campaign. It discusses the criteria for determining a winning ad, which includes comparing its performance to the worst-performing ads in the main campaign. The speaker shares strategies for scaling successful ads by gradually increasing the budget and monitoring performance over time. The script concludes with a reminder to continually monitor winning ads and to repeat the AB testing process with new creatives to maintain an effective ad strategy.

🎉 Conclusion and Additional Resources

The script concludes with a brief mention of additional resources, specifically a cheat sheet available in the video description, which viewers are encouraged to download for reference during their AB testing process. The speaker reiterates the importance of the steps covered in the video and bids farewell to the audience.

Keywords

💡AB Testing

AB testing, also known as split testing, is a method of comparing two versions of web pages to determine which performs better. In the context of this video, AB testing refers to comparing different Facebook ad creatives and strategies to identify which generates a better response from the audience. The script mentions AB testing as a crucial part of managing Facebook ads, with the presenter showing the process they use in their agency to test creatives and improve ad performance.

💡Facebook Ads Manager

Facebook Ads Manager is a platform where advertisers can create, manage, and track their ad campaigns. The script discusses using the Ads Manager to conduct AB tests, emphasizing its importance in the process of optimizing ad performance and making data-driven decisions based on the results of these tests.

💡Creatives

In advertising, 'creatives' refers to the visual and textual content of an advertisement, such as images, videos, and copy. The video script focuses on testing different creatives in Facebook ads to determine which resonate more effectively with the audience. The presenter explains how to test variations of creatives like images, videos, and copy to improve ad outcomes.

💡Campaign Objective

A campaign objective defines the goal of an advertising campaign, such as increasing sales or generating leads. The script emphasizes the importance of setting the same campaign objective for both the main campaign and the testing campaign to ensure that the AB tests are relevant and can be accurately compared.

💡Ad Set

An ad set in Facebook Ads Manager is a group of ads that share the same targeting, budget, and scheduling. The video script describes structuring ad sets for testing purposes, including setting budgets and choosing the right audience to ensure that the AB tests are conducted effectively.

💡Broad Audience

A broad audience in Facebook advertising refers to targeting a large group of people without narrowing down the audience with specific interests or behaviors. The script mentions using a broad audience for testing to ensure that the ads are reaching a wide range of potential customers and to make the test results more generalizable.

💡Budget Optimization

Budget optimization in Facebook ads is the process of allocating the campaign budget to ad sets that perform better. The script discusses ad set budget optimization as a strategy for managing the testing campaign, allowing the advertiser to control where the budget is spent and to identify high-performing ads more effectively.

💡Lookalike Audience

A lookalike audience is a group of people who are similar to a source group, such as existing customers. The script mentions that lookalike audiences were commonly used in the past but have become less effective compared to broad audiences, which can now be more efficiently targeted using Facebook's advanced technologies.

💡Testing Campaign

A testing campaign is a specific type of ad campaign designed to experiment with different variables to determine what works best. The script outlines the steps to create a testing campaign in Facebook Ads Manager, including setting up a new campaign, choosing the right objective, and structuring ad sets for effective testing.

💡Scaling

In the context of advertising, scaling refers to increasing the budget or reach of a successful ad to maximize its performance. The script advises viewers on how to scale a winning ad set by gradually increasing the budget when an ad is performing well, thus capitalizing on its success.

💡Winner

A 'winner' in AB testing is the version of the ad that performs better than the other versions. The script discusses identifying the winning ad set or ad within a testing campaign by monitoring performance metrics such as cost per result and then integrating the winning creative into the main campaign to improve overall ad effectiveness.

Highlights

Introduction to AB testing Facebook ads using the same process as Vertex Marketing Agency.

Importance of AB testing in generating and refining creative ideas for ads.

The presenter's preference for testing and its role in managing Facebook ads for clients.

Explanation of using broad audience targeting in Facebook ads for better scalability and testing.

Advantages of broad audience targeting over lookalike audiences in current Facebook ad strategies.

Details on setting up a typical ad account with a main campaign and multiple ad sets.

Manual AB testing preference over Facebook's built-in AB testing feature due to flexibility.

Instructions on creating a new campaign for AB testing with the same objective as the main campaign.

Importance of keeping the campaign objective consistent when duplicating successful ads.

The role of adset budget optimization in controlling spend and testing different ads effectively.

Guidelines on structuring the testing campaign with appropriate naming and broad audience selection.

Recommendation on the number of variations per ad set, ranging from two to six for effective testing.

Strategies for creating different creative variations to test within the same ad set.

Advice against mixing different types of creatives in the same ad set for accurate testing.

Process of launching and monitoring ads for a minimum of 5 days to gather sufficient data.

Method for identifying the best performing ads and deciding whether to scale, duplicate, or pause them.

Emphasis on continuous monitoring and optimization of winning ads within the main campaign.

Final thoughts on the rinse and repeat process of AB testing to continually refine ad performance.

Transcripts

hey everyone SED from vertex marketing

agency and in this video I'm going to be

showing you how to AB test your Facebook

ads in ads manager and I'm going to show

you the exact same process that we take

in our agency to test creatives and

answer some of the common questions like

how long should I leave a test running

what do I do with my winners what do I

do with my losers and a lot more and I

thought actually this video would be the

perfect time to wear my this t-shirt so

I'd rather be a testing cuz it's true to

be honest that's probably my favorite

part about my job which is you know

managing uh Facebook ads for clients is

the AB testing set of things and I think

that's also the most important part

because it's really all about generating

new ideas like with creatives and copy

and implementing on those ideas right so

getting those let's say those ugc videos

those images and just going in ads

manager and testing them and hey was it

a good idea or was it a bad idea if it's

a bad idea you learn from it so that you

Tred to put those ideas on the side and

just think about the good ones in the

future but then if it's a good idea

perfect you scale that and then you

think huh like how how could I replicate

this or how could I create something new

that is similar to this and it's like a

loop you just want to keep repeating

that process but it all starts by

knowing how to AB test and how to really

structure everything and that's exactly

what I'm going to be showing you in this

video okay so I am now Inside My Demo ad

account and I first want to talk about

the typical ad account account setup

that we usually will Implement for

client so we will always have like a

main campaign and that main campaign

that well that's the one receiving the

majority of the bud buget um and inside

that main campaign usually we'll have

like maybe one two to three different

ads sets but like 99% of the time we

will have a broad adset and what does

broad mean broad just means that we're

just selecting the specific let's say

country or maybe states that we want to

Target or if it's local advertising

we'll select the city um and we actually

don't include anything else in the

audience but instead we do the targeting

with the ad so now with the advancement

in technology Facebook can actually just

understand a your copy but B your image

and your video and the lending page and

really just understand what you're all

about what you do and who your customers

are and because you have tracking set up

and you're sending let's say a lead

event or a purchase event back to

Facebook it knows exactly who your

customers are so by using this

information they can just go ahead and

kind of like create your own lookalike

audience which that's what we really

used to do like back in would say 2019

is we would create you know lookalike

audiences but now those tend to not work

as well as something as like a broad

audience so that's number one um and

then another reason why we like the

broad audience is because once you can

make it work with the broad audience is

really easy to scale because you don't

have a limit with the frequency I mean

if you do it's more of like an issue

with Facebook meaning like you're

spending just way too much and Facebook

just doesn't have a big enough audience

for you which at that point you know

like that's a whole other issue so

that's why we really like the broad

audience and and it makes it also really

easy to test but I'll get into that in

just a second I just want to show you

kind of like what the typical add

account structure will look like I'm not

saying that everyone just only gets one

campaign one adset but usually the

majority of the budget is going to go

towards that main campaign um but it

doesn't mean that we don't have like

another campaign like doing like a

catalog sales or other specific type of

campaign maybe targeting specifically

like different countries and we just

want to have those separate again what

I'm saying is most of the time there's

going to be one campaign in your at

account that is receiving the majority

of the budget um and that's what we call

like a main campaign but with that being

said now let me show you how you would

go ahead and structure that uh testing

campaign so now creating your testing

campaign there's a few different ways of

doing it I actually like to create just

a brand new campaign and doing it a

little bit more manually uh but if you

talk to like a meta specialist they

would probably say or refer you to this

feature right here which is AB testing

but I'm personally not a big fan of that

and I've talked to a lot of other

advertisers and you usually have the

same opinion and the reason I'm not a

big fan of the ab test feature here is

because once you create it you have to

set a start and an end date and what I'm

going to show you is if ever your ad set

or really your test is performing well

like well don't pause it and then a meta

specialist will usually say well yeah

just pause it and bring that inside your

main campaign and you know the ad is

going to be running there just with

experience and also if you have

experience running Facebook ads you'll

notice often that your test might

actually perform really well in your

testing campaign but once you bring it

inside your main campaign it doesn't

have the same momentum that it had

inside your testing Campaign which is

fine that will happen but that's why you

want to just ideally just keep that test

running for as long as it performs well

and with that feature it just doesn't

really allow you to do that so that's

one of the main reason why I'm not

really a big fan of that feature so I've

made this notion document and this is

kind of like a cheat cheat and uh the

link of this document is going to be in

the description of this video you don't

need to give me your email or to

download anything to get it just click

the link and you'll see it but what

we're going to do is we're going to

follow this cuz that's actually uh what

I follow to AB test are uh creative so

the first thing you want to do Step One

is setting up your campaign so you're

going to go to ads manager you're going

to create a new campaign and the

campaign objective needs to be the same

objective that you have inside your main

campaign um so if you've selected sales

in your main campaign I'm going to

select sales here and that's what I've

done but if you're service based company

in your main campaign you've selected

leads you're going to select leads um

and that's really important because if

you don't select the same objective when

you do find a good test you're going to

want to duplicate that ad to bring it

inside your main campaign and if it

doesn't have the same objective you

won't be able to duplicate it someone

could make the point that okay well just

you know just upload and create that ad

inside your main campaign sure you can

do that but if you do that that's going

to be a different ad ID and I mean at

this point you're kind of losing some of

the learnings that that ad as made um

but then the other thing and with that's

probably for me that least the main

reason is you're losing the engagement

so when you're duplicating an ad uh from

an ads set like your testing ad set to

bring it inside your main campaign if

you uh select select the same campaign

objective and you also select the option

that says keep existing engagement

you're going to be able to keep that

engagement and I mean social proof right

CU you're paying for those likes those

comments and those shares and when you

get ideally good comments it's free

social proof and actually helps it gives

your add a little boost so that's why I

always like to select the same campaign

objective so I'm going to select sales

here and I'll usually just select manual

sales campaign and uh go to the campaign

level here so first thing I say here is

like name your campaign testing so I'm

going to go ahead and do

that testing and then this is going to

really depend on your you know your

business model so are you in a special

ad category credit employment or housing

if you are obviously you want to select

that if not you're going to get your ad

account disabled this here use a catalog

I always turn that off um and then

create AB testing I do not use that

feature so I leave that off I always AB

test my ads manually so the way that I'm

showing you in this video and I do not

use that feature because with this

feature you need to add a start and an

end date so we're not going to do that

and I do not turn this feature on right

so that's actually step uh or Point

number two here budget optimization so

choose adset budget optimization so

adset budget optimization is when the

budget is on the adset this is a

campaign right so we don't want to set

the budget here if I were to turn this

on what would happen is it would allow

me to set the budget on a campaign and

it's fine if you only have one test that

you're running but what I'm going to

show you is how to potentially have like

two or three different adset testing

different types of ads and what that's

going to do if you have this on it's

just going to spend the money on the

adet that has the lowest cost per result

which you might think oh well that's

fine like I want to spend the money

wherever it makes sense yes but if one

ad set has been running for longer than

another ad set that one's going to have

momentum and probably have maybe a a

better cost per result than another ad

set and now that new test that you want

to run to see if it performs better and

every time you're running a new test

that probably means that you think those

ads will perform better than any current

ads that you're running so you want to

at least have a minimum amount of spend

that you can dedicate to that ad set and

have control on potentially increasing

that budget decreasing that budget and

forcing basically a specific amount of

send towards that adset and with

campaign budget optimization right this

option right here it's just not really

going to allow you to do that in the

same way that adset budget optimization

would do so I'm going to turn that off

and uh now we're ready to go on the

adset level so for the adset level I

usually will do this so I'll name this

testing and I'll say Broad and the

reason I'm using a broad audience here

is that inside my main campaign I'm also

using a broad audience I usually want to

look at the results that I'm having

inside my main campaign and compare the

results with the results I'm getting in

my side my testing campaign here but if

in my main campaign I'm using Broad and

here I would you use let's say a

lookalike or retargeting audience that

would mean that I'm not looking at the

same thing right what here I'm testing

and what I'm showing you how to do is

how to abest your creatives if I guess

you wanted to abest audiences then what

you would want to do is you would want

to use the same ads that are inside your

main campaign bring those ads inside

that testing ad set and now only change

the ad set so you have one variable and

that's what I'm a big fan of have one

variable so are you testing the audience

or you testing the creatives most of the

time I am testing the creatives that's

why I'm I'm not you know spending too

much time showing you and talking about

this different like audience variations

I'm going to be talking more about

different creative variation because to

me that's what I found just drives the

most amount of results for like you know

input versus output I will usually use a

broad audience and if you're not using a

broad audience inside your main campaign

make sure that the audience here is

Broad enough so don't have a bunch of

Interest targeting like if they need to

meet this and this and this and this and

this in order to fall in this audience

because that means that the audience

size is going to be a little bit too

small okay so just keep that in mind

and another reason you want to have a

big enough audience is because here I'm

using the same audience I'm using inside

my main campaign and inside your main

campaign you're only doing retargeting

most likely what you will see is an

audience overlap so you want to try to

avoid that as much as you can by using a

big enough audience so that hopefully

there's not an overlap but outside of

that you're going to come here and if

inside your main campaign you're sending

people to your website you're going to

select that option here if you're doing

a website and Shop in your main campaign

you're going to select that option right

so you really want to duplicate what

you're doing with your main campaign on

the adset level right here so you're

going to go ahead and select obviously

the same pixel the same conversion event

all the same settings um and this is

where you set your budget so in terms of

budget it kind of depends on what your

cost per result is so if your cost per

result is really low you can get away

with a lower budget because let's say

you're only spending I don't know $100 a

day in this test and your cost for

purchase is $10 well you know every day

you're going to potentially have about

10 purchases but if you let that run for

about 5 days you're probably going to

have 50 purchases and if you only have

four ads 50 purchases dividing with all

those ads like it's going to give you

enough purchase to be able to clearly

understand and identify a winner right

but since it's a demo I'm just going to

set it to let's say 150 a

day and that's pretty much it in terms

of what I would do in the uh ad set but

obviously if you're targeting United

States you would go ahead and change the

location here but this is demo I'm just

going to leave it to default and for me

that's Canada and again I don't I'm not

changing anything anything else here so

let me let me go back here I want to

talk about number of variations so each

adet should have a minimum of two and a

maximum of six ad set so the reason for

that is just like what I said so if you

have too many uh ads it's going to be a

lot harder for you to actually make a

decision on which ad is actually

performing best because what's probably

going to end up happening is some of

them will not get enough spend for you

to be able to make an informed decision

and you have two little ads then I mean

you're just not testing enough at this

point so you The Sweet Spot is between

two and six and if you want to find like

maybe even a better middle ground for

your company again it depends on your

cost per result and your spend so if

your cost per result is low and your

spend is really high you can probably be

closer to the five or six ad that you're

testing per ad set but if it's the other

way around then you know you probably

want to aim around two or three but

never go below two and I don't recommend

companies to go above six unless you

know you're a really really large

company and you're spending a lot on

Facebook ads um the second thing here

that I'm going to talk about is the

different types of variation so what you

want to do is maybe just try different

hooks so it could be like the first 5c

of the video if it's an image maybe it's

just like the the headline that you're

changing maybe it's a copy so same video

same videos or same images and you're

just trying out different copy so I

usually like to grab like let's say a

video concept or an image concept and

just change a lot of different things

about that image or video to create

different variation but what I don't do

and that's what I explain here I try to

avoid using different types of creatives

inside the same adset let me give you an

example I will not have one video and

then testing two different carousels

because at that point they're different

types of creatives and they're not

really equal but usually what I found is

that carousels when they're mixed with

videos and images they sometimes suck up

all the budget so what you want to avoid

is mixing different types of creatives

together so if you're testing images

like I said test different hooks um test

different colors maybe in the image but

test only images and the concept should

be similar and same thing with videos if

you're if you're testing like let's say

you have a specific video you want to

test just make different versions if

it's a ugc video ask the Creator okay

like record this content but the first

like let's say your hook let's try this

hook that hook and like three different

hooks so now you have three different

versions to test but just stick with

Carousel if that's what you want to test

with that ad set okay so for for step

two create enough but not too much

creatives and I give you different

examples of what to test so it could be

a different headline it could be just a

different hook or it could actually be

the same video but like one video is

longer let's say 55 seconds and the

other one is like cropped to 25 seconds

so you're testing long versus short uh

so those different types of tests you

can make and again the biggest takeaway

here is don't mix and match so number

three is launch and monitor so

everything's good with your ad set and

your ads you're going to go ahead and

and click the launch button what you

want to do is you want for the next 5

days to just monitor your ad just look

and see okay our purchases or our leads

flowing in but you don't want to make

any changes yet and just please note

that the review process or the launch

and monitor process will depend on your

sales cycle so if you have a short sales

cycle then regardless I don't recommend

that you go below 5 days maybe 4 days

would be acceptable if you have a really

really uh short s cycle but definitely I

I I wouldn't really recommend it like

honestly 5 days I think is perfect um

but if you have a longer s cycle then

maybe you want to let the ads that run

for let's say two weeks so it really

depends on your company and if ever you

do have a really long sale cycle and

where it's like maybe you're you're at

the end of the day you're trying to see

if those leads are converting into deals

because you're a service based company

then maybe at that point you're not

optimizing for the deals because that

will take too long maybe you're trying

to like identify how many let's say

valid lead this adet generat right so

you're trying to find maybe a metric

that is closer to let's say the deal

stage but is not really the deal stage

because maybe that just takes too long

and if ever you have an Ecom store where

your product costs I don't know $500 so

there's a long s cycle then again maybe

you want to just wait or wait a little

bit longer or find another metric that

you could use to identify like early

success in an adet or with an ad but

guys that's a launch and monitor stage

identify the winner all right so that's

my favorite part so where what you're

going to do is you're going to go inside

ads manager so let's pretend that this

one has been launched we have my testing

ad set and my regular broad campaign

right both of them have the same

audience and I would just go to the

result tab obviously this demo account

so there's no data but I would look at

how many results they generating but

more importantly what is the cost per

result and if you're in the Ecom space

and maybe you're also looking at things

like the return on ad span right you're

adding that to your view and you're just

kind of looking on the adset level now

which one is performing better if if

your main campaign is performing better

than your testing adset don't give up

yet just go now select your main

campaign and select your testing adset

and you want to go to the ad level and

what you're trying to see here is are

any of my ads that have enough data what

I mean by enough data is inside your

testing adset when you're looking at

your results you want to look at this

and say okay because of my budget and my

cost per result and what I see in let's

say a 5day uh period I can actually

confidently see that this ad or that ad

is actually performing well or maybe who

knows none of them are performing well

right but you need to have enough data

to be confident that okay this was a

good test or this wasn't a good test and

if this was a good test let's say one of

your ad inside your testing ad set is

actually performing better than any of

your ad inside your main campaign what

you're going to do is you're going to

duplicate that ad and you're going to

bring it inside your main campaign and

if ever it did not perform well then

what you're going to do is after the f

day again if it has been enough time and

you're confident about the data you can

go ahead and pause that testing ad set

right and then let's say inside of your

testing ad set you find an ad that is

performing well but when you compare it

to your main campaign it doesn't match

the results of your best ad inside your

main campaign well then the next

question you want to ask yourself is it

better than the worst ad inside my main

campaign if that's the case same thing

here you're going to duplicate that ad

and bring it inside your main campaign

and now the question that I get all the

time is okay Cedric so once I brought

that good ad inside my main campaign

like what do I do now well if the ad is

performing well inside your testing

campaign leave it there right let that

adset run and sometimes what I'll even

do is I'll even scale a testing adset

what I'll do is for as long as it's

performing well every two days I'll go

inside my adset and I'll increase the

budget by let's say 20% and I'll can

just continue scaling scaling scaling um

until I see a drop in the cost per

result or again Ron as Bend depending on

the metric that you're using but don't

pause something that works right don't

try to reinvent the wheel so if ever

testing ad set is performing well leave

that running and you can what you can go

is you can kind of just like monitor it

and that's what actually the the last

step is right here is keep an eye on the

winner so whenever you do find a winner

just keep an eye on it monitor it and if

ever let's say you have three ads that

are performing really well inside your

testing ad set and after let's say 3

weeks let's say one of them stops

performing well but the other two are

still doing well then just pause the one

that is not doing well and you're going

to keep running those two that are doing

well again until they don't perform well

and it's kind of like a rid and repeat

process and that's why I say to use

adset budget optimization and not

campaign budget optimization because I

really wish for you that you find an

adet that performs well and that you're

going to leave running for let's say a

month so for that one month period you

don't want to just have one ad set that

is running because you want to test

other stuff so what you do is you kind

of like rinse and repeat so you would

create another adset and in that adset

it would just be like three or four or

five or six different creatives that

you're testing and you're going to do

the same process that you just did with

the prior ad set right so you let it run

for enough days look at your data is it

a good test or is it a bad test and it's

really uh a Rin and repeat process so

guys that is it for this video um now

hopefully you know exactly how to AB

test your creatives and if ever you want

that cheat sheat again it's in the

description of this video I recommend

that you take a look at it even maybe

like download it and when you're AB

testing just reference that sheet and

make sure that you're following all the

different steps but guys that is it for

this video bye for

[Music]

[Music]

[Applause]

now