LongCut logo

I Found the BEST Facebook Ads Strategy after ANDROMEDA Update (+43% ROAS)

By Antonio Ventre

Summary

## Key takeaways - **Classic Facebook Ads Strategy is Failing**: The traditional testing and scaling campaign structure for Facebook Ads may lead to a decrease in ROAS following the Andromeda algorithm update. [00:05] - **New Strategy Boosts ROAS by 23-41%**: A newly tested Facebook Ads strategy has been implemented, improving ROAS by 23% to 41% for e-commerce clients. [00:17], [00:22] - **Creatives Drive Targeting Post-Andromeda**: The Andromeda update shifts focus to creatives for targeting audiences, making traditional audience segments and lookalikes less crucial over time. [02:11], [02:21] - **Consolidate Ads, Minimize Ad Sets**: The optimal strategy involves decreasing the number of ad sets within a campaign and maximizing the number of diverse creatives (15-50) inside each ad set. [03:20], [03:44] - **Full-Funnel Targeting in One Campaign**: A single campaign with minimized ad sets can effectively target top, middle, and bottom-of-funnel audiences by leveraging creative diversification and broad targeting. [07:31], [10:43] - **Creative Diversity is Key for Algorithm**: To succeed with the new algorithm, creatives must be diverse in concept and angle, targeting different user awareness stages to avoid repetitive targeting. [02:31], [08:07]

Topics Covered

  • Meta's Andromeda Algorithm: Your Old Ad Strategy Is Dead.
  • Creatives, Not Audiences, Drive Meta's New Targeting.
  • Fewer Adsets, More Creatives: The Optimal Meta Structure.
  • Meta Auto-Optimizes Full-Funnel, If Creatives Are Right.
  • Meta's Budget Allocation: New Audiences Still Dominate.

Full Transcript

If you're still running the classic

testing and scaling campaign after the

meta new Andromeda algorithm, bad news

for you, your rowas might go down soon

if most probably is not already down.

And we have tested a lot of different

strategies across all our e-commerce

clients so far and we found the best

strategy that is basically helping us to

improve our ROS by between 23 and 41%.

And in this video, I'm going to show you

guys exactly what is this strategy. So,

what is the new winning meta account

structure, which is basically based on

what Andromeda algorithm wants you to do

to maximize the performance. I'm going

to show you some test and differences,

some before and after in terms of rows

that we have generated with this new

strategy. I will also go through the

exact way that this new algorithm works

and why you should keep this into

consideration when you build the ad

structure and strategy overall for your

e-commerce brand right now. Okay, let's

jump on my screen. So, this is the

newest Facebook ad strategy uh that is

working post andromeda algorithm that we

use of course to uh help our clients to

scale with a very high return on the ad

spend. So, as you can see this is

literally just the before and after. Uh

this is the performance of this account.

uh on July before we applied the new uh

Andromeda strategy and this is was the

after. So as you can see the overall

return aspend increased by 23%. We went

from a 3.83

to 4.71. And on some campaign like this

one specifically we went from from a 430

return on ad spend to a 6.08 return on

the ad spend. And this one in red again

is the old ad strategy that is not

working anymore. And this one is the new

ad strategy that is working especially

well nowadays. So let me actually

explain what the new meta andromeda

algorithm is. So this new algorithm

basically is going to uh maximize the

target efficiency. So the way that you

target your audience uh towards content

and so what you basically need to do uh

the basically like the overall outcome

of this is that meta is now improving

the targeting and even more of your

targeting is done through creatives

right now. So audience and segments and

lookalike and all these things will

matter less less and less with time and

each of your single creative will target

different audiences. So what is the main

thing to prioritize here as you can see

is the create is the creative

diversification. Creative

diversification means how diverse are

your creatives and diverse I don't mean

changing the background or uh just a

simple thing like the color of the CTA

and they need to be actually diverse in

terms of angle concept and everything.

And this is basically how it works. You

see this is directly from meta guys. As

you can see, meta takes all your ad

creatives, reorganize them into

different buckets of audiences, and then

sub audiences, and then it will show the

related ads to the audience where this

ads is most relevant for. Here, this is

like a summary of how you should adapt

your strategy based on this new

algorithm. So, basically, you don't need

to split creatives by angular messages

anymore into separate adsets. So your

goal overall now is to decrease the

number of adsets inside your campaign

and maximize the number of ads inside

each single adset. So what is working

right now is basically not splitting

anymore the ad creatives by different

angles putting like six, seven, eight

different variations of ads inside one

adset and then test another angle in one

different adset. But now what you can do

is basically just put all the creators

inside one adset. You can be like 15,

20, 30, 40 different creatives, even 50

maximum creatives inside one adset and

increase the budget within that single

adset and create new adset when you have

reached the maximum capacity or when I

suggest to do it is basically when the

learning uh of the adset is maximized.

That means that it's not learning

limited anymore. So the the the adset is

performing well with a good return

aspect. In this case, you don't want to

touch the overall adset. And this is

when we basically need to create a new

adset with new creatives inside. As you

can see, what meta says itself uh is

that you can basically put different

angles and messages into the same

campaign or adset EDL advantage plus of

course and and let the system optimize.

So what is the uh what is the uh test

the results from the test that we got.

So as you can see this is the first one

and look here this is June uh 1st until

30th of June we got a 4.68 68 return on

the ads per on this account. And as you

can see, we had 1 2 3 4 5 6 seven

different ads. What we did, we basically

just increase the number of ads within

the same adset rather than splitting

into a new adset. And um and then the

rows, sorry, this is still July, so it's

June and July. Nothing happened. Same

number of of ads within the same adset.

And then we have increased the number of

ads to 11 from 7 to 11 uh in August and

the return ad spent went to 6.08

and uh what a lot of people were doing

right now is just like creating a new

adset and everything so it get optimized

again. But I think is that the the adset

already has a lot of uh data and so what

you can do when you put input new ads

within the same adset, you're going to

increase the number of um success rate

within that specific creative because

you're putting those creative inside an

adset that already has a lot of data

acquired. Now the thing is that uh the

ads the go like our goal to stabilize

the performance of the ad account is to

make sure that we have as many adset as

possible outside from the learning phase

and the learning phase itself happens

within the adset level. So if you're

splitting by different adset even if you

have a CBO it's going to be difficult

for each single adset to collect 50

let's say purchases within a week. And

so what you want to do, you want to make

sure that they are all stacked within

one single adset. And there is also

another example here, another different

account. And as you can see, uh this we

had three different creatives inside

this adset with a 4.24 return on the ad

spend in uh between June and July. So

two different months. And then what we

did, we had just added new creatives

inside. And as you can see, we went to a

550,

sorry 5.43

on the ad spend. And in total we had 13

creatives. So we add 10 new creatives

inside the same adset. Now uh when you

should do this and when you should not

do this. So as explained before you

should avoid doing this when you have

already an asset with 20 30 40 different

creatives inside and the adset is

already has already exited the learning

phase. You have another batch of 10 20

30 different creatives. Rather than

putting this creative inside an existing

adset, what you can do is just to use

another adset. Since you have 30 20

creatives, you can put all these

creatives inside the new adset. Um, and

the goal, your overall goal is to

minimize the number of adsets and to

maximize the number of ads within each

single adset. So, this is going to be

your goal. The campaign structure so

that is working right now as you can

probably understand is basically just

having one campaign. Uh and in this

campaign we have one or more adset

within a CBO setup. Um and uh the number

of adsets is basically minimized. So you

want to have as minimum adsets as

possible and maximize the number of

creatives. This is the campaign

structure and this is very important

because what you want to do here which

what you can read here we do not we do

not exclude

audiences. We do not exclude existing

customers. But in order for this to

work, you must you must create creatives

for different user awareness stages

because if you only create creatives for

the bottom of funnel uh awareness stage,

then your ads will keep targeting the

same people over and over again. So the

strategy only works if you do two

things. The first one is this one that

I've just mentioned. And the second

thing that you have to do is basically

make sure that you are uh putting all

your audiences here because what meta

does now basically meta targets

different people based also on um on

what you need like meta knows that you

want to grow the business in order to

grow the business. Meta knows that they

like you cannot spend just on existing

customers to give you the best spend. So

Meta nowadays, especially with your

Andromeda, they knows that you need to

have enough top ofunnel audience and

traffic that can sustain the middle and

bottom of funnel. And so what you need

to do is let Meta understand what this

middle and bottom funnel are in terms of

audiences. So Meta knows that it doesn't

have to push a lot of budget towards

this creative sorry towards these

audiences. So if you don't do this so

basically if you don't add your website

visitors to cart the checkout initiated

excluding the purchases into the engaged

audience as well as uh the clab your

list of uh you know all the contacts

excluding the existing uh customers into

the engaged audience uh meta is going to

basically probably push all your budget

over this audience because meta doesn't

know that this audience is an engaged

audience. So you must do it and then you

should do the same for existing

customers. In this case, what you want

to have is the pixel data purchases last

180 days and uh the email list in sync

with Cleio or any other tool you're

using. And you want to filter by all

customers. So a list of people in Cleio

that have made at least one purchase

from your website. So all your existing

customers basically make sure this is a

dynamic list and also this is a dynamic

list because if it's not a dynamic list

it's not going to be updated and uh it's

not going to keep you know the

performance. Um and then finally the

asset level actually very very broad no

custom exclusion and everything. Now a

lot of people might be saying oh um meta

is going to spend for sure the budget to

existing customers and everything but in

reality it's not true and uh I'm going

to explain with this example here below

but before going towards uh that example

here I want to you know just finalize

the um ad structure that you should use.

So all the other things right now

becomes optional because meta

automatically do the retargeting and the

retention within the prospecting CBO. So

the prospecting CBO doesn't even become

a prospecting CBO. It becomes a single

campaign like a fullfunnel CBO where

meta spend budget towards new customers,

engaged audience and existing customers.

Right? So you might need to use some

retargeting uh campaign or some

retention campaign if you run specific

brands just in specific cases. So uh

needs to be you know you just you kind

of need to understand the different

situation where and when you should use

it. Uh but what you want to do is just

keep it very simple. Manual targeting

for the retargeting uh manual campaign

uh website vistors Facebook engages

cleavio audience only inside exclude

existing customers because you are

retargeting only everyone that didn't

buy anything and uh broad placement one

adset 15 to 50 ads per adset and uh the

ideally you want to have one single

adset here as well and uh the retention

campaign manual targeting existing

customers only use P surplus clio data

broad placement one adset 15 to 50 ads

within the single ads. Why? Because

since these audience are very small, the

frequency will be very high. And so you

need to make sure that you have enough

content that keeps you know retargeting

the same people just to avoid the the

same audience see the same content over

and over again. Otherwise they will

start they will start reporting you and

if they reports you it's going to be a

huge problem because your CPM will go

very high and the only way to actually

see this is you basically need to have a

meta representative and speak directly

with them because they can only see it.

Another thing that you need to do is

when you have promo campaign. So let's

say you have like a Black Friday

campaign. Cool. So in this Black Friday

campaign, what you have to do, you still

need to keep your overall campaign set

up active. Um so one campaign CBO per

single product uh evergreen and then you

have an additional campaign where it's

going to be a broad setup similar to

this one. And then what you can have is

have a retargeting campaign website vis

engagement only if there is a specific

promo. like like in this case for Black

Friday uh you don't need to exclude

audiences in here and you don't need to

exclude audiences in here so we keep the

broad open the retargeting open and the

retention open as well I mean the

retargeting you still need to target

website visitors engagers and cleavio

audiences but you do not uh you don't

have to exclude any existing customers

as well as in the retention you don't

have to exclude existing customers why

because this one this campaign here so

you might be like Okay, but this

audience then is going to target the

same people uh that we're target here.

In reality, this is not going to happen.

Why? Because since you have this

campaign already here, then meta is

going to push the broad more towards um

new audiences compared to this one. And

this one is probably going to perform

very well for just like a couple of days

or maybe like towards the end of the

campaign. That's why usually this one we

launch it then we keep it active for

like 3, four, five days at the beginning

at the beginning of the campaign. and

the performance will drop and then we're

going to turn everything off and keep

only the broad active. U this is not

mandatory. I would extend the optional

even to the retargeting part because the

broad if you only keep the broad is is

also good because the broad will also be

able to um you know retarget the engaged

and retention. If you don't put these

two like the broad itself will push more

budget towards this especially during

the beginning of the campaign. Uh and

this actually good from one side. Why?

Because when you launch the campaign,

Meta will get a lot of data and we'll

see probably a lot of purchases coming

from existing audiences uh existing

customers or like all their targeting

audiences and so the optimization of

that specific campaign will be much

faster towards new audiences because

meta is going to leverage the data

collected kind of a lookalike concept.

So meta will leverage all the purchases

and analyze all the audiences uh like

all the behaviors and you know pattern

and everything of the audiences that

just bought your product. So probably

going to be retargeting audience and

existing customers and then it will

start pushing more towards new audience

that are similar to this and so the

targeting of the of the campaign will be

improved basically because the adset

already has a lot of data and um now

let's go to the concept of existing

customers. So here this is a screenshot

from our uh campaign right we don't

exclude anyone and this is what is

basically happening here so we spend 18k

in this account and as you can see 65%

of the budget went towards new audience

then

31% of the budget we went towards

engaged audience just want to remind you

that engaged audience is

still prospecting so is net new

customers at the end of the day because

these people, they never bought for

anything from you. And then you have the

existing customers and only 3.4% of the

total budget went towards uh the

existing customers. And as you can see

here, uh this is the return on ad spend.

So of course it's going to be higher for

the existing customers. It's going to be

uh the lowest for the new audience, but

overall it will be balanced. And this is

how we are basically doing a 4.56 return

on the ad spend because we want to

maximize for the return on ad spend

because even if you do some retargeting

campaign let's say email retargeting SMS

and everything you're probably not going

to reach all your existing customers and

so you still want to have some of your

campaign like to target your existing

customers especially if you sell

different products. So why meta here and

there's no spending uh on existing

customers and the answer is short

because we know what we're doing that

basically means that we create ads based

on different awareness level. So if you

don't do it you will mess up because

meta will start pushing towards existing

uh audiences like engaged audience and

existing customers a lot and everything

is not going to work. Now meta uh is

basically going to create a fullfunnel

uh targeting inside your single

campaign. So basically the al is

basically what I just mentioned before

that the algorithm knows exactly that it

must be a top of funnel to keep the

campaign profitable over time not just

today. Okay. And so it creates basically

a flywheel of content. You put all the

con inside and meta will allocate and

target different people on different

stages based on the messages that you

create within the ad creative strategy.

So the creative strategy becomes very

important because if you don't know how

to do it if you don't know how to create

messages for different audience segments

and for different user awareness stages

this probably is not going to work

right. So just make sure to follow this

up. uh and just make sure to shape and

use the proper messages and to create

creatives properly for different user

awareness stages. Otherwise, this is up

here. Probably it's not going to work

and you still need to exclude exist

existing customers. Of course on the

side what you can do you can also have

specific campaigns that push towards

prospecting fully called campaign

because maybe you want to have an

incremental reach there because probably

you're already spending a lot of money

on ads and so you want to have a

campaign that is basically going to

increment your reach new reach as much

as possible and so you probably need to

exclude a lot of different things like

engaged audience website vising existing

customers and everything because you

want super cold audience you want to

force it But again, most of the time

this is already working by itself on

Meta and we're seeing pretty much good

results even like 250 100k spent per

month US dollar. So this strategy is

what we're doing right now. These are

the results and this is basically it for

this video. If you like the video,

remember to leave a like, leave a

comment, subscribe to the channel, and

I'll see you in the next

Loading...

Loading video analysis...