Latest / AI DTC WTF / The Ads That Spend $1M Are Still Written By Hand | Harry Delmege

Transcript

The Ads That Spend $1M Are Still Written By Hand | Harry Delmege · AI DTC WTF

0:00
We all hear a ton about AI. We all hear about
0:03
what it can do and how amazing it all is. And I don't
0:06
think that there's anything incorrect about that.
0:08
But this episode is brought to you by an AI brain
0:12
that is absolutely incredible for your marketing
0:14
team called Runeth. It's from Motion and therefore
0:17
you know it's awesome because it's from
0:18
Motion. But really what Runeth does is connects every
0:22
single function of your marketing department into
0:25
one place. It can do so many different things. It
0:28
build you ads. It can verify that your brand
0:31
looks proper. It can verify that what your top
0:34
creatives were and you can ask it. You can have
0:36
it live in Slack or you can have it live as a CLI
0:39
so you can interact with it any way you want to.
0:41
The thing that's really insane to me about Motion
0:44
is how absolutely intelligent it truly is and
0:47
how it works cross-functionally across the team.
0:49
There was recently a marketer that Reza talks
0:53
about in the landing page on Runeth which you
0:55
should definitely check out where basically she
0:57
went on vacation the CMO did and said if Runeth says
1:00
the ad is okay to run then it is okay to run.
1:04
So she uploaded all the context for the team.
1:06
They ran it into Runeth and Runeth said yeah that ad
1:09
is totally on brand. It's absolutely crazy a number
1:12
of things that this can do and I think the fact
1:14
that we all live in silos and the fact we all live
1:16
separately is really really challenging and one of
1:18
the hard parts about AI. So that's not what this does.
1:21
This brings it all together with shared context,
1:24
shared resources and it also looks at thousands
1:28
of ads every day so it's always up to date on
1:30
what actually is the best performing ads that it's
1:33
seeing across Motion and so that's another whole
1:35
thing that it does. It's really crazy because instead
1:38
of one more dashboard you get a teammate in
1:40
your Slack that knows your performance data cold.
1:42
As I said you can ask it really anything and
1:44
answers instantly. Weekly recaps, top creatives,
1:46
which creators your audience trusts. It learns
1:49
your brand, your playbooks and gets smarter every day.
1:52
Ecommerce and D2C teams are using it to turn
1:55
weekly reporting from a two-day job into a two-minute
1:58
conversation. Head to Runeth.com and see it for
2:01
yourself and it also has a 30-day money-back guarantee.
2:04
So if you need a cheat code in ecommerce and D2C
2:08
with your ads and your ad creative and your marketing
2:11
team check out Runeth.com. 30-day money-back
2:14
guarantee. It's absolutely incredible what this
2:17
thing can do and I encourage you to give it a shot.
2:19
What's up everybody? We are live recording a
2:44
podcast of AI D2C WTF which is our new podcast.
2:49
podcast if you haven't heard this or checked it
2:50
out and I'm here with my man Harry Delmage who
2:55
is has always been a good has has always been
2:57
an awesome person and I feel like whose stock is
3:00
absolutely just exploding right now talking about
3:03
so many different things especially with ads and
3:06
yappers. The episode we did with him on scalability
3:09
school is everyone wanted to see it. It was it was
3:13
just hot hot hot hot hot. So Harry thanks for being here man I appreciate it.
3:20
Oh good man thanks for having me happy to run it
3:22
back. We're here talking about you know why AI isn't
3:28
great for creative and what happens in the creative
3:31
process so we're actually kind of you know we talk a
3:33
lot of about AI on the podcast obviously but for
3:38
you it's not something you use very pretty much at all
3:42
in terms of making great ads. There might be
3:44
some areas where you use it in your process
3:46
but tell us kind of where you use AI and where
3:50
you don't and how you view that in terms of its
3:54
negative impact almost on creating better ads
3:56
sometimes. Yeah so I mean to be fair I wouldn't
3:58
want to say like I don't I don't use it I
4:00
definitely use it but like I think like we use it more for
4:03
for execution right like you know but like
4:06
less ideation but like because I think that's like
4:09
what a lot of bottlenecks are it's like if you're
4:11
a great strategist you have all these great ideas
4:14
but you're just like well I just don't have
4:15
enough time to write these all down but I think with AI
4:18
you you do not have it so it's like you know for
4:21
example where we do use it is like I'm constantly
4:23
building skills or trying to improve skills
4:25
that are like format dependent so we have like a
4:27
Yapa skill a VSL skill like a native ads you
4:31
know they're like the long form copy ones like have
4:33
all that but really what we don't use it for is
4:37
ideation like that's like a lot of manual stuff
4:40
just because I think there's no way that maybe I'm
4:43
wrong but like right now the AI can come up with a
4:45
better ad idea than you I think like you know
4:48
because it has different guardrails it can't like
4:50
you know all this different stuff baked into
4:52
it so I think you know we're very bullish on
4:57
like manual ideation and then just like AI
4:59
execution but saying that we also do find like
5:02
write one bit of manual copy like per week and
5:06
I'm probably actually going to start doing bounties
5:07
for that like whoever writes the best banger
5:10
that's all manual like gets like a like end of
5:12
month bonus or something on top of like the
5:14
commission they already get nice nice so what you're so
5:19
you have skills that are let's walk through that
5:21
so you have like a skill that knows how to write
5:24
yapper ads and you built that skill in Claude and
5:28
it and it it has every yapper ad you've ever written
5:32
along with instructions from you I assume uh
5:35
not every yapper ad we've ever written but like
5:38
all the all the best ones uh and it's got like a
5:42
lot of different contexts right like you can build it
5:44
like you know I ended up doing quite a few consulting
5:46
calls um and a lot of them around yapper ad so I got
5:49
all the firefly notes and then baked that into like
5:51
a new skill um you know took the breakdowns from all
5:55
the winning ads like broke down like you know
5:58
copywriting fundamentals basically like the way
6:00
that we write yappers like it's going to change
6:02
it's like baked into that skill but yeah only take
6:04
the winners and then you know if there's any
6:06
fireflies like any team meetings about it you just
6:08
like get Claude to like scan that and see if it
6:11
can like add anything in I think recently we were
6:14
comparing output and from the original one that
6:17
I've written um only now have we beaten the output
6:21
with like the new skill and I do think that came
6:23
from like all the firefly notes and stuff like that
6:25
so I think if people aren't putting those into
6:28
their skills they're they're missing out for sure
6:30
I mean so you go through and you have that skill
6:35
and then you have what are the other skills that you
6:36
have like VSL um native long form like you know
6:41
those ones where it's like the the image and then
6:44
that like really long ass uh primary primary copy
6:48
uh that basically for any format you can think of
6:51
we have that as a skill um you know we have like
6:54
you know ad analyzer skills uh anything you want
6:58
to do with um as you just basically want to scale
7:02
and give it as much context as possible so you don't
7:04
have to keep like you know prompting or re-prompting
7:07
stuff it just knows how to do it the way you want it to
7:09
right so you have these built in and you're
7:13
going through and then within the process of ideation
7:17
of creating better ads um well I mean we don't
7:20
have to get into that actually let me put a pin in
7:22
that for a second why do you think the ads that
7:26
you're manually writing are better than what it comes
7:31
out with is it because of the human experience
7:34
is it because of you know AI is just it has those
7:38
guardrails so it doesn't know how to think
7:40
outside of that like what is it about it because I think
7:43
you know you look across the we look across all
7:46
creative right now and performance creative and
7:48
those ads that are working why are they working
7:50
what's what is part of it that's work like that
7:53
does it and I think you've had a higher hit rate
7:55
than a lot of I mean there's a lot of people out
7:57
there that have high hit rates I guess I'd be
7:59
curious of your angle has been we're going to
8:02
manually do this and do you know work on copy and
8:05
make sure the angle and that like this there's a
8:06
long form narrative almost to it what else are other
8:10
people doing do you think along or and also how are
8:12
you thinking about it in terms of AI and what it
8:14
does yeah I don't know so many different ways I guess
8:17
I could answer that but I guess like the reason
8:19
we're very bullish on manual research and we also do a
8:26
manual research bounty as well so like every month
8:29
we run one like from manual research like you post
8:33
it into the channel the thing you found out and if
8:34
that like brings around like the biggest unlock for
8:38
that account you also get like uh like an extra
8:41
bonus at the end of the month I think the reason
8:44
is like obviously as we touched upon like some of
8:46
the ideas it just it just can't come up with like
8:50
I don't know maybe if they're like pretty like
8:51
crazy or out there ideas like maybe like gray hat
8:55
ideas whatever you want to call it like you know
8:57
it's it's not going to want to do that too much
8:59
I also feel like you know I've played around with
9:01
it so much and tried to like extract stories from
9:05
reddit with AI and it just I don't know for whatever
9:08
reason maybe I'm just doing it wrong I could be doing
9:10
it wrong but like it doesn't really pick up on any
9:13
good stories and it just skims everything rather than
9:16
actually getting like any kind of good good
9:19
thing like obviously based on the keywords it might I
9:22
don't know I just feel like it just doesn't give
9:23
me what I'm actually looking for it just kind of
9:25
gives me like oh this kind of thing could happen
9:27
and then also like you just have to trust AI like
9:30
face value and like I don't know maybe other
9:32
people are seeing I'm certainly seeing them like
9:33
talk about it like Claude's output seems to be
9:37
unanimously getting worse so it's like well it's going to do the
9:39
same for research and if it's just like I feel like
9:42
I don't know maybe again I'm using it wrong but the
9:44
amount of times I'm like no take your time make
9:46
sure you found this add in these guardrails like
9:49
by that time I might as well just gone on to like
9:51
some subreddit found the story myself and then I'm
9:53
just like off to the races you know I mean like
9:55
I don't think you can get deep research like deep
9:59
knowledge on something just from like research
10:01
alone especially if you're not like a like if you're
10:03
cracked at ads like I'm sure you could the amount
10:06
of time you're spending going through back and forth
10:08
with AI anyway you might as well just fucking
10:10
gone on reddit in my opinion maybe I'm wrong but like
10:13
you know yeah it's almost like you have you have
10:15
to prompt it and give it so many instructions on
10:17
what you're trying to do that it takes a long
10:19
time and you know I also think there's a piece of the
10:23
the way that you connect dots as a human being of
10:25
like what actually is interesting in terms of an
10:28
angle that maybe that you know you can mine I
10:30
mean I think you're using it if you use it to mine
10:33
reviews and to mine you know certain things
10:36
like data that you have on post-purchase you know
10:39
surveys or whatever then that's one thing
10:42
because that's not something you want to do manually
10:44
but in terms of going and doing it on reddit
10:46
like it's yourself I think that a lot of times it's
10:49
harder for Claude to do that anyway like and it's
10:52
easier if you go in and look around and say oh it's
10:54
kind of interesting and whatever how do you I
10:56
mean okay so right now looking across others that are
10:59
really you know I know you're you're looking at
11:01
other high-performing ads a lot what do you think
11:04
are components that uh AI has played into those
11:08
to make them successful not necessarily from you
11:11
but from others like what are they doing is it
11:14
is it like that their AI has helped them to create
11:18
uh rapid iterations of animated ads or you know
11:22
what what is it that they're allowed to do um and
11:26
I'm just curious well yeah I mean I think a
11:29
lot of it comes down to like what you can do
11:31
visually like there's a lot of constraints
11:33
where like if you do want to come up with these wild
11:35
stories you know it's kind of hard to do
11:39
sometimes with a real person you know I mean so like you
11:42
know something I've been seeing like crush is just
11:44
like you know Pixar stuff but even like Pixar music
11:47
song stuff you know I mean like you can't really
11:50
do that with a person um then like other things
11:55
where like you know some people are using like AI
11:58
b-roll to show situations that again with a person
12:01
like I saw one where I think it was about a story
12:05
about you know how someone went abroad and it like
12:08
led to this like discovery of something well getting
12:10
a UGC creator to do that in any short amount of time
12:13
is just going to take you I mean it might be fucking
12:16
nigh impossible to get someone to do that then to go
12:18
like get the b-roll in this other country but like
12:21
with AI you can have that tomorrow you know I mean
12:24
so again I've seen that work really well uh I've
12:27
also seen AI UGC work really well um you know but
12:32
the one thing I will say with it that maybe again
12:34
others haven't seen that I have is like it fatigues
12:36
quickly you know I mean like you have to
12:39
constantly with the disclosure and everything like it's
12:43
because now in the U.S. like a lot of ads have a disclosure basically uh default
12:48
so you've seen you've seen those work even
12:51
with the disclosure on there but they die faster
12:54
uh I mean to be fair like I'd have to look at like
12:57
the disclosure again like we don't really do many
12:59
of them so like we're never in a situation we
13:01
have to like disclose um but I'd have to check but
13:05
even before that like they were like working good
13:07
but like it just would die out quick but you know
13:08
there was that trend where someone was like
13:10
speaking on stage you know what I mean and as like a
13:13
presenter like that that worked good but I think
13:17
just like whereas like if we write a yapper and
13:20
it like bangs like that can last for months and
13:22
months and months like at the top of the ad account
13:24
and like we're seeing that now rather than being
13:26
like okay it's crushing then we have to like whip
13:29
something up uh like there's like an account we're
13:31
in right now where it's like they have like a UGC
13:33
that's crushing we've done some other stuff while
13:35
we're getting the the appers it might take over for
13:38
like a couple of days but then that real UGC just
13:40
comes back to the top again you know I mean so it's
13:42
like you know I'm sure when we get like more of
13:44
the real UGC like we can defer on it like you know
13:47
properly but again it's just something I see is
13:50
like it can work but it just doesn't have the legs
13:54
so I'm just like well why not just do that and do
13:57
the hard thing and just like you know real people
14:00
real research I don't know yeah yeah yeah totally
14:04
I agree it's interesting like how do you think it's
14:08
going to evolve in terms of looking across a
14:11
successful ad account like is it is it going to
14:14
be um you know for example would it be that
14:21
you'll look in a successful ad account that's spending
14:23
200 grand a month and they have you know
14:26
the yappers they have uh ads that are you know
14:32
you like generated animated stuff from you know
14:36
uh ai um they have a bunch of ai generated statics
14:40
like what is what is it going to look like I
14:42
mean because I know it's sort of some of places look
14:45
like that now but I'm curious what your take
14:47
is yeah I mean as you know I'm a fan of a fan of
14:51
diversity and I think ai has made that like so
14:54
much like I think that's genuinely like a huge net
14:58
of ai is like now the new amount of like diversity
15:01
you can bring to accounts like you know we did this
15:04
uh ads that like popped off for this like pet
15:07
brand and we had like the dog talking it and like a
15:10
claymation and stuff like that like that just
15:11
wasn't possible like you know for the average advertiser
15:14
like three years ago so the fact that you can
15:17
just like you know pretty much do that and have that
15:19
generated like in a couple of days now is insane
15:22
um yeah I mean like to 200k I think you can just
15:26
you know just pop off with like one format but
15:29
I definitely see the ones that are like you know
15:31
when they hit like you know big spend is it's
15:33
definitely like very diverse content like founder
15:37
content uh like yapper content maybe like more
15:40
performance UGC statics like ai like you know you're
15:44
definitely seeing a lot more of those come in um
15:48
at those like higher spenders and like all of them
15:50
like you know I guess the ideal thing is like
15:52
they all spend like a decent amount and you can just
15:55
you know keep increasing spend with it rather
15:57
than just like them fighting for spend and just
16:00
the spend level staying the same yeah so
16:03
basically uh there's the you know it's right it's going to
16:08
continue to be used for diversity in what you're
16:10
saying you know um which is good so one thing I've
16:15
been curious about with you is let's say you know
16:17
we're looking across Andromeda and Lattice and others
16:21
right um and it's we know that meta wants diversity
16:25
and that's what we've been aiming and building for
16:28
do you think that as we're looking forward if
16:31
I run a creative shop what's your advice to me
16:34
in reference to like uh how do I approach building
16:41
a bunch of different ads for clients is it like hey
16:45
let's let's make sure that we're doom
16:47
scrolling a lot that we're going on tiktok a lot
16:51
to unlock new styles and immediately you know
16:54
then we're rapidly iterating those for clients
16:57
um like what's the how do I sort of set up
17:01
my organization to continue to make great ads
17:04
utilizing ai but also not utilizing ai yeah cool
17:08
I guess I'll give an answer for both I mean also
17:10
someone's like I've been explained like Lattice
17:12
like three different times I'm still not even really
17:14
sure what it means um I'm not even really
17:18
sure what Andromeda means apart from diversity
17:20
I don't know uh so I mean I'll be honest I'll
17:24
be honest with you it's like uh but yeah I mean
17:27
in terms of and this is where we come back to
17:29
ai right like you know we talk about like review
17:30
mining a hundred percent just do that with
17:33
ai that means you can doom scroll uh what your
17:37
prospects feed looks like even more you know I
17:39
mean like you can type in the niches and watch even
17:42
more like instead of just going through
17:43
every fucking review one by one and being like oh
17:46
that's another person that said very good okay
17:49
that's like pointless like I think what it looks
17:52
like in the future is just like really working
17:55
out where is great to use it and that's still
17:58
something that we're trying to work out as well
17:59
and that also might change in six months because
18:01
maybe ai will be able to pull stories like
18:04
absolutely perfectly and then take that and just
18:06
crank like an amazing yeah presence like okay
18:09
great but like I think going forward it's like and
18:12
what it is now is it's like you know people like
18:16
using a blend and working out okay if I can review
18:19
like where can I save time and where can I where
18:22
should I be putting more manual time and that's
18:24
the thing is like you know obviously we've had
18:26
like a pretty successful like first 10 months of being
18:29
open and like I can't tell you how many brand
18:32
owners I talk to that like have a pretty like poor
18:37
agencies purely from the fact they just say like
18:39
all they give is like ai slot like they can tell
18:42
the like scripts are written by ai slot so as
18:44
again like you know learning how to make sure like
18:48
you know you're gonna have to put in time manually
18:50
to learn how to be a great copywriter and stuff like
18:52
that you know I mean like that's all stuff yo
18:54
what's going up man what's up sorry I'm late
18:58
no problem by the way will we're live on x right
19:01
now so oh what's up people yeah so feel free to go
19:05
on rt me to get more viewers yeah it's no I mean
19:08
I I think with one thing that's interesting harry
19:11
um that you just said is you know I think that
19:15
there has been a lot of folks that they see volume
19:20
as the game and it's not about volume right
19:22
it's about the quality of the create of what you're
19:26
creating um and you know ai allows you to create a
19:30
ton of volume but that doesn't necessarily mean that
19:33
it's going to net you better results in any case
19:35
and I also think it's really challenging because
19:37
as has always been our industry there's a lot of
19:40
people that are in our industry that give advice
19:42
and that could be on well you know bidding or
19:46
audiences I mean that this is as tales old as time
19:48
like beauty and the beast but like the the thing
19:53
that I see with that that basically they're huge
19:57
accounts um and they are spending tens of millions
20:00
of dollars and they tell everybody what to do and
20:03
then these brands that are spending 50 to 50 to 100
20:06
thousand dollars a month which is a large majority
20:09
of e-commerce brands to be honest like they are
20:11
trying to implement a system where they're creating
20:14
400 net new concepts a week or something and you're
20:17
just like you know that's not going to be it right
20:19
and so it like lowers the bar enough and I think
20:22
that's not good for anyone really that's not good
20:26
for for for the you know any of it um no yeah that
20:31
was like that was like stuff I used to hate is that
20:33
you wake up to like some fucking x screenshot from
20:37
like some like of like your favorite gurus but oh man
20:41
like you know the accounts they're talking about are
20:43
way bigger than yours so you're in completely different
20:45
different situations but luckily that doesn't
20:47
doesn't happen to me anymore but yeah that like
20:49
back in the day that was a that was a painful one
20:51
because like yeah they would want the same thing
20:53
it's like I don't think you even cracked like what
20:56
your core messaging is yet or what actually makes
20:58
you different from other people right and I just
21:00
think you expanding it go ahead yeah and it's also
21:03
like I don't think you know the cpa that they
21:05
can afford to have compared to you so like you know
21:07
like it's a whole different conversation um but
21:11
yeah I mean like it's weird actually talking about
21:13
like the volume thing okay yeah obviously AI
21:15
is great for volume and like execution uh but I
21:18
actually saw a post about uh from a guy called guy
21:20
called Jacob uh a friend of mine you know he's he's
21:24
he's pretty active on Twitter and he's always been
21:26
like pretty anti like volume for the sake of volume
21:30
um and he made what I really liked was a great
21:32
post about the fact that he's actually in like the top
21:35
one percent of like advertisers with like the
21:38
amount of ads live and he was like that's not because I
21:41
like launch loads of volume it's just because I
21:43
make great ads and they just stay active for like
21:46
ever they've all got like a job to be done so I'm
21:48
like you know people are like oh I need to launch
21:50
like x amount of ads like all these ones uh but
21:53
it's actually you can just make bangers that just
21:55
have jobs to be done and just stay active for for
21:58
ages and you just compound that way instead which
22:00
is probably the way I would prefer to do it
22:03
personally um but again some people are going to be like no
22:06
that's stupid just make infinite ads um but also
22:10
a lot of those people who say make infinite ads
22:13
never really seem to supply results that come
22:16
with that so I'm not fucking know yeah well I'm not
22:20
trying to start drama but it's like you know you
22:22
always say make infinite ads but I'm like post some
22:24
post some fucking spend on those ads buddy come
22:26
on the so first of all the um for those of you that
22:30
tuning in uh this is a live stream episode that
22:32
shall be recorded as well of ai d2c wtf which is a new
22:36
podcast that will and I and our friend in Norway
22:39
uh Thomas Mullen have started uh and Harry Delmage
22:43
is on here talking about a whole bunch of stuff
22:44
so Harry thanks for being here um the so let me ask
22:49
about I'm gonna ask one more question and then
22:51
will you can kind of hop in so one of the things that
22:53
I think has been I was talking last night with uh
22:55
Hannah Haug and um Matthew Gatozzi about creative
23:00
in this whole ecosystem um and how I think that
23:05
there's this there's this really hard push and pull
23:08
between brands and agency operatives and or
23:12
operators and owners that want creative strategists to
23:17
be putting out more and volume is the name of the
23:19
game and the creative strategist that actually is
23:22
like I need to make sure that this is good and that
23:27
we're getting back to basics of creating great ads
23:30
how how do you see this right now and how do you
23:33
think that we as organizations can be better in terms
23:37
of uh helping our strategists write better briefs and deliver ultimately better ads
23:49
one of the biggest questions that comes up
23:51
inside the foxhole founders community is around hiring
23:54
how do you know and how do you find great people
23:56
that you can actually trust to do the job that's
23:58
why I want to talk to you about propel propel has
24:02
placed more than 75 employees for members of foxhole
24:04
founders some of the hardest roles our members have
24:07
been trying to fill for months they filled quickly
24:09
and efficiently we're talking from video
24:11
editors to graphic designers media buyers and more
24:14
in latin america to managers directors and heads
24:18
of department and vps in the u.s they have done
24:21
it all they can help you build that team that
24:23
scales faster I know that this is an ad read and I know
24:25
that it's totally one of those things you're
24:27
like well it's whatever it's an ad honestly though I
24:29
literally use this and have used propel and the
24:32
team over at propel for all of these things they've
24:35
filled offshore they've filled domestic like I
24:37
just said it's absolutely incredible and honestly the
24:39
founder Leandro used to run an econ brand and
24:42
performance marketing agency so he gets it that's
24:44
one of the things I've always loved about
24:45
Leandro and what I love about dealing with him I have
24:48
never even heard a negative review about working
24:50
with them one time somebody said it wasn't working
24:52
out and Leandro immediately replaced the candidate
24:54
that he had put in the position for a person that's
24:56
like the only feedback I've ever heard the way that
24:58
they work is simple it's performance based you only
25:00
pay a one-off fee once you actually hire someone
25:03
there's no retainers no ongoing costs and every hire is
25:05
backed by a three-month guarantee so go to the
25:08
show notes check it out let me know what you think
25:10
and if you have questions about this feel free to
25:12
hit me up andrew at foxwelldigital.com I'm happy
25:15
to talk about it if you go to the link in the
25:16
show notes you get 10% off your first hire with them
25:19
it's a great deal really special for podcast
25:21
listeners so I look forward to hopefully helping
25:23
your brand grow and scale
25:29
is that question for me or for Will yeah for you
25:32
damn I thought you were talking to Will that whole
25:34
time I'm not gonna lie to you I feel I thought
25:36
you said Will at the start so I was like okay he's
25:38
asking that but okay no I'm Will can answer that
25:41
and then you can chime in if you'd like I mean I'm
25:43
I'm with you I know I have to answer but sorry I
25:45
was just like waiting for Will to talk because I was
25:47
like I don't want to just talk over but no I can I
25:51
can answer yeah I mean like I think it comes back
25:55
to the thing that so much has rapidly changed
25:57
over three years like I was saying before like three
25:59
years ago I would have had like no way of
26:02
knowing that I'd be able to create all this like wild
26:04
Pixar or Playmation looking stuff that like
26:07
genuinely does well like that just just wasn't
26:10
something that I thought when I was like you know
26:12
three four years ago and I was like new to create
26:13
a strategy that was like okay this is a possible
26:15
thing to to do um so I think it's like a natural
26:20
pendulum swing right it's like you just like oh
26:21
my god this is so cool I can just like get every
26:23
single one of my ideas down and I can just like
26:25
get everything out and now it's coming back to that
26:28
thing where it's like oh shit I need to actually
26:29
make sure this is grounded in like proper direct
26:32
response proper marketing I need to QA this all
26:35
properly because like you know depending on the day
26:38
it will either give me something that's like
26:39
actually pretty good or just it doesn't even make sense
26:43
I think it's just a natural like pen and string
26:45
back where like people thought they could just
26:47
like one shot or was hoping they could one shot
26:49
a lot of stuff with AI and it'd be a banger like
26:52
that would be the best situation if I just have
26:54
like a random idea and I'm like okay go write a
26:56
yeah right about that and it spends like seven
26:57
figures that's that would be the best thing ever
26:59
like I'm not saying I don't want that I'm just
27:02
saying it's like realistically it's not there so I think
27:04
people are like now coming back to being like
27:06
okay no we actually need to go back to fundamentals
27:09
because like anything right it's like whatever
27:12
sport you do if you stop doing the fundamentals
27:15
and the basics like you will lose that skill you
27:17
will forget how to QA a script you'll forget how
27:20
if an ad is even good or not like you just won't
27:22
know you will look at it and be like yeah that's
27:24
fine but that's just because you've been like using
27:26
AI for every day for the last six months and you just
27:29
have no idea if that's got better or worse
27:31
because it's just your new standard so I think it's just
27:33
naturally naturally coming back to that and I also
27:35
think like brand owners especially I mean you know
27:39
again I'm not speaking for all people here just
27:42
like you know they're they're coming to me and
27:44
like the agency for quality like that that's what
27:47
they want some people like are like oh yeah like
27:49
we want some more volume but like they only want
27:50
more volume once the quality is proven you know I
27:52
mean that's when they're like hey guys can you
27:54
give us more like it's not like you know I feel like
27:57
if I told people what their like monthly like
28:00
deliverables were like a year ago people were like
28:03
that's nowhere near enough how do you expect us to
28:04
scale off that and I just don't think people got the
28:06
results they wanted to were like oh shit okay
28:08
maybe we got to go back to basics here so that's that's
28:12
my view on it yeah I definitely agree I think you
28:16
know we have been actively hiring creative strategists
28:19
and I think one of the most eye-opening
28:22
experiences for me is that like a lot of creative strategists
28:24
will just continually write copy with Claude and
28:28
like in the application process you'll be getting a lot
28:30
of responses with Claude and if you ask to
28:32
brief out a handful of statics and videos you'll get
28:34
Claude responses and you know I'm obviously the
28:37
first to admit that like we are like sort of like
28:39
an AI enabled agency but the reason we are AI
28:42
enabled is so we can free up strategists to actually do
28:46
the hard work right like we want to be able to
28:49
produce 80% of your ads bottom of funnel statics
28:52
with AI we still have editors on all of our all
28:55
of our statics but then that other 20% needs to be
28:58
like entirely human driven like copywriting is
29:01
undoubtedly the most important part of creative
29:03
strategy in my opinion and you can't just like
29:05
skip that step and I think we're just like in this
29:09
still in the stage of like AI psychosis where
29:11
it's just like it's a really fun tool to play with
29:13
and I again I'm the first to sort of fall into
29:16
that grouping but like where you ultimately end up
29:20
winning is sort of utilizing AI to uncover like
29:25
angles personas emotions formats you should be testing
29:30
giving the reins to like a really good
29:32
copywriter uh who can actually deliver um and so yeah it's
29:38
definitely a combination of the two and like
29:40
finding good strategists that actually have copywriting
29:44
jobs that are actually willing to sort of do the
29:48
hard work is becoming much more challenging it's just
29:51
I think people just assume that they can just uh
29:54
skip steps by using uh you know language learning model
29:59
so yeah I completely agree with Harry there um
30:01
I guess my question to that is like when you're
30:05
looking for a strategist now Harry like what is
30:08
like if you had to sort of distill the the the the job
30:11
down into like one most important quality of a
30:15
person like what do you think that is one that's tough
30:19
one important man okay I guess just willing to
30:25
do the hard work like that's just just it like
30:29
you know because at first I was like looking
30:31
for people who uh okay how how skilled are they
30:36
where have they come from like you know I want
30:39
someone that's got like quite a bit of experience
30:40
but like I'm also finding that to be somewhat
30:42
sometimes again not always like a bit of a negative
30:45
now because I've now got to like make them
30:48
unlearn all the bad habits they've picked up
30:51
whereas like actually uh you know onboarded
30:54
some like really like hungry people who were like
30:57
not green by any means like at all but like not
31:00
these people have come from these big places or like
31:03
you know had this like you know years of experience
31:06
and they're like fucking crushing but just because
31:08
they're hungry to learn they want like actually
31:09
want to learn the craft they don't think they know
31:12
the craft you know I mean whereas a lot of people
31:13
are like oh yeah well I've done this for x amount
31:15
of years now I I can just keep cranking out I
31:18
know everything I already need to know and if like
31:20
they've given up learning still you know I mean
31:22
like they're not going to go read like breakthrough
31:25
advertising for the sixth time do you know I mean
31:28
they're just not going to do that because in their
31:29
head they've like completed it so I think just
31:31
like willing to do that hard work and go that extra 10
31:34
percent is like are you willing to like QA your
31:38
scripts that bit more or are you just going to
31:40
one-shot it with AI and it's like yeah that's
31:42
good enough um so yeah probably that's the one
31:46
thing I'm looking for but it always changes you
31:48
know like first I was like thinking this is the
31:50
case but then in three months I might find a new
31:53
type of person I'm trying to hide but like yeah
31:55
hungry people willing to do the work and like
31:58
generally still curious and like wanting excellence
32:01
whereas like I feel like that's died a bit not
32:03
saying it's because of the use of AI but like
32:06
it seems like a lot of people have just become
32:08
happy with good enough like even like you know
32:11
as for example when like I like I always do
32:13
trial and I always make someone write a yapper at for
32:15
trial like even if it's like they're not very
32:17
comfortable with because it takes time it's longer
32:18
like and we had someone um send one over the other
32:23
day like instantly just replied and I was like write
32:25
that again like you know and just made them do it
32:27
again make them go longer because it's like no I
32:29
don't want it just like oh with AI like just go
32:32
back and do it again yeah I mean I think I think
32:34
that that's I think it's true it's like that that
32:37
desire of somebody who wants to and knows that it
32:40
doesn't it it's um it's going to take time right
32:43
um and also knows that there's always this constant
32:47
desire to continue learning and to continue
32:49
refining right like um if you asked I don't know Michael
32:55
Jordan at the height of his career what's
32:57
the number one thing you would say to somebody
33:00
you know to continue to learn I mean I think or
33:02
continue to grow like it had to be something like
33:05
you have to keep training you have to keep
33:06
developing this to developing it which I think is is the same
33:10
because I think it's it's just very easy to fall
33:12
into that the AI trap of of it's doing it for you and
33:16
you think that you're improving but in reality you're
33:18
really not so I think that that that's is like a trap
33:22
that a lot of folks are caught in and they're out
33:24
there seeing like oh I hooked up all these things
33:27
and I have all this you know data and all these
33:29
MCPs um and I'm pumping out more ads than I ever have
33:33
but it's it's like it just loses that human element
33:36
and it loses the hard work that it takes to actually
33:39
copyright really probably um and I think that
33:43
creative strategists know this I guess is what I'm
33:45
getting at and I think a lot of leaders don't
33:47
know this as well right so I think this is a call to
33:51
to leaders to be like hey this is different
33:54
right like we need to not just be pushing on
33:57
the volume button or just generally saying
34:00
personas right but like what does that even mean
34:03
and what are you trying to do I guess a question
34:06
I have for you Harry is you know let's say that we
34:08
look a year down the road too I'm you know I'm
34:11
always in the interest of equipping people properly
34:15
we go a year down the road what are folks doing
34:19
that uh you know has has improved their ads with
34:24
AI like is it just that people are utilizing AI in
34:27
a year more efficiently or that they've integrated
34:30
into their systems better or you know what is it
34:34
that it's that they've done with AI that's made
34:38
their lives easier and better I think yeah
34:39
there's like a case for efficiency like you know again
34:42
go back to the review money but I think the people
34:44
that use it the best and the people that can like
34:46
really use it to bring their imagination to life
34:50
you know what I mean like I think that's the case now
34:52
and I think it will be the case like even more so
34:55
as like you know it gets easier and easier to like
34:57
create these wild things but like some of the AI
34:59
stuff that I see working like you know people would
35:02
do like fake like security camera stuff like you
35:07
know as like a hook or like you know again we talk
35:10
about like that like the pics are like musical
35:12
stuff like anything like that where you can use it
35:16
because before it was such a constraint and like
35:18
now it's not and I think it's only going to get
35:19
easier to bring your imagination to life but I
35:22
think to do that you need to still keep that like
35:25
creativity alive and like I think a lot of that
35:28
comes from like manual stuff or like experiencing
35:33
things and trying to like then turn that into to
35:36
adverts but like yeah obviously things will get more
35:38
efficient reporting like all that stuff like the
35:40
grunt stuff is gonna um basically make it easier
35:45
and then like the things that you'll probably do
35:47
is be like okay well I'm gonna go out for like two
35:49
walks today on those two walks I'm gonna try like
35:51
ideate two fucking crazy ads or something and I'm
35:54
gonna wait for like I'm gonna be ruminating over
35:57
everything I've been thinking about and then wait for
35:59
one to be like airdropped to me or something
36:01
like so I think yeah it's just gonna become like you
36:04
know the people that are crushing it the people
36:06
that are still like you know keeping their like
36:08
imagination and creativity cranking and then they
36:11
can just like use AI to come up with this like mad
36:14
shit that looks awesome you know what I mean yeah
36:19
100% and like sort of piggybacking out of that I
36:22
think a lot of people are just like using AI wrong
36:25
right like AI has opened up this like so many avenues
36:31
of content right like as you were discussing
36:33
earlier like you can do a full sort of end-to-end
36:36
top of funnel animated you know claymation you
36:39
know Minecraft whatever sort of medium you want
36:42
it to be but like fundamentally you really need a
36:44
strong script there like advertising hasn't changed
36:47
at the end of the day we're all just telling a
36:49
story right and now we just like have additional
36:51
mediums to do it so like I think where people get
36:54
what gets so wrong with AI in general it's like oh
36:57
this format is crushing right now it's not the
36:58
format that's crushing it's the story that you're
37:02
telling behind the format and like again that's
37:04
why people have this like sort of AI psych psychosis
37:06
where they're just like building fucking
37:08
dashboards and just like you know spinning their wheels and
37:11
like oh yo look at my fucking sick dashboard and
37:14
it's just like it's so infuriating so it's like not
37:16
accretive that's not ROI positive like what's
37:19
ROI positive is like using these tools to like
37:22
underpin like where are my holes in my uh in the
37:26
ads I'm running now like who who haven't I been
37:28
talking to and then like using Harry's method
37:30
where it's just like I'm actually writing end-to-end
37:33
scripts you know I can use a creator to do a
37:35
yapper ad or like if I want to even test it like I can
37:38
spin up like an animated ad top of funnel ad in an
37:42
hour right it's just like it's so people are just
37:45
like are so of the mindset that like everything that
37:48
they're creating with cloud code is is god's gift to
37:50
the earth and like again this is sort of this
37:53
psychosis mindset and it's just like you have to
37:56
sort of go down to fundamentals like a good script
37:58
is a good script and like nine times out of ten it's
38:01
written by a human being but like then you can
38:04
take that script like do a minecraft one and if that
38:06
works well like then get a creator to go do it
38:08
it's just like it gives you the ability to sort of
38:10
catalyze testing in a way we've never been able
38:12
to do so before um and and Harry I'm curious like
38:17
when before you like go and create a yapper ad
38:20
like are you testing that concept with like an
38:22
animated sort of top of funnel uh no okay so
38:26
you're going straight to creator straight from script to
38:28
creator yeah I think so just because like
38:32
honestly like a lot of people can kind of do that like
38:34
animation stuff pretty pretty well I feel like um
38:39
because I think like a lot of the visuals can do
38:41
the carrying there you know I mean like because
38:43
they look so like you can tell so much of that story
38:47
with like you know it's a bit of like a cheat
38:49
code for that kind of stuff um but like still a
38:53
lot of accounts are in like you know haven't quite
38:55
cracked yappers and like we're pretty confident with
38:58
how we write it and how we QA it so we can just
38:59
send it straight to a creator and just be like fuck
39:01
it let's let's let's run it um because it's still
39:05
like a thing that I think people struggle with so it's
39:07
just like well while people are still struggling
39:09
why not just just go for that uh but then if it does
39:14
turn it into like AI stuff and just like you know
39:16
spin that concept and that story uh as many ways as
39:18
possible um you know for sometimes for example
39:22
though like on certain accounts we would do like
39:24
you know the long form native image copy stuff
39:27
if that pops off we'll take like the core story of
39:29
that and then apply that to a yapper because it's
39:31
just like a cheap and easy way to test that story
39:33
before you then uh go to that uh just like full
39:37
warning for anyone that does that don't just like
39:39
copy and paste the thing just because your yapper
39:41
script will turn into like a 12 minute long thing
39:43
and people can read way quicker than they can
39:45
consume information and getting a real person with
39:48
like not any b-roll or anything like that being
39:50
charismatic for 12 minutes straight is uh pretty
39:54
pretty difficult I wouldn't I wouldn't necessarily
39:56
recommend it so yeah like that's like a really good
39:59
way to test it but then it's more like once we
40:01
have done that um you know I guess not that I ever
40:04
think about like do the hard thing first then we can
40:06
like do the do the easier stuff and just spin that up
40:08
like different ways um and then there's just
40:11
some stories that like again like I've spoken about
40:14
it's like you know if you're in a pet like pet
40:16
niche I would just spend more of my time doing
40:19
like AI animation stuff because you can make
40:21
dogs like talk and shit like that which is like so
40:24
awesome do you know I mean like that's so cool
40:26
you can like yeah you know I think we did one for a
40:28
brand it was like you know it was like for like
40:30
some joint shoes or whatever uh and like you know we
40:33
show the claymation where like the mum gets like
40:35
really upset when like the dog starts limping because
40:38
like got old and the dog's like I don't
40:40
understand why mommy's sad and all this stuff like I'm just
40:42
trying to be like a good dog uh so it's like you
40:45
know you just can do that with like you know like
40:47
loads of people are commenting like this made me
40:48
tear up like this like you know and it got like a
40:51
pretty pretty healthy amount of spend but it's
40:52
like you just can't do that with like yeah like the
40:55
apparatus like you know I mean so it's like that
40:57
kind of stuff we just don't necessarily bother with
40:59
the apparatus and we just go straight to AI
41:00
animation yeah but like fundamentally you still have a
41:04
human being writing that script right like that
41:06
is like the research needs to still come into play
41:08
like it's not the animation that's carrying it
41:11
it's it's it's a good script presumably yeah exactly
41:13
I think like more what we do is like we're really
41:15
like heavy on human QA because like I like I was
41:19
saying like earlier like you know one of the great
41:21
things with AI is like especially of skills is it
41:23
like really lowers that barrier to like execution
41:26
where you can actually get like and with skills you
41:28
can get a pretty decent first draft but I think
41:30
that's the problem is people look at that like oh
41:32
that's pretty decent and then just like leave
41:33
it there but it's actually the ads that like even
41:36
when we've curated and I spoke to um like spoke to
41:39
my friend like Luke the other day he's like head of
41:41
performance like another and he said like the ads
41:44
that he sees pop-offs are the ones that where you
41:46
add the tiny details to it you make the small
41:48
changes because like the nuances really matter so it's
41:50
like for execution like and I fully agree with
41:52
that I see the same things like even if it's not big
41:55
changes you're just asking like so what can we
41:57
add a little bit more detail here can we make that
41:59
more emotive and like you need to in my opinion
42:01
you need the human human QA for that um but yeah
42:05
I think most people get to that point where it's
42:06
like okay cool that's like a six out of ten we'll
42:08
just ship it but if you can get to that six out
42:10
of ten and then make it like nine out of ten like
42:12
great that's like kind of like the way you wanted
42:14
to do it because yeah starting from zero and being
42:17
like fuck now I gotta write like a four minute
42:18
script like while you've got all this other stuff going
42:21
on it's like it's not a fun feeling so it's like
42:23
if you can get that first draft great then cool but
42:25
you but that's why you need to still be studying
42:27
still be breaking down winning ads manually like
42:30
you know like I got like a bloody like this here
42:32
like I got the I got the prompts that I do to run
42:35
like winning ads but I also watch winning ads
42:37
manually and like write down notes myself so you
42:39
can still just know what makes a good ad like I go
42:42
on Kalo data and I write down what the affiliates
42:45
are doing when they're like overcoming objectives
42:47
like I'm still doing that manually so that I can do it
42:49
when I get my script back and I've noticed when I
42:51
don't do that my QAing gets worse because you just
42:55
over-reliant on AI and you're just starting to
42:57
accept what it's giving you because you're like
43:00
ah fuck it it's been fine but a lot of the time
43:03
it's not and then as it keeps going it's worse and
43:06
worse you know what I mean totally it's such a
43:09
negative feedback loop and like I definitely agree
43:12
like journaling every morning like sort of keeps
43:14
me humble and like you know puts puts the earth
43:17
underneath my feet like so for your team like are
43:20
there any things like you recommend for your team
43:23
to members or like anyone else in this space
43:25
like how do you sort of keep that grounding and not
43:28
sort of lose yourself in the psychosis of it all
43:31
yeah I mean I just like ask for for proof around
43:35
stuff I'm just like where did you get that from and
43:37
if like sometimes they're like AI I'm like okay go
43:40
and verify it yourself but like also what we do is
43:43
we do like a manual research challenge every month
43:46
where there's like bounties and it's like if
43:48
you get the biggest unlock from research uh manual
43:51
research um then you get you get a bounty
43:55
like you know I think like prizes go up to like
43:58
like right now like five hundred dollars on top of
44:00
like the commission they already get anyway so it's
44:02
like you know it's all right it's not like it
44:04
probably could be better and then also what I'm
44:06
going to implement alongside that is if you write
44:08
a manual ad that pops off alongside the commission
44:12
you'll also get like a bounty to do with that
44:15
as well so like really just trying to get them to
44:17
get into the habit of like keeping the tools tool
44:19
sharp um and you know I'd love for people to do it
44:23
just because they think I'm like awesome and it's
44:25
the best company to work at ever but like you know
44:28
like money talks you know I mean so like just
44:30
giving people like financial ascent like incentives
44:33
especially when it's like something that's like
44:35
becoming harder to do because it wasn't we never like
44:38
we had mental friction to do it before but we had
44:41
to do it you know I mean there was like well I have
44:43
to just write this manually like yeah it kind of
44:45
is like annoying that I have to do it but like I'm
44:47
just gonna fucking sit down and get it done like
44:49
that doesn't exist anymore you can just get AI to
44:51
do it so it's even more mental things about oh
44:53
fuck's sake I really have to write this manually it's
44:54
just like okay cool I just I I gotta give you money
44:57
for that effort you know what I mean yeah I mean I
45:01
think that that raises a good point which is one
45:02
I've said a lot uh in coaching calls other things which
45:06
is with employees making sure that you are
45:09
incentivizing them for the behavior that you
45:12
want it's just like anything else but I mean like
45:14
seriously it matters when you go through and say
45:17
this is the type of behavior we're trying to reward
45:20
that's not done enough people do that in they think
45:23
it's under job expectation um which in some cases
45:26
it is but if you are doing competition or you're
45:30
gamifying it or you're you know doing some sort of
45:33
bonus they'd say hey look like the more I see this
45:36
better I am happy to incentivize you um and
45:39
here's what it means like you will get x number of
45:41
dollars like that will drive better results so I
45:45
think that's kind of an interesting one um as you
45:47
go forth one question I have Harry is on on those
45:53
people that are utilizing you know AI in a way that
46:00
makes sense uh in their organization do you think
46:06
that they have is there some creative director or
46:11
brand the brand owner has a good idea of what
46:14
they are okay with with the brand in reference to AI
46:19
like there's or is there like an AI captain in the
46:21
organization or are they just saying you know what
46:25
Harry we trust you you're the expert go for like
46:28
which I just am wondering where people have been
46:33
held back right are you talking about like in
46:36
terms of like like you know I'll like the my company or
46:41
terms of like like when I work with brands like what
46:44
they're saying about the use of AI yeah I'm wondering
46:47
specifically about the people that are um how do
46:53
I say this I'm wondering specifically about those
46:55
organizations that are really good at AI and they're
46:57
like really good at making your job easy in terms of
47:00
getting out better ads do do they have some sort
47:03
of filter that says yes that can happen with that
47:06
that's fine that's on brand for us um go for it and
47:09
they trust you or it's like they have a process of
47:12
okay this goes to this person they approve it it's
47:14
good um like and there's almost like an AI captain
47:17
internally I'm just curious of your take uh well
47:21
yes okay if I've got it confused let me know but in
47:24
terms of like you know when people I've like
47:27
listened to that again I was like this could be a number of
47:29
things but it's been a long day a lot of calls a
47:32
lot of calls you gotta bear with me um you know it's
47:35
a little bit late for me over here um yeah so in
47:38
terms of like when I'm like sending of if you're talking
47:41
about when I'm sending over ads to clients and like
47:43
how they're talking about like are things on brand
47:45
and like with the use of AI I'll be very honest
47:48
like we mostly work with people that are pretty chill
47:52
with a lot of stuff um you know we work with like
47:56
a lot of like supplement brands and like that kind
48:00
of thing um so I would say like and a lot of people I
48:04
think would that come to me know that like we're very
48:06
rooted in like direct response and like we like to
48:09
push it a little bit um but yeah we've definitely
48:13
had it before where people are like that's not
48:15
on brand and like to be honest if it's like stupid
48:18
stuff that I don't agree with then like we it's
48:20
just not like I don't want to say stupid but like
48:23
just stuff that isn't aligned with me then like
48:25
the relationship's just not going to last very very
48:27
long like we do best with people that are kind of
48:30
just like yeah like fuck it you guys can kind of
48:32
and like run whatever you want essentially like
48:35
whatever you think is best because at the end of
48:37
the day they have a freedom to it they're just
48:39
saying that's fine I'm just wondering because I
48:41
feel like there's a big thing that happens with AI
48:44
people are either like there's a lot of places it
48:46
has to get routed and that hurts them from
48:48
creating better ads and there's other places that are like
48:50
the extreme of your side which is like go for it
48:52
whatever we don't care right and so I'm trying to
48:55
figure out like there there probably needs to be
48:58
a middle ground of this of like getting things out
49:01
to because there's going to be more governance
49:03
around this so that's what I'm thinking yeah yeah
49:06
for sure like yeah I guess because I mean like with
49:09
some brands like again with like some with fashion
49:11
like makes sense like you know you don't like you
49:14
know you don't be doing necessarily like pixels and
49:16
it might work I don't know but like sometimes you
49:18
see a bit more branded stuff with like that but I
49:20
guess it kind of makes sense because fashion it
49:23
makes sense you know I mean but like in terms of other
49:26
stuff like I've been speaking to people like maybe at
49:29
like bigger brands and they just can't even get like
49:31
AI animation stuff approved by brand and I'm just
49:33
like that's just the fucking death of advertising
49:35
like what the fuck are we doing here like this is
49:37
crazy like yeah you know like I get on the other
49:40
side yeah and I get on the other side where people
49:42
are like doing like fake doctors and stuff like you
49:45
know it's a bit like whatever like you know
49:46
people have like you know views on that so it's like
49:49
you're right there definitely probably like is a
49:50
middle ground but like I'm always on the thing of
49:53
like lean on the like side of like a bit more
49:57
aggressive you know I mean I think yeah you know you look at
49:59
brands like crushing now and a lot of them that
50:03
are up and coming again can't talk full of them
50:06
because I don't see inside their ad account but
50:07
like they are doing these like you know being more
50:09
loose of it like because it's all like
50:11
trend-based as well like you know there's only so long that
50:14
these ads will like kind of work like claymation
50:16
and stuff like that is like still work because in
50:18
some cases falling off a bit so they're just like
50:20
missing trends because they've got some people in
50:23
brand who don't even know what direct response is
50:25
being like oh we can't run that like well it's stupid
50:28
it's obviously it's advertising it's so well
50:31
and that's that's so funny I mean I I have this
50:34
I we should name this like to come up with a
50:36
law for the name the the the something law of like
50:39
the bigger the organization the worse they are to
50:42
work with usually and the and the more they shoot
50:44
themselves in the foot like I'm gonna say nothing
50:48
because there's so many layers you know you're like
50:50
oh my gosh um yeah well Harry thanks for
50:54
joining us thanks for joining the live stream everyone
50:56
um if you came on live stream and uh we'll look
50:59
forward to uh doing this again but Harry thanks
51:01
for for coming on today yeah no all good I have
51:04
no idea if I said anything helpful at all zero you
51:08
didn't say anything good it's terrible yeah yeah no
51:11
I just just complained about volume and big brands
51:15
that was it that's it yeah people people got
51:19
their their times worth with that one let's let's go
51:22
you