Best Bits: Save Time & Money on Facebook Advertising with Dynamic Creatives and Broad Targeting

Send us a text

Best bits from Charley's episode - full episode here on Spotify, Apple, and more.

Ever wondered how top brands streamline their advertising campaigns on Facebook? Are you curious about how to make the most your time and money with Facebook Ads? 

We actually use an exciting example of how this method could be (and has been) utilised for a brand like Hexclad, Gordon Ramsay's cookware line. If you want to make your Facebook ads more effective and efficient, this is the episode you've been waiting for.

My guest Charley dives deep into the 3-2-2 dynamic creative method, a unique approach that lets Facebook decide the best combination for the end user, making your ad campaign work like a well-oiled machine. By using engaging content that already worked organically, we go over how this strategy can increase brand reach and fuel the sales funnel.

Enjoy this best bit! If you want the full ep, it's here on Spotify, Apple, and more.

Want more Ellen Ave? Head over ellenave.com now and subscribe to the newsletter. It's the best way to get everything.

---

Website & Newsletter: ellenave.com
LinkedIn Profile: Finlay Barnett
Instagram: @ellenavepodcast
Youtube: Ellen Ave Media
TikTok: @ellenavepodcast

Episode Transcript

Speaker 1 (00:00):
And that 3-2-2 thing.
Do you want to walk thatthrough?
So it's three creators, twocopies, two headlines.
What's like the sort of processof it.
That's what takes like yourapproach of let Facebook decide,
be real broad, let Facebookknow it's best.
This is like a really good,actual application of it.
So how does it actually work?

Speaker 2 (00:19):
So the way it works is dynamic creatives pair all of
your ads together in a way totry to give the end user the
best possible experience.
So if you have three creatives,two headlines, two copies, you
do the math, that's 12 totalvariations.
Now you could make 12 total adsand kind of do it manually
First off.
That's painstaking and you'reprobably going to make a mistake
.
Lord knows, if I have to make12 ads, I'm going to have a typo

(00:39):
on at least one of them.
So the point of that is whenyou build 12 ads, that data is
being collated into 12 differentfiles.
If you look at it as like afolder on your computer, it's 12
different files.
When you make it a 322 ad,dynamic creative, the Facebook
is just looking at the folderitself, so it's not letting each
.
All the ads work in conjunctionwith each other and they're

(01:01):
complementary in compoundedvalue.
So you have in compoundinginterest in the data set versus
cannibalizing each other, wherethe ad that works is probably
the one that got a couple ofgood comments early or like had
a nice like hook rate orsomething that has nothing to do
with performance.
The way we look at it is thesecreatives need to be.

(01:22):
Ultimately, each ad needs toanswer a question and really
what we're doing is justbreaking down elementary school
science class.
It's scientific method.
I have a problem.
So I have a video coming out onthe channel, but I'm making an
ad for Hexclad, the you know thepots and pans company, right,
gordon Ramsay's cookware brand.

(01:42):
So let's assume the problem iswe're having trouble scaling
because we're having troublefilling the funnel.
Okay, so that's our problem.
Our hypothesis is going to bewell, if we use really engaging
content that people really wantto see, we're going to reach a
whole lot of people and get themreally interested.
So that's going to kind of fillthe file, okay.

(02:03):
So then the video is.
I go to the organic social and Ifind, hey, there are three
videos that kind of look similarBecause I'm trying to solve the
problem the same way.
There are three videos where,kind of Gordon's being a, he's
goofing off.
They're kind of, you know, he'staking a piss at himself, he's
just having a blast, right.
He's like he's driving aforklift and like hit something
or he's just kind of goofy,whatever it is right.

(02:24):
They're like kind of fun,stupid little videos and they
all basically have like amillion views, cool.
So let's what?
Let's find what worked and hadlike a million views that are
all of the same concept, right,the concept is Gordon's kind of
being a goof and they're highlypublic.
So they look and feel the same,although the content itself is
a little bit different.
So now our hypothesis is if werun these ads that earned a lot

(02:50):
of reach organically, wherethey're all kind of Gordon being
a goof, we think that will geta lot of spend because people
clearly want to see it, becauseit works well organically and,
as a result, it will fill thefunnel and let all of the other
ads that kind of do someretargeting or stuff like that.
It'll let them have more peopleto touch.
So we're going to reach morepeople and let more people know
that hexclad exists and get theminterested.

Speaker 1 (03:12):
Awesome.

Speaker 2 (03:14):
So now the scientific method applies and like, okay,
well, we have a controlenvironment, our best ads and
we're going to run this againstthat in a CBO campaign and
they're saying, great, now thetest goes live when we launch
this, does it earn spend?
And if it earns spend, does ourbank account look better?
And the answer is yes, awesome.

(03:35):
We have an end and what we cando because we have the three
creatives.
If it doesn't work, we can sayif any of the delivery was,
you're going to have one or twothings, either one ad is going
to get all the spend or thespend is going to be kind of
evenly spaced amongst all three.
Now, if the spend was evenlyspaced amongst all three, then

(03:56):
you know it's not really thecreative on why it didn't work.
Like, just this is an ideadoesn't work.
Now, if all the spend went toone video and it did poorly,
well, we can remove that videoif we wanted to.
Or we'd like, hey, no, thisvideo isn't it Right, let's try
another one.
That'd be fair.
Probably I would just turn itoff and move on.
But we have the choice.
We can be very binary.

(04:16):
Let's say we have seven or tenof them in there it would take
forever.
For us it's not well.
All three were even and one gota lot.
It's probably going to be.
One gets 80%, 70%, one getslike 20, something else gets 15.
And you have no idea why.
But when it's three it levelsoff or it becomes hyperbolic
really fast.
And the same thing happens inthe copy.

(04:38):
Maybe one headline gets 90% ofthe spend, or they're both even.
If they're even, then you know,hey, the headline doesn't
matter.
If the copy's the same, cool,they're the same, and this copy
doesn't matter, which meanswrite two new sets of copy for
the next test, because thisclearly has no impact on
performance.
It was even.
None of them were better thanthe other ones.

Speaker 1 (04:57):
It didn't really move the needle at all.
No clear winner yeah.

Speaker 2 (05:01):
Yeah.
So we have a binary choice.
Did it get spent?
Yes or no?
If it got spent, did it helpthings out?
Yes or no?
And if it did, if it earnedspend but it wasn't helpful, was
one copy or one creative getall the spend in the world, yes
or no?
Or was it even yes or no?
Like all the answers are yesand no.

(05:23):
Logic trees yeah, and you'll beable to know that answer really
quick.
So the beauty of it is itallows you to just make what
used to be an art form and, tobe fair, I'm not saying the
creative director's notimportant.
Look, people are really good atmaking ads.
When they use dynamic creatives, they're phenomenally better
than me making an ad.
I'm terrible at making ads.

Speaker 1 (05:46):
Good at making them spend, though, oh, I'm good at
that, absolutely.

Speaker 2 (05:51):
So, yeah, that makes it just so easy.
It becomes that there'sobjectively no reason to not do
it, because ultimately, allyou're going to do is, if it
spends and makes your bankaccount better, then every
iteration improves yoursituation, and as long as you're
continually getting better,then awesome.

Speaker 1 (06:06):
And then you have this next step as well, of when
you find that winner.
You put that into like awinning ad set or campaign a,
but then you introduce new onesto test.

Speaker 2 (06:15):
Yeah depending if you need to.
I mean, if your target cost fora sale is 50 bucks and you're
coming in at 42, just spend moremoney.
The reason you were launchingnew tests is because you're not
good enough.

Speaker 1 (06:27):
If it's working, why fix it?

Speaker 2 (06:28):
I think one of the biggest problems people do is
they launch creative tests forno good reason.
They try doing a landing pagetest for no good reason, like if
the biggest problem you have isyou have an untapped potential.
Just have the potential spendmore money Like job done.
We'll come back next week.