YT YouTube2Text

YouTube2Text → Transcripts → Nos STRATÉGIES d'ADS chez Fincut : Podcast avec Hamdi et Erwan (CEO - Agence OhMyGads)

Nos STRATÉGIES d'ADS chez Fincut : Podcast avec Hamdi et Erwan (CEO - Agence OhMyGads) — Transcript

by Alex Shane · 17,722 words · 2,755 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00Actually, for me, a brand means you're
0:02able to sell without running ads. It's
0:04an identity that you manage to
0:05broadcast repeatedly without stopping.
0:08For a beginner, I think AI is one of
0:10the worst things that could have hit
0:11the market.
0:12>> But I admit, giving up, it did cross my
0:14mind, you know. After a while, you feel
0:15like you're giving it your all. It's
0:16never enough.
0:17>> We are not at the service of AI. AI is
0:19at our service. That's something you
0:21have to understand and really
0:22internalize because if you let it do
0:23everything for you, you're done.
0:27>> Ladies and gentlemen, hello. I hope
0:29you're doing well. We're here today for
0:31a great podcast with two exceptional
0:33guests. Amdi, who you already know very
0:37well, CMO of Fincut, wears many hats, a
0:39very, very strong person in their field
0:41and it's a real pleasure to be
0:43surrounded by this person. But also
0:46someone you've never seen on the
0:48channel before, maybe a small
0:49appearance in a recent vlog, who is the
0:50founder of Oh My God, a creative agency
0:52we've been working with for a few
0:53months now, about how long? Since April
0:56, you know.
0:57>> Since April.
0:58>> Uh, it's a creative agency we're very,
1:00very happy to work with. It's going
1:02great. So, nice to meet you, Herwan,
1:05could you introduce yourself quickly,
1:06please?
1:07>> Yeah, thanks for, thanks for having me
1:08on the podcast, it's a pleasure. So,
1:10Herwan, it's been a little less than a
1:12year now that we founded Oh My God, so
1:13you are among our first clients. Thank
1:15you for trusting us, and so our goal,
1:16really, is to become the benchmark in
1:18e-commerce. And for that, we're moving
1:20a little cautiously because there are
1:21reputation stakes involved, and we want
1:23to create things that are beautiful but
1:25also high-performing at the same time.
1:26So, that's the goal. There you go. Yes.
1:28>> And for those who don't know you, can
1:30you introduce yourself quickly?
1:32>> Yeah, I'm Amdi. I am the creative and
1:34marketing director at Fincut. Hm.
1:37>> Hm. Uh, and as you say, I wear many
1:40hats, I take care of everything,
1:42management, the strategy to implement,
1:44checking that all our providers are up
1:47to par, uh, defining the upcoming
1:49collection, all that stuff.
1:51>> Really a very, very big role at Fincut.
1:54And what's cool about having the two of
1:57you at the table is that, honestly, the
1:59two people who were exchanging the most
2:01, really, it's clearly you two; you met
2:03, and you kind of grew together over at
2:05Omegads. That's that's very cool.
2:07Actually, it's good that you're between
2:09us both because Amy is kind of my
2:10sparring partner, you see. Meaning that
2:12>> I know that that's a lot. Amy is a very
2:14, very demanding partner.
2:16>> I saw I saw I saw the the posts on
2:17LinkedIn, you know.
2:18>> Yeah, that's exactly the term I used,
2:20actually. He challenges us, so it's
2:22good. So for you, it's a bit like a
2:23sparring partner. You always have to
2:24raise your game to be at the level of
2:25the opponent. But are you happy? Yeah,
2:26it made us improve. That's what we're
2:28looking for.
2:28>> No, honestly, that's what's great with
2:30Amy, is that at the beginning it's a
2:31bit complicated to manage because
2:33you're very demanding and so I was
2:34getting a lot of backend messages from
2:35quite a few partners. But at the
2:38beginning when I had a bit of trouble
2:40with that, it turned out that when I
2:42really gave 100%of my trust to Amy,
2:44actually trusting him with that. We saw
2:46that the stats went up, that the
2:47quality is much better, and so on. So,
2:49it's true that being demanding at first
2:50glance can be scary. But it's true that
2:52often agencies, in fact, they will
2:54settle for the minimum and charge the
2:56maximum, no matter the agency, right,
2:58no matter if it's creative, email, etc.
3:00But they will settle for doing the
3:01minimum and charging the maximum to
3:03make maximum profit. And having a
3:05profile, like Amy, who really regulates
3:06, for that matter, well, it's both
3:07great for the brand, but also for the
3:09agency which will be able to evolve and
3:10head in the right direction because of
3:12it.
3:12>> Yeah, completely. I mean, there's a
3:13side where you actually remind everyone
3:15of the mission. Yeah.
3:16>> Like, I I I pay for a service, I it's
3:19it's like at a restaurant. If I pay, I
3:21don't know, 100 € for a dish, I don't
3:24I don't want it to be cold. I didn't
3:26pay for 80%.
3:27>> You see? Yeah. Yeah. That's what that's
3:29what you told us when we when we saw
3:31each other, by the way; we've already
3:32met with Amy in the past. But indeed,
3:34it raises the bar, and at every level,
3:36even when you think you're good enough,
3:38you know, because we released creatives
3:39with drones, we shot beautiful golf
3:41footage, etc., we send it off, we're so
3:43proud. He tells us: "Guys, the color
3:45grading..." and actually, it’s silly,
3:46but it’s that kind of detail that
3:47makes you learn to pay attention
3:49everywhere. And so, it really shook
3:50things up for us internally, because
3:52suddenly, you're forced to bring
3:53everyone to a high level in your
3:54production value chain: everyone has to
3:55improve a little. Except today, well,
3:56we've developed this level of demand we
3:58didn't have before, because you always
4:00have to keep questioning yourself. But
4:01now, it's something we can offer to all
4:03brands, and we can say it's partly
4:05thanks to that; it also comes from our
4:07humility to accept feedback. For me,
4:09that’s what I liked about you: when
4:10many people would have gotten
4:12frustrated, you decided (well, it must
4:13have been a bit frustrating at first,
4:15too), but you decided to take the
4:17knowledge from it rather than digging
4:18your heels in and saying, "Yeah, no,
4:20actually, I'd rather not," but you
4:21really wanted to grow with it. In fact,
4:23now we have creatives that are
4:24incredible and are constantly getting
4:26better. Actually, it's so cool if you
4:27can share good creatives with your
4:29clients in the market, in the French
4:30market, and so on. It’s top-tier,
4:32clearly. That’s clear. But I admit,
4:34giving up did cross my mind. You know,
4:36after a while, you feel like you're
4:37giving your best. It's never enough.
4:38But today, you see that all the service
4:40providers are benefiting from it, and
4:42everyone is saying, like, OK, My God is
4:44starting to take its place. It's
4:46>> the little troublemaker that comes in
4:47to shake up the market a bit, and in
4:49the end, everyone, everyone, everyone
4:51wins, at any rate.
4:52>> Exactly.
4:52>> In any case, what I found cool was that
4:54you managed to take the feedback into
4:56account. I mean, I received your
4:57messages where you said that as soon as
4:59I get a WhatsApp from Amedi, my heart
5:01races. But on the other hand, you know
5:04that now 4 minutes
5:06>> you can you can I mean you can call me
5:07anytime and you know I can give advice
5:09and you see the relationship is no
5:10longer the same because actually
5:11there's a sense of trust that builds
5:13and I know that from the moment you
5:14take that into account
5:15>> well for me it's fine you see
5:17>> absolutely.
5:18>> Gentlemen, I have a good a first
5:19question a first question to ask you.
5:21In your opinion, what is a good
5:22creative?
5:23>> Start.
5:24>> Go ahead go ahead please. For me, a
5:26good creative is a creative that we
5:29don't I mean that we don't identify as
5:31being an ad.
5:34>> Which means that for me the hook, you
5:36have to be because today we are so
5:37solicited that actually you must have
5:39that same hook and and then you have to
5:42say I've been had, it was an ad, now
5:44I'm intrigued and I'm going to go check
5:46it out. For me, it's it's basically
5:48that.
5:50>> very very native.
5:51>> Yeah. It shouldn't be too polished, not
5:53too pretty. Hm
5:54>> hm. Uh it shouldn't be like cinema.
5:57>> Yeah, I agree.
5:58>> There's a balance between the two. Me
6:00>> you know, I'm a I'm a teacher's pet. I
6:02read books on advertising and if you
6:04listen to the parents of advertising
6:05like David Ogilvy etcetera, they will
6:07tell you the only goal of an ad, so a
6:08creative and an ad, is to sell.
6:10>> Everything else, that doesn't count.
6:12>> Now, I would still nuance this
6:13discourse a little bit because as AMD
6:15says, you try to fly under people's
6:16radar, be careful, but ultimately it's
6:18to sell. So in the end, we agree on
6:20that definition. But I am evolving a
6:22little bit with precisely the more
6:23polished creatives that we are
6:24developing for you today where I don't
6:26necessarily have the objective of
6:27selling immediately but of making an
6:28impression so that other creatives come
6:30to sell. So short short term it's to
6:33sell. Longer ah it depends.
6:35>> Yeah. I'm a bit in agreement with both
6:37of you in the sense that actually
6:38everything depends on the uh level of
6:39awareness of the client. If you want to
6:41do bottom or or top
6:43>> is that if you want to do bottom the ad
6:45shouldn't necessarily look like an ad
6:46or for middle something fully native.
6:48You get brained and you're like what
6:49the? Actually, it said sponsored
6:51at the bottom, whether it's middle or
6:53bottom. After bottom, you can be more
6:55aggressive, so you can pass more for an
6:56ad. But it's true that at the top of
6:59the funnel, you can afford to do
7:01concepts that are a bit more polished
7:03which, I agree, shouldn't be perceived
7:05as an ad, but rather, you know, almost
7:07like a...well, I don't know, a little
7:09moment you spend with the brand, a nice
7:12little thing, and so on, but not
7:14full-on. You really get the feeling
7:16you're on TV or something like that.
7:18But it's true that mixing the two is
7:20quite, quite important.
7:21>> Exactly. And as an agency, you know, we
7:23also question ourselves on what we do
7:24differently from others, what can we
7:26offer? And, well, precisely, we try to
7:28have these two extremes. Creatives made
7:29to sell which, perhaps, are less
7:31polished, less refined, and that way,
7:32you fly under people's ad radars. So
7:34that people associate the brand,
7:35discover it, and think: "Wow, this
7:37creative really hit me, hey, that's
7:38cute." And in fact, by constantly
7:41churning out that kind of creative, you
7:42build your brand image, and behind that
7:44, you have your entire armada of sales
7:45creatives so you don't sacrifice your
7:47profitability.
7:48>> In any case, if I had to start tomorrow
7:50, I would take that strategy.
7:53>> Full native.
7:54>> Yeah, full. I mean, an ad that doesn't
7:55look like an ad.
7:56>> Even at the top of the funnel, what
7:58kind of format would you do, for
7:59example?
8:00>> Something super educational, you know.
8:01You see.
8:02>> Why, uh, why I had this problem, and in
8:04fact, how I solved it, and everything
8:06to eventually arrive at a product.
8:09>> It's true that right now in the US,
8:10there are plenty of "what the" ad
8:12concepts starting to work.
8:14>> Whether it's music—the guys, it's
8:16just a 3-minute song, and in the song,
8:18it sells you the product with a
8:21>> big story behind it. They spend a lot
8:23of time developing the problem first,
8:24by the way.
8:25>> With a—Yeah, exactly. With a nice
8:26animation behind it, where you actually
8:28see the person experiencing the scene,
8:30living the music, and so on. VSL videos
8:32of 5 to 6 or 7 minutes to take you
8:34through a whole journey before you even
8:35start seeing products.
8:37>> And honestly, these are concepts that
8:39are starting to be top spenders in the
8:40US, which are quite incredible, and
8:42they keep getting longer and longer. I
8:44really feel that the longer you can
8:46capture a person's attention on an ad,
8:48the better you'll be able to sell to
8:51them, and it's true we're seeing more
8:53and more of this "fulling" format. It's
8:56true that now it's really about
8:57education, a person in front of a
8:59whiteboard, uh...
9:01>> a guy with a phone you know walking
9:02down the street filming himself,
9:03talking to you, and you feel like he's
9:04going to tell you a secret, but in the
9:05end you realize he's actually selling
9:07you his product.
9:08>> Yeah, honestly, I totally, totally
9:09agree. What's funny is that ads are
9:11becoming easier and easier to make.
9:13Yeah,
9:14>> honestly, but on the other hand,
9:16there's more and more competition
9:17volume, and really, you still have to
9:19manage
9:20>> to be a pioneer in ad concepts, or else
9:23be the absolute top of the market. And
9:26that's what's becoming harder and
9:27harder. We'll talk about concepts a bit
9:29later, but really: how do you manage to
9:31consistently stay ahead of the market?
9:33Because actually, you know that now
9:34with AI, you’ll get copied in 4
9:35seconds, so you have to be the first
9:37one to breathe, to breathe life into
9:39the thing in your market, in your niche
9:40, and constantly be aware of that. And
9:42that's a lot of work on the, on the
9:44creative side. And we, I'd like to
9:47understand your market analysis process
9:49a bit to grab concepts, to grab what's
9:51best out there. And actually, how do
9:54you iterate on advertisements, for
9:56example on what's working, how do you
9:57find new concepts, et cetera.
9:58>> Hm.
10:01>> That way he gives me time to think.
10:03That's wonderful.
10:04>> Don't hesitate, when you need to think
10:05you can, you can answer right away.
10:06>> Wow.
10:08>> Does it get to that point for me?
10:09>> No.
10:09>> Damn, what did I break? It's not
10:11working anymore.
10:11>> That's not my job here. I don't...
10:13>> Can you connect the Wi-Fi? So, I'm
10:17stuck. It's good.
10:20>> Yeah, I think for market monitoring,
10:21it's always kind of the same thing
10:23actually. There are different levels of
10:24analysis. There's what your direct
10:25competitors are going to do. So
10:27typically, you'll go check, uh, on the
10:28...if we take Fincut as an example,
10:30you'll go look at the US market for
10:31those selling well-fitted t-shirts and
10:33what you can borrow from them, keeping
10:34in mind that the market isn't the same,
10:36that the promises, the claims, and the
10:38format of the creative won't be the
10:39same. But you test your angles, you can
10:41identify your marketing angles that way
10:42. In any case, it gives you your first
10:44level of analysis. The second is
10:45indirect competition, meaning those who
10:47might be in fashion, for example, but
10:49are selling something else. So, you
10:50know, shirts. Well anyway, today your
10:53product portfolio is evolving, but
10:55shoes. For example, those who sell
10:56shoes to the same target audience as
10:57you, and this time you ask yourself the
10:59question: you're going to look at the
11:00formats, less so the marketing angles
11:01because it's not the same product. And
11:02you're going to look at the formats and
11:04what type of people they sell to, and
11:05actually, you'll say: "Okay, well I see
11:07that they're ultimately selling a
11:08premium-style format, but in the
11:09creative, they only feature 40-year-old
11:11men."
11:11>> So okay, there you tell yourself that
11:13this is something worth testing. And
11:14what I really like doing is indirect
11:16competition. It's the level of analysis
11:18I prefer. You go see a brand that has
11:20nothing to do with your industry but
11:21sells to the same people,
11:22>> you know,
11:23>> like pharma,
11:24>> pharma for example. Actually, you can,
11:26you can go look at anything, cars, you
11:28can imagine anything. And I like that
11:30because it's the same with formats:
11:32you'll see things that keep you, uh,
11:33from staying too focused on e-commerce,
11:35because you're looking at brands that
11:37have nothing to do with it and that
11:38avoids a bit of the echo chamber. And
11:40that's what can make you come...it's,
11:41it's the expression of...I have to, I
11:43have to, I have to give it to them,
11:45actually. If you want your brand to
11:46break out of this overly easy
11:47e-commerce perimeter and actually
11:49become a brand, you need to see what
11:50brands are doing. And, uh, that's where
11:52I, that's my favorite level of analysis
11:54, but sometimes it's a bit subtle to
11:55take a well-crafted creative and bring
11:57it to e-commerce with your budget
11:58constraints, which are the ones
11:59inherent to e-commerce. But we manage
12:00to bring out some pretty interesting
12:02things. So, uh, those are my three
12:03levels of analysis. Direct, indirect,
12:04and completely unrelated competition.
12:06Market just, well, similar target.
12:08>> And uh, when you release a batch for
12:10Fincut, what’s the split between the
12:12percentage of creative iterations of
12:14what works, because we’re safe with
12:16that, and new concepts we’re going to
12:18launch? In this case, we pretty much
12:22follow market standards, which is 70/30
12:23, because if we make highly polished
12:25creatives like we’ve done in the past
12:26—the whole lower funnel is like that
12:28—if it doesn’t perform afterward,
12:29we’re out as an agency, which is
12:31normal, you see. So with iteration, you
12:33take fewer risks
12:34>> and so we try to have a significant
12:35share, but honestly, I’d say we’re
12:37more at 60/40—60%iteration, 40%—in
12:39fact, that was the split for the latest
12:41concepts we proposed to you. And you
12:43have to ask yourself how you define
12:45that ratio? Well, what’s your raw
12:46material? You see, there’s no general
12:48rule, actually. If the brand has little
12:50history, you’ll have to create
12:52something new. So, you start from what
12:54works in the market, from the three
12:56levels of interpretation I gave with
12:57Fincut. There’s quite a bit of
12:59history, but at the same time, you also
13:01wanted to change the messaging or the
13:02image. So, there are things that are
13:04difficult to iterate on. That’s why
13:06we still proposed something new.
13:08Typically, when we arrived with the
13:09first batches with the whiteboard, the
13:11scientist, and so on, we could consider
13:13it new, but it was just the format that
13:14was new, while we kept the same
13:15arguments inside. And incidentally, we
13:18can talk about that, finally. What are
13:19iterations? What is it, what is it,
13:21what can we consider an iteration, or
13:22at what point does it become a new
13:24concept? Because in the e-commerce
13:25world, there is
13:26>> it’s true that there’s a pattern
13:28that is changing. Back when you could
13:30just change a hook and actually it was
13:31just the first three seconds of the
13:32video, the first five seconds that
13:33changed. Uh, well that worked, but now,
13:35honestly, less and less, with Andromeda
13:37, and everything that’s happened
13:39recently with MTA. And I feel like now,
13:41actually, if you want to iterate on an
13:43ad that works, beyond just changing the
13:45hook, you have to change the entire
13:47body of the video, in a way. But, for
13:49example, you change the environment,
13:51you change the persona, you basically
13:53keep the exact same ad, except you
13:55change something really obvious. So if
13:57for example you were in a closed room,
13:58well, you're going to be on a private
14:00jet. I don't know, some random
14:01. Uh, you use two different characters.
14:03If you had a guy, you use a bigger guy.
14:04If you had a white guy, you use a Black
14:05guy. I mean, you just don't give a damn
14:06. You just test different big things in
14:09your creative and that creates a real
14:11iteration accepted by MTA. Whereas a
14:13few years ago you could just change the
14:15first three seconds and it would be
14:16enough, but now the ads really have to
14:18be different and that's what people
14:19still struggle to understand. Even
14:22though everyone sees Andromeda here and
14:24there, people don't get that the ads,
14:25even in static creative, really need to
14:27be distinct from one another and not
14:29just micro-changes that make MTA
14:31compare the two ads, see they're almost
14:33identical, and just stop one of them.
14:36>> Yeah, that's it, and I think Eddie has
14:37some things to tell us about that. No,
14:40I'm, I'm, I'm very aligned with that
14:42vision, which is that we tend to focus
14:44too much on the hook, but essentially,
14:46once you've studied a BM, you're
14:48supposed to know what the top 10 are.
14:51So essentially, you know how to catch
14:53people regarding your niche, your
14:54product. I mean, normally it's, it's
14:55quite simple actually. Behind that,
14:57it's really about saying, "OK, I spoke
14:59to one persona, a 40-year-old Caucasian
15:01guy, a dad, and so on." Actually,
15:04what's the, the next iteration? Well,
15:06in your winers, careful, you only
15:08iterate on the ones that work, you see.
15:10Well, there, you, as you said, you take
15:12a Black guy, the same one who has the
15:14same family situation, actually in a
15:15slightly different context, meaning the
15:17setting. That is an iteration and,
15:19actually, it’s supposed to do the
15:21same...I mean for us, the thing is: you
15:24take the guy who, who’s in business
15:26and, well, you could go with a guy who
15:28does sports versus a man who is, I
15:31don't know, a chef. Actually, for me,
15:32that’s what an iteration is. It’s
15:34that the message stays the same, the
15:36hook stays the same because we’ve
15:37found the formula, you see, it’s hook
15:39script, you see. But behind that, you
15:41have to change the setting and the
15:43persona. H
15:44>> And that’s what’s crazy. Actually,
15:46I’ll just jump in on that because the
15:48reason I mentioned Dedy is that when we
15:50worked with him, he developed a
15:51philosophy; he told us he didn't want
15:53the same b-roll in two different videos
15:55. And it’s also precisely for that
15:56reason, for the algorithms, so that the
15:57videos don't look the same. And so I
15:59think you really have to go further
16:00than just saying, well, you have your
16:02bank of b-roll that works well, well
16:03yes, but you have to break out of it,
16:05you have to renew it.
16:07>> Yes, because a b-roll where you show
16:08all the benefits of a man, I don't know
16:10, from North Africa, you can't put
16:11other people in
16:15>> in that b-roll because, in fact, we can
16:16no longer identify with the...the
16:18message was embodied and so I no longer
16:19identify with the message
16:22>> next, because behind the ad, well, the
16:23landing page is supposed to be
16:24dedicated to him too, you see. Yeah,
16:26>> so for me, you also have to be very,
16:27very careful with the b-roll.
16:29>> But now, what’s completely crazy with
16:30iterations currently using artificial
16:32intelligence is that we’ve actually
16:34just finished developing a big part of
16:36our internal software for that. You
16:38take an ad that worked well, you
16:39literally change the persona you want,
16:42and it reproduces exactly the same
16:43video but ultra-realistic. You don't
16:45even—it’s almost face swap, you see
16:47. You don't get the impression that
16:48it’s a new, a new creative.
16:51>> And in fact, it duplicates your winning
16:52concepts automatically. So, in fact,
16:54the role of creative agencies is almost
16:56just to find working concepts on repeat
16:58, and iterations are almost no longer
16:59an issue to manage because, in essence,
17:01the agency internalizes it or makes it
17:03cheaper or something like that. Anyway,
17:06we are entering a quite different
17:08period in terms of creative strategy
17:10and just creative work in general;
17:12thanks to artificial intelligence,
17:14everything is moving faster. And we
17:17realize with Jabert that the tool is
17:19just insane.
17:20>> Like, you put in a creative
17:22>> that worked, you say what you want to
17:24change, and the creative is
17:26pixel-perfect, it’s incredible, and
17:28actually
17:28>> Persona and background.
17:29>> Persona, background, for example,
17:30>> but it can say the same hook, the same
17:32pitch,
17:33>> lips pixel-perfect,
17:34>> pixel-perfect in every language.
17:36>> Excellent.
17:37>> And that’s it, for me it’s
17:39>> they’re scratching their heads. Yeah,
17:41I’m in “let’s do that” mode.
17:43>> Yeah.
17:44>> No, but exactly, for me, no,
17:47>> because that’s what we were talking
17:49about earlier. We’re about to launch
17:51Fincut worldwide, in the main major
17:53markets across the globe. And we
17:56actually still need human beings who
17:59will create full concepts. This means
18:02we will need even more new videos and
18:04new concepts than before, because
18:06that’s what will fuel the artificial
18:08intelligence behind it. That’s
18:11something I repeat in several of the
18:13last videos I post: we aren’t at the
18:15service of AI, AI is at our service.
18:17And that’s something you have to
18:19understand and really internalize,
18:20because if you let it do everything for
18:22you, you’re dead. Like, you don’t
18:24even have a small chance of succeeding
18:26in a field where there will be
18:27ultra-competition, where there will be
18:29very intelligent humans using AI to
18:31evolve. Actually, that’s what you
18:33need to keep in mind. You always need
18:35humans who are constantly searching,
18:36searching, searching, trying to create
18:38the best concepts, trying to create
18:39volume with real people, with real
18:41stuff, with real knowledge, to then
18:42feed a machine that you create and you
18:44direct. If you start doing the opposite
18:46, you’re dead. And that’s why we
18:47were talking about it yesterday. We are
18:49going to try to reach 100 creatives per
18:51month pretty quickly on Fincut to feed
18:54this AI, to have more and more concepts
18:56, and also the pod system we are going
18:58to set up. We can talk about it a
19:01little bit later, but a pod system that
19:03allows you to put in even more humans
19:05than before to constantly create
19:08incredible creatives. And that's what,
19:11for me, separates a good e-commerce
19:13seller from a bad one. The bad seller
19:15will do everything with AI, understand
19:16nothing, launch en masse, not
19:18understand why it’s performing poorly
19:19, and iterate on something they didn't
19:21even understand that worked by luck,
19:22and so on.
19:24>> Versus e-commerce sellers who have
19:25their structure, their sound strategy,
19:27who know exactly what to analyze, what
19:29to duplicate, what to iterate, and then
19:30they use AI to transfer to different
19:32markets, and then do market analysis to
19:34really understand what worked in the
19:35market, why it worked, and so on. And
19:38in fact, it's really a logic of
19:40intelligent scale thanks to artificial
19:42intelligence. Powered by human
19:45knowledge and human analysis.
19:47>> Exactly. And I think that's why,
19:49ultimately, even if in terms of
19:50iteration, when you talk about AI and
19:52the fact that it's going to accelerate,
19:53I don't feel that threatened because,
19:55in the end, these are processes I can
19:56use internally too. But you gain in...
19:58well, the human still has their place,
19:59and as a result, you actually gain
20:00striking power. We will manage to do
20:02more work with less time and fewer
20:04human resources, but every human
20:06working with us will be augmented, kind
20:08of like bionic people. Actually,
20:10generally speaking, if you don't input
20:13your own intelligence, you've lost.
20:16>> Yeah. That's clear.
20:17>> If you let AI do everything and bring
20:19its own intelligence, you're dead.
20:21However, if you bring your own
20:22intelligence to it—meaning your
20:24analysis, how to understand each KPI—
20:26it's not flat, you see. There are still
20:28many things it doesn't provide, so
20:30you're dead. And today, many people are
20:32doing things automatically.
20:33>> If you don't bring the intelligence,
20:35the intelligence is your intelligence.
20:36>> For me, you're dead. That's clear, and
20:38I think people underestimate the amount
20:40of context you need to give an AI so it
20:41gives you a relevant answer. And that's
20:43why at all levels, even e-commerce
20:45sellers who automate their market
20:46research a bit with AI by saying, "I
20:48want this product, I found this product
20:50, tell me what the market thinks," They
20:51are going to get insights that are
20:53wrong. And I've seen e-merchants who,
20:55on the contrary, out of passion for the
20:57market, go and look at everything;
20:58they’ll go watch creatives on TikTok,
21:00do market research on Reddit, forums,
21:02and so on, and they are the ones who
21:03are right. OK, it takes longer, but
21:05they are sure—well, they increase
21:07their chances of finding something
21:08relevant. B
21:09>> I think there are many people watching
21:11this video who will say to themselves,
21:12"Damn, how long has it been since I
21:13really analyzed my shop, analyzed my
21:15stuff, since I left everything to AI?"
21:16>> Really dive into your brand, dive into
21:18what's being done on the market, look,
21:20open Trend Track, don't just use the
21:22MCP, you see? On the library, things,
21:24analyze, look at the comments. And
21:26actually, when you look at that
21:27yourself and not AI—because AI will
21:29do the work, but it will do it in a
21:31different way, in a way you don't
21:32understand—above all, how will you be
21:34able to...at best, in the short term,
21:36OK, it might work, having AI do
21:37everything, but how can you scale a
21:39brand that you yourself don't
21:40understand? In fact, you own a brand
21:42that you don't understand, so
21:44essentially, you own almost nothing. If
21:46you don't have the knowledge of your
21:47brand, you don't know what works, you
21:48don't know where it's going. Even for
21:50your future product decisions, your
21:52next marketing decisions, you'll be
21:53doing it a bit by gut feeling, while
21:55you've handed all the intelligence to
21:57someone who isn't you, who can make
21:59mistakes, and you are losing your added
22:01value as a human.
22:02>> Exactly. Yeah.
22:04>> So it's dangerous. It's great and it's
22:07very dangerous.
22:08>> Clearly. But it's like if you want to
22:09make a contract, you ask AI to write it
22:11for you. To you, the contract looks
22:12fine,
22:13>> but you give it to a lawyer, they'll
22:14tell you there’s no structure, no
22:16legal foundation behind it. It doesn't
22:17have the right vocabulary. But it's the
22:18same for market research, it's the same
22:19for scripts, and so on. You can't rely
22:21on it. You clearly have to dig to
22:23understand it well yourself, and in any
22:25case, that will prevent you from asking
22:26stupid questions where you get AI
22:28outputs that are useless to you.
22:30Actually, it's up to you to drive it
22:32once again, it really comes down to one
22:34word. You have to bring your
22:35intelligence to the intelligence, or
22:37you're dead.
22:37>> Yeah, it's the same, the most
22:39interesting insights really come from
22:40the customers. Calling them, sending
22:42messages, looking at customer service.
22:44Good luck, Alia, finding that.
22:46>> Yeah, seriously. OK, maybe you can
22:47automate the analysis of your customer
22:49messages if you have a ticketing system
22:50and whatnot, but at some point, a real
22:52conversation with a customer,
22:54>> sometimes one sentence, and you find a
22:55new marketing angle where people aren't
22:57buying at all for the reason you
22:58imagined, they just liked the product;
22:59you can't know that if you don't call
23:00them, your market research won't give
23:02you that.
23:03>> So, do you do you need to have access
23:07to what data yourself to analyze, well,
23:11what you deliver?
23:12>> You mean what do you rely on? We gave
23:14you access to the BM, we gave you
23:16access to everything, but what do you
23:18actually check yourself, and at what
23:20frequency? Before finding my insights,
23:22you mean? Or even just to figure out
23:23what you actually need to do,
23:25>> new concepts, new stuff, we spend a lot
23:27of time looking at brands 'organic
23:29content, even yours, just to see what's
23:31happening in the comments. Actually,
23:33we're looking for human insight, and
23:35whenever we can, since we're targeted
23:37by your ads, we look at them when we
23:39come across one of your creatives. I
23:41read the comments, every single one, I
23:42look at what's being said today. Well,
23:44the comments are a little filtered,
23:46it's a bit like reviews. We try to
23:48highlight what works best. So there's
23:51that human insight, and there's always
23:53the Business Manager. And here we can
23:55talk about metric analysis because we
23:57have a kind of cascade that tells us, "
23:58OK, if it passed the first filter, we
24:00move to the second." If it passes the
24:01second filter, we move to the third,
24:02and that's when we decide what to do
24:04with it. And the third one is Trend
24:06Track. I spend my days on Trend Track,
24:08on the tools, the aggregators. And
24:09that's what I was just saying, it's my
24:11favorite part, going to look at
24:12industries that have absolutely nothing
24:13to do with us.
24:14>> After that, I take all the inspiration.
24:16>> And what are your filters? Because it
24:17could be super interesting for everyone
24:19to understand. Yeah, totally. Actually,
24:20if you want to iterate on a winner, you
24:22need to define what a winner is.
24:23Obviously, there's the ROAS. I find
24:25that interesting. Everyone says, "Yeah,
24:27we're anti-ROAS and all.""In the end,
24:29that's what everyone looks at, and so
24:30for ROAS, I'd even say the NC ROAS, the
24:32new customer, everyone looks at
24:34standard ROAS."
24:36>> But for me, it doesn't mean much on its
24:37own. For example, we recently launched
24:39shirts, we're getting ROAS of 5, 7, and
24:41...
24:42>> etc.
24:43>> But when you look closer, it's all
24:45returning customers, whereas a brand is
24:47actually healthy when you have a higher
24:49NC ROAS that allows you to gain new
24:50market share, etc. That's what you
24:52really need to look at to realize how
24:54profitable you really are, or else you
24:56risk running out of steam quickly on
24:57your project.
24:59>> Yeah, but you can't find that on Meta,
25:01it seems to me.
25:02>> Yes, you can actually get it.
25:03>> OK.
25:04>> In fact, you can filter the segments,
25:06and on the segments, you can select new
25:08customers and, uh, and well, prospects,
25:10sorry, excuse me.
25:11>> OK. Well, we stop at ROAS at this stage
25:15. Actually, it also depends on the
25:16brand's level of maturity. But indeed,
25:18in any case, when you look at how
25:20healthy the business is, you start
25:21asking questions about NC ROAS, MER,
25:23etc. Just metrics that differentiate
25:24what you invest from what you actually
25:26acquire in terms of new customers. But
25:28so, we look at the ROAS and we always
25:30put it in light of the spend, because
25:32if you say, "Well, okay, I have a 6
25:33ROAS, but the creative only spent €
25:35100," it might be a bit early to draw a
25:36conclusion depending on the size of
25:38your BM. And that actually gives you a
25:40scalability index that we calculate.
25:42And then we look and ask, "OK, where
25:43does this index come from?" Is it
25:44because it generated a lot of clicks?
25:46Is it because it managed to hold
25:47attention for a long time until, etc.?
25:49"So, first metric, we look at quality
25:51or click-through rate times hold rate,
25:53and we actually create an index
25:54relative to the rest of the creatives
25:56in the BM." That way, you don't just
25:58compare it, saying, "Go, it's at 10%,
25:59that's good." No, it's about how much
26:01it outperforms the others. You look at
26:03your CTR, you also look at your CPM to
26:06ask, "Did it reach an audience?" Maybe
26:08it found an interest within an audience
26:10, so Meta distributes it cheaply,
26:11affecting your CTR and all that.
26:13Actually, you create ratios as you move
26:16forward, going down the funnel,
26:18>> you know, you’re going to look at
26:20your conversion rate and such. And in
26:21fact, when you have all these metrics,
26:23it gives you a number of clues, you
26:25compare them to the rest of the BM, and
26:26let's say a creative sold well, but the
26:28attention quality isn't very good. This
26:30means that for the few people it
26:32reached, the marketing angle was super
26:33convincing. And there are others, we'll
26:35look, we'll see very good attention
26:37quality, but very few clicks. So, in
26:39fact, it touched on something, but it
26:40didn't manage to convert them. And from
26:42there, you decide what you test. So for
26:43example, if you see that it held a lot
26:45of attention but didn't convert, maybe
26:46the value proposition comes too late.
26:48I'll make a creative where I rework it,
26:49bringing the product or the proposition
26:50in sooner. And that's it, it creates an
26:52analysis grid for you
26:55>> which is a really interesting topic to
26:57develop, but you can't just apply one
26:58funnel to everyone, but I think you see
27:00, you've understood the, yeah,
27:02>> of course
27:03>> the mechanics. I'm currently testing a
27:05software whose name I won't mention
27:07because I don't have enough data yet to
27:09know if it's really good or not. But I
27:12find the concept quite interesting,
27:14basically, they connect to your
27:15comments, your customer service, your
27:17BM, and they really connect to your
27:18reviews and everything to
27:20>> Your reviews, really all the
27:21touchpoints with your customers, and
27:23they also collect all the concepts that
27:25work well from ads on the other side,
27:27and based on that, create creative
27:29concepts that you can then give to your
27:31creative strategists so they can
27:33develop them even further, and so on.
27:36But honestly, that's the real
27:38foundation of creative strategy for me:
27:39looking at all your customer
27:41touchpoints, all the frustrations, all
27:42the problems, everything they love, and
27:44then iterating as much creative as
27:45possible based on what the market likes
27:47or dislikes about your product. And
27:50actually, in any case, you've always
27:52got great creative to make because a
27:53customer who isn't happy is a great ad
27:55to make. A customer who has a question,
27:57well, that’s a superb ad. A customer
27:58who is satisfied, well, that's even
27:59better. And in fact, you have tons and
28:02tons of ad ideas that flow from what
28:03your client says, and you're constantly
28:05heading in their direction. That’s
28:09what really lets you have a staggering
28:10number of creatives. I would add a
28:13nuance, though, in the sense that you
28:15say you're constantly heading in that
28:16direction, and that might be a bit of a
28:18problem: you end up narrowing your
28:20creative range and repeating the same
28:22thing over and over. And that’s why
28:24the ratio you asked me about earlier is
28:26good.
28:26>> Yeah, OK, do 70%of your ads by leaning
28:28on what the market gives you. Keep 30%
28:30where you say, "OK, I’m exploring,
28:32I’m testing new things," and so on.
28:34And that’s the nuance that keeps you
28:35from getting locked into something
28:37where you’re completely segmented and
28:38only talking to one type of person.
28:40>> Yeah.
28:40>> Because, really, who is going to give
28:42you signals on your ads, your comments,
28:43and all that? Well, the people you’ve
28:45managed to reach, and since you’ve
28:46managed to reach them, you’re in a
28:47bias, you see.
28:48>> Yeah. Also, the people you manage to
28:49reach are often prospects you haven't
28:51converted yet. So, in fact, these are
28:52also new pockets of audience you need
28:54to go after, and if you manage to
28:55unlock them, well, that’s actually
28:56pretty good for your brand, rather than
28:58people who aren’t aware at all.
29:00Actually, in your comments, you find
29:02your customers and you find your
29:04prospects. And for me, prospects aren't
29:06necessarily iteration in the sense that
29:08they are new pockets of audience that
29:10are already there, that are almost warm
29:12and just waiting to be converted. And
29:15actually, if you constantly focus on
29:16your new prospects, they become clients
29:18, which creates new prospects, and
29:19that's how you manage to grow. Of
29:21course, you still need to reach people
29:22who have nothing to do with it, who
29:24don't know the product at all. But if
29:26you can already manage your prospects
29:27well and convert them, that's already a
29:29big step for your brand. And it allows
29:31you to constantly be aware of new
29:32prospects coming in and how to convert
29:34them. And it's basically an infinite
29:36loop that allows you, step by step, to
29:37eventually capture the whole market
29:39because one prospect leads to another,
29:40and so on, you see.
29:41>> Yeah, that's interesting. I have a
29:44question, it's—we see a lot of
29:45creative strategy agencies and there's
29:47a lot of ambiguity about their
29:48organization. Behind this type of
29:52agency, it's something that interests
29:54me a lot, especially today, because I'm
29:56currently working on setting up
29:58different pods. We need to talk about
30:01it later with Amy and so on, but the
30:03question is: how does an agency,
30:05especially with many clients and really
30:08a lot of deliverables, do it? How does
30:12an agency manage to organize itself
30:13well to produce a high volume of
30:15creative work while maintaining a
30:16significant level of quality? Yeah,
30:18that's a good question. It might echo
30:21the pods you mentioned. So, is it okay
30:23if we talk about it?
30:24>> Of course, of course.
30:25>> Yeah. OK. First thing, a disclaimer. My
30:27agency isn't even a year old. You have
30:28to be honest about that. I'm not going
30:29to give entrepreneurship or structuring
30:31advice on things I haven't done, but
30:32I've already encountered the first set
30:34of problems, and we are part of an
30:35agency incubator with agencies that
30:37produce 400 to 500 videos per month. So
30:38I was able to talk to those founders to
30:40try to understand. So, I'm giving
30:41feedback as I see it today, as I
30:43imagine it. And potentially, anyway,
30:45everything I said today might be—well
30:47, I might believe it to be wrong in 3
30:49months—because my perspective evolves
30:51as I evolve with the market, and so on.
30:53Now, current status, I'll tell you
30:55what's happening. Indeed, your pods—I
30:57call them APUs, autonomous production
30:59units. Meaning, what do you need for
31:00someone to be autonomous? You need
31:02someone capable of thinking up briefs
31:04and writing scripts. So, a creative
31:06strategist. OK. Then it depends on the
31:08volume they can handle, but you see,
31:09for a creative strategist to stay
31:11creative, you won't give them 80
31:12scripts.
31:12>> You'll stay around 30, let's say, for
31:14example.
31:14>> OK.
31:15>> An editor can handle
31:16>> a script—that is truly a concept.
31:18>> Yeah, that's it. Let's talk about
31:19concepts later. But a script is a
31:21concept in terms of creative strategy.
31:24Exactly. We'll talk about that later,
31:26>> not about mixing existing creative
31:28works and so on. It's really something
31:30new.
31:31>> It depends on your definition of a
31:33concept. You could imagine that
31:34remixing, doing B-roll, voice-overs,
31:35and all that is a concept. Uh.
31:37>> Yeah, I'm trying to re-explain what a
31:39concept actually is.
31:40>> Yeah. So.
31:41>> According to you, what is it? Just so
31:43we're on the same page about that.
31:44According to you, what is a concept? I
31:45mean, what's the diff?
31:46>> Actually, for me, a concept is
31:48something innovative. It's something
31:50new. It's
31:51>> a new concept.
31:52>> It's a new concept. Actually, you have
31:54concepts that you repeat; those are
31:55iterations that stay within your
31:56concept, you see. Just because you
31:58changed a character—because that’s
31:59where our previous conversation got
32:01stuck. Changing a character
32:03>> is not a new concept.
32:04>> It is not a new concept. A new concept
32:06is a different approach to what you are
32:10doing. It's, uh, for example, doing a
32:1390-second mini-series, I don't know,
32:16with drama. Uh, that's a concept. Doing
32:19"us versus you," that's a concept. Uh,
32:23staging a fake protest with a megaphone
32:25selling your product, with everyone
32:27like that, it's a concept, with a guy
32:29on stage as a comedian.
32:32>> Exactly. And in the end, he makes a
32:33joke about how he dresses. Boom, that's
32:35a concept. And in fact, whether the
32:38actor is white, Black, Arab, or Chinese
32:41, that's just an iteration. And
32:43actually, that's where we didn't
32:45particularly agree. And, in fact, what
32:47I was saying is that we're getting a
32:48bit too lost. A concept is something
32:50innovative. So, telling me that someone
32:52can put out 88 concepts in a month, I
32:55think to myself that Steve Jobs didn't
32:57release that many iPhones. I mean, you
32:59know, like the Hermès Birkin, I mean,
33:02you see what I mean?
33:03>> Yeah, absolutely. Actually, you really
33:06have to focus on the fact that for me,
33:08new concepts need to be, uh, how can I
33:11put this? You have far fewer of them to
33:14release than iterations, actually. But
33:16>> actually, you make fewer concepts but
33:17more volume per iteration because the
33:19iteration will be super profitable in
33:21any case. Well, yeah.
33:22>> And so it's true, might as well focus
33:23on a few less concepts and more
33:25iteration volume to really go
33:26>> because imagine a concept you release
33:28and you send 15 personas, so your
33:30iterations, actually maybe only two
33:32personas will hit, and that's when it
33:34gets interesting.
33:35>> You see?
33:36>> Because it's true that for every new
33:37concept or expenditure of energy on a
33:38concept, it's a shame to only test it
33:39with one persona and say it's a flop or
33:41not, when maybe the persona next door,
33:42the neighbor, is the one that would
33:43have worked, you see.
33:44>> Yeah.
33:45>> Yeah, but that's why the level of
33:47concept definition a definition only
33:51makes sense in its context. You see, if
33:53I ask you to give me the definition of
33:54a foot, you'll tell me for example: "
33:55It's the end of the leg." But actually,
33:57if I talk to a guy who takes
33:58measurements all day, he'll tell me: "A
33:59foot, yes, it's a unit of measurement,"
34:00and so on. It's the same word, it
34:01doesn't have the same definition. So I
34:03asked myself the question after all
34:04these discussions and as said, to know
34:05what a concept actually was. And
34:06actually, I arrived at an analysis
34:08where, in fact, there are three levels
34:09of reading a concept. The first is the
34:11strategic level. You're going to say
34:13basically, what message are you putting
34:14in? What do you want to say or what are
34:16you trying to convey? The second is the
34:18creative concept. And there, I think
34:20that's what you're talking about, it's
34:21the envelope in which you put your
34:22message. And the third, which was
34:25ultimately the level I called a concept
34:26because it's somewhat of an e-commerce
34:28market trend, is the creative strategy
34:29concept, and you'll tell me, "Okay, you
34:31just put the first two together and
34:32think that's it." But actually, in
34:34e-commerce, the bias and what keeps you
34:36stuck in e-commerce and prevents your
34:38brand from evolving with insights, is
34:39you. You, we're moving past that, we're
34:40stepping out of that echo chamber.
34:42It’s that, in fact, an iteration is a
34:44concept between a persona, a format, a
34:46level of awareness, and an offer. A
34:48product. If you change one of these
34:51things in creative strategy, you
34:52consider it a new concept. That's,
34:54that's the nuance. And so, actually, I
34:56think that when I started thinking in
34:58creative concepts, following
34:59discussions with Amdi, yeah, I told
35:00myself: "Well, it's not enough to just
35:02say, instead of doing, instead of
35:03talking about a 5-pack of t-shirts, you
35:04talk about a 2-pack, that's it, I've
35:06made a new concept," that's not it. So,
35:09I evolved on that vision, and uh, and I
35:11like it, because in my opinion, for
35:12those listening, it's what takes a
35:14brand from e-commerce to an actual
35:16brand, because you start producing
35:18things that are really different from
35:20one another. And I really, I specify
35:22also, well, and I think Alex is pretty
35:24much in agreement with me, that when
35:27you test a concept, a real concept,
35:29mind you, and you've uh, you've made uh
35:31hook variables, that is, three
35:33different hooks, it's been delivered.
35:36Uh, actually what can be problematic
35:38sometimes is just the persona. It's
35:41that for this same concept, you have to
35:43run all the iterations for every
35:46possible persona in your audience.
35:48There’s surely one or two that will
35:50hit. But often we say, "Ah, the concept
35:52didn't work, yet we had three hooks."
35:54No, no, sometimes it's just the persona
35:55, actually.
35:56>> It's just that you spoke to the wrong
35:58target or the audience pocket that
35:59Andromeda delivered to wasn't the right
36:01one. So that's where you need to
36:03diversify.
36:04>> Absolutely. Agreed.
36:05>> So if we go back to the production
36:07value chain.
36:08>> Yeah.
36:09>> So, well, the person.
36:11>> There you go, now everyone has the same
36:13level of definition. So to recap, how
36:14do you scale your creative volume? You
36:17need autonomous production units, or
36:18APUs. In concrete terms, this is
36:20someone who develops the briefs and
36:21writes the scripts. So, they create a
36:24strategy. Meaning it’s not just about
36:26producing for the sake of it, but
36:27asking what question or hypothesis my
36:29brief and script are answering. So that
36:31is the creative strategist. You need
36:32one of those. To keep them creative
36:34enough, you can't overload them. That's
36:35why we try to aim for around 30 per
36:37month maximum, including iterations in
36:39that count. Actually, what we do is
36:41tell them, look, 70%is iteration, so
36:43there's less creative thinking required
36:44. However, for the last 30%, push
36:47yourself and try to propose something
36:49truly new. And that’s when we start
36:50thinking about the creatives we’re
36:51going to release.
36:52>> For all our clients, or at least those
36:54whose budget allows it, we still manage
36:56to put out really high-quality work. OK
36:58. And when I say the budget allows it,
37:00today we aren't limited with a brand
37:01like Fincut, for example, but there are
37:03much smaller e-commerce brands that
37:04trusted us from the start, whom we
37:06continue to support. There, we are a
37:07bit more constrained. OK?
37:08>> You also have to be realistic; when you
37:10have big ideas, you have to put up the
37:11budget. So you have one creative
37:13strategist who does 30, and one editor
37:15can handle 30. But well, in case there
37:17are questions or whatever, we always
37:18have two editors so one can take over
37:19from the other and so on. And actually,
37:21from there, you understand that you can
37:22produce, and that's enough. You see,
37:25you also need your production team, but
37:26for me, for example, a production
37:28manager or a cameraperson can move
37:29between several autonomous production
37:30units. You see, for instance, a
37:32production manager will be able to
37:34schedule around twenty shoots. Each
37:35shoot produces between 4 and 5 videos.
37:37So, in fact, you'll be able to produce
37:3880 per month, you see. H
37:40>> So you have your creative strategist,
37:42you have your editor. The three of them
37:43are super close, you see. They work
37:45together all the time, they exchange
37:46feedback, and behind them, you have a
37:47production team with a cameraperson. In
37:49general, for social media and socials,
37:50that's enough. You just need a camera
37:52operator who knows how to hold an
37:54iPhone or a camera and, uh, your
37:55production part is then handled
37:57in-house.
37:58>> Okay, that's the base level.
37:59>> And sorry, in terms of production, you
38:01have the creative strategist duo plus
38:03editor.
38:03>> Trio. I still like having three editors
38:05, though. Two editors and one creative
38:06strategist trio. Well, of course, of
38:08course. Two editors and one creative
38:09strategist. And on the side, you have
38:10the production team, which is also an
38:12internalized team that will handle all
38:14the shoots in Paris, for example, etc.
38:15That's a production manager. So, their
38:17role is to lead, find the actors, find
38:19the locations. It's really a production
38:21manager's job.
38:22>> Yeah. Production manager. And above
38:23that, there's the head of production,
38:25but we aren't there yet. And that's
38:26what a production manager is. They are
38:27the logistical conductor who brings the
38:29script to life. Exactly. They handle
38:30the transition of the script, which is
38:32ultimately a living object that passes
38:33from hand to hand, telling you I need
38:35this environment, this location, this
38:36prop, this person. And they bring it to
38:38life with a shoot, summon everyone, and
38:39boom, we're on set.
38:41>> They also find the technical crews.
38:42>> I say technical crews because our needs
38:44today are just, ultimately, a camera
38:46operator who knows how to manage a bit
38:47of light and sound. But tomorrow, if
38:49you do much more ambitious productions,
38:50you need your sound crew, your lighting
38:51crew, and so on.
38:52>> Okay?
38:53>> Exactly. It's good to have it in-house
38:55because, since your script is a living
38:57object passing from hand to hand, you
38:58need to be able to internalize the
39:00processes. So we prefer to have someone
39:02rather than just a production team.
39:03>> Of course.
39:04>> Exactly. And uh, afterwards, well, you
39:06need resources for actors, you need
39:07resources for locations, and so on. So
39:09all that needs to be documented. But
39:11with this info, you can run a shoot.
39:13>> Okay.
39:14>> Who is in charge of finding the
39:16characters? For example, if you need a
39:1840-plus character, etc., who is it?
39:20Whose part of the team is it to find
39:22the right actors?
39:24>> First of all, in my opinion, or at
39:25least what we do, is that the creative
39:26strategist is the one who makes the
39:28need emerge precisely. They will tell
39:29you, you see. But finding the actor,
39:31contacting them, giving them the
39:32information, scheduling an appointment,
39:33that's the production team, that's
39:34production.
39:35>> OK.
39:35>> There you go.
39:36>> OK.
39:37>> Interesting. And so that was level 1.
39:39You mentioned higher levels, what would
39:40those be for you? Actually, the higher
39:42level—and here it's more agency than
39:45brand—Hm.
39:46>> The higher level is when you start to
39:48have a volume of 100, 200, 300 videos
39:50per month, who validates the videos,
39:52you see, who ensures the quality of
39:54what you want to deliver to your
39:56clients.
39:58>> If I ask the creative strategist, they
39:59step out of their creative role,
40:00they're a bit...they'll waste time
40:02watching videos when they should be
40:03starting to think about the next
40:04iteration.
40:06>> Yeah, exactly. But it's a bias, because
40:07since it's their work, they have the
40:09idea they want to create in their head.
40:11And in fact, production works with the
40:12constraints they have as well. So in
40:13the end, when they deliver, they'll say
40:14, "This isn't what I expected at all, I
40:15can't do anything with this anyway."
40:17They won't necessarily put in as much
40:19effort as someone who is impartial.
40:21We're going to...I'm going to explain
40:22exactly how that works. So, even if
40:23it's their work, and basically for me
40:25it was them who validated the videos,
40:27they don't have the skills to say: "
40:28Here I need this sound effect, here I
40:30need this zoom, here I need..." And
40:31that's why the person who validates the
40:33videos is what we call a lead editor,
40:35and for us, it's someone with a
40:36background in editing. Because since
40:38they receive the finished object, which
40:40is the edited video, they have no bias
40:41regarding what it should have looked
40:42like.
40:43>> They see the object as it is and see
40:44how to improve it from there.
40:46>> OK?
40:46>> Exactly. And that's actually why you
40:48have to ensure to your creative
40:49strategists that at the script level,
40:51the right level of information is
40:52provided. Hm.
40:53>> Because in the end, the client receives
40:55a video that answers a brief we
40:56presented to them beforehand, and so
40:57the creative strategist must have been
40:58able to say exactly if there were
41:00important points that absolutely had to
41:01go to production—it's in the script.
41:03>> Exactly, if they want a transition, a
41:05visual effect, a sound effect at some
41:06point, it's in the script.
41:08>> There, that's why I say it's a living
41:09object that passes from hand to hand.
41:11>> Yeah, I understand. For those who
41:12didn't quite understand what a pod was,
41:14just because we mentioned it but didn't
41:15really explain it.
41:16>> Yeah.
41:17>> The goal of a pod, well, it's exactly
41:19as you said, it's a creative team
41:20within your creative team. This means,
41:23essentially, the goal for a company is
41:25to have different pods, different
41:27production units, that are each
41:29independent from one another to
41:30constantly propose new concepts, offer
41:32the best edits, offer the best AI
41:34videos, the best human videos, and so
41:36on. Because it's true that we've talked
41:38a lot about human videos with you, but
41:40not yet about AI videos. I don't think
41:42it's an angle you've taken in your
41:44agency yet regarding that, but in terms
41:46of creative strategy, it's also a major
41:47, major direction to take. It’s a
41:49creative strategist. AI, or rather
41:51human, but who creates strategies for
41:53AI videos, with two or three AI editors
41:56who handle the production of this
41:58person's idea. So there's that, and in
42:00fact, the goal of a pod is to put
42:02different pods—meaning different
42:03production units—into competition to
42:05create a bit of a rivalry between them
42:07and see who has the best data. And, in
42:09fact, for the people who have the best
42:11data, the goal will be to reward them
42:13with potential bonuses, incentives,
42:14etc., to really constantly push people
42:16to create the best creatives, come up
42:18with the best concepts, so that they're
42:20constantly driven to crush it and
42:22become the number one pod team. And the
42:25long-term goal, you see, is to have 5,
42:276, 7 different pods, seven different
42:28production units that will constantly
42:30hammer the market with tons and tons of
42:32ultra-high-quality creatives. And
42:35that's where you actually start to
42:36become a very, very big e-commerce
42:38merchant and can start to take on the
42:39global market without, without, without
42:41any problem. Basically, just the little
42:43>> Yeah, I would even say to easily
42:45multiply pods, honestly, for merchants
42:46who can afford it, I recommend, well,
42:48first, having their production in-house
42:49. You need to have the processes,
42:50otherwise, you're held at the throat by
42:51your service providers. That's why I'm
42:53happy to work with brands that can
42:54produce things themselves. But second
42:55thing, you shouldn't hesitate to have
42:57several agencies on the account if you
42:58can afford it, because they will
42:59challenge each other. It's very simple.
43:01Meaning at some point, you want to be
43:02the agency that performs better than
43:03the others. That way, your client is
43:04happy. If you only have one, well,
43:06there's no challenge. If you have two
43:08on the account evolving at the same
43:09time,
43:09>> ah, it kicks things into gear.
43:11>> The goal isn't to save as much money as
43:13possible in this field of creative, of,
43:15of advertising. Even if you have to not
43:17give a damn about spending money on
43:19that. The best, the best stack to have
43:21is five pods. You split your volume
43:24between several internal pods and then
43:25a stack of 5 or 6 agencies, which will
43:27be in another pod, but it's an agency
43:29pod with a point person for the
43:30agencies only who will make sure they
43:32all deliver well, etc. They will
43:33compete with each other to see which
43:35ones are actually kept, which ones we
43:36give more volume to, and so on.
43:38Actually, you distribute your volume
43:40based on the best agencies that
43:41delivered the product, and that's when
43:42you have a machine that is just
43:43stratospheric. Because you have, you
43:46have, you have creative internally, you
43:48have your partner agencies, so they
43:49will cost more but they will also have
43:51a kind of external creativity, and that
43:53allows you to do volume, volume, volume
43:55, volume, and really start to take over
43:57the market in a...
43:59>> pirate-like way.
44:00>> Yeah. Costing more, uh, in truth, you
44:02see the price of an agency, it depends
44:04on your agency, but costing more isn't
44:06always the case, eh.
44:08>> Because actually, a good creative
44:09strategist, you already need a
44:10qualified profile, that costs money,
44:11quite a bit. Let's say you're doing, uh
44:14, UGC, for example. Let's say we take a
44:16pack of 10. I actually did the math, I
44:18asked myself the question: you take
44:19your pack of 10, first off you're going
44:21to pay your creative strategist, let's
44:22say € 2,000 a month, which is an
44:24entry price for a well-qualified
44:25profile. You're going to want to
44:26produce 10 videos, so they will be able
44:27to make 30. Well, you divide your 2,000
44:29by 30 to know your cost per video for a
44:30creative strategist. After that, you're
44:32going to call on, for example, UGC
44:33creators, which is the easiest solution
44:34when you're working with an agency. UGC
44:35creators will charge you around € 150
44:38per video, you see. So, if we stick to
44:40150, your 10 videos cost you € 1,500.
44:42>> So the creative strategy cost you €
44:442,000. You need your editors and you
44:46pay your editors fairly. 10 videos,
44:48that costs you € 50 per video, €
44:50500. 2000, well 1500 and 500, that’s
44:53€ 2,000 for the editor plus the UGCs.
44:57And honestly, these are conservative
44:59prices. Yeah, but he won't really lower
45:00the price for you in general. I mean,
45:02it's hard for a qualified editor doing
45:04good work to go lower.
45:06>> No, they are all above that.
45:07>> Oh yeah.
45:08>> Yeah. I don't do UGC, so I don't think
45:10so. But from what I've done, I saw that
45:12it was still higher.
45:13>> I feel like at one point yes, but now
45:14AI is almost good enough to bypass that
45:16. That's
45:17>> OK, but you need a profile capable of
45:19doing that then. That makes for a
45:20second profile to hire. That's where
45:22it's important to have—the most
45:23important thing now is to hire AI
45:25editors who know this trade perfectly
45:26and are really, really, very good at it
45:28. That's it. That's where you can
45:30actually save on costs. But the point I
45:32want to make is that we, you see, a 10-
45:33video pack, we do it for € 4,000, you
45:34see, you no longer need to think about
45:36the script or the brief. Plus, we
45:37produce at a level of quality that a
45:39UGC creator isn't capable of. What's
45:40cool with agencies is that you just
45:41give the money and it's done.
45:43>> Yes. Yes. It's handled. Uh.
45:46>> But yeah, very interesting. Very
45:48interesting.
45:48>> There you go. So, the pod, it's good to
45:50think about it that way; the question
45:51is, what is the minimum staff I need to
45:53produce my videos? And actually, you
45:55think, if you have your UGCs, you
45:56remove the production level and you
45:58need a creative strategist and a few
45:59editors.
46:00>> And uh, AI, is that something you plan
46:01to implement in your agency or not for
46:03now?
46:05>> That's a good question. Actually, I
46:07implement in my agency whatever the
46:09market needs. The
46:11>> market.
46:11>> Because yeah, the market is king. If my
46:13clients tell me, if there’s a real
46:15need from my clients to generate AI
46:16content—we’ve already done it for
46:18the US market, for example—we’ll
46:20set it up, we’ll put processes in
46:21place. But my agency vision is: what
46:24can we do that brands can’t do
46:25themselves? Because otherwise, you have
46:28no added value. I mean, sure, there are
46:29some who can’t be bothered, so they
46:30outsource everything. To me, I think
46:32being an UGC agency...well, you can do
46:33UGC in-house with people who know how
46:35to do it quite easily. Being an AI
46:37agency, well, it’s the same; if
46:39it’s important to the brand, it would
46:41be more cost-effective for them to hire
46:44someone to handle it. So that’s why
46:47we focus more on real shoots and things
46:49that are perhaps more polished, which
46:51brands can’t produce themselves.
46:55>> Do you have phases where you think AI
46:56can help you find winning ads more
46:58easily, you know, for creating your
47:00scripts and so on? Because I personally
47:04believe in AI videos, in the sense that
47:06we can test many more at a lower cost,
47:09and as soon as one takes off, the
47:11agency should recreate it with a human.
47:14Is that a process you can implement?
47:16>> Yeah.
47:17>> Yeah, that’s a good question.
47:18>> Testing with AI and...
47:20>> more iterations, basically.
47:21>> Exactly. You see, the iterations...
47:24>> AI gives us a persona, a concept, three
47:27hooks,
47:29>> but we could take the 15 personas we
47:31have with AI, send them out to test
47:33them, or
47:34>> as soon as one takes off, we say, "Hey,
47:36the 55-year-old guy who’s like this,
47:38like this, like this—that’s a hit."
47:41Can you find him for me and do it for
47:42real?
47:42>> Yeah, that’s a good, that’s a good
47:44question.
47:45>> I would’ve stood up to shake your
47:46hands, guys, but I’m with you in
47:48spirit.
47:48>> But yeah, that’s actually it. Even
47:52beyond delivering finished creatives,
47:54if you deliver the brief, since your
47:56brief is complete and can be
47:57AI-generated, it allows the internal
47:59team to see how far you can go, you
48:01know, because you have to protect your
48:03business and so on, you see. But,
48:05actually, if you deliver your strategic
48:08creative briefs, they can be tested the
48:11next day using AI because we’ve
48:13developed our own tool. Basically, you
48:15input the brief and it outputs the
48:17creative in 40 minutes.
48:19>> And that's it. Yeah, that's
48:20>> really, the most important thing is to
48:22constantly, constantly, constantly have
48:23new briefs, new concepts, new things
48:25>> that are well-documented. For example,
48:27we want this environment, but you need
48:28to make sure the brief includes the
48:30environmental photo, the image, and so
48:31on, and then you can start doing
48:33something.
48:34>> Okay. Well, that's interesting if you
48:36have your own tool that helps save time
48:37. Because until recently—and actually
48:40, every market shift makes you change
48:41your mind—but until recently, I was
48:42saying: "Yeah, given the time it takes
48:44to actually make an AI video that's
48:45decent and that the market can accept,
48:47almost without noticing..."...
48:49>> you might as well do your own shoot,
48:51bring someone in, make 5 creatives,
48:53it's easy to organize, I mean it was
48:55simpler, and it costs 11 dollars.
48:57>> Yeah, I, I'm more, I'm not really
48:59aligned on that. For me, really, I
49:01think tomorrow's future, I mean if
49:03tomorrow I were at your place in your
49:05company, at Oh my Gods, I would tell
49:07you: "Actually, shoot your three, I
49:09mean shoot your video, make the three
49:11different hooks, and you end up with 15
49:13deliverables because you've reached
49:15everyone, and you deliver to the client
49:17and say:' Here, in the pack I'm giving
49:19you, there are three different hooks
49:21with one single persona. '""You have
49:25variants using AI." We test, and as
49:29soon as one character performs well,
49:31for the next batch,
49:31>> we know we'll do it for real.
49:33>> We know we'll do it for real with that
49:34one.
49:35>> Okay. So, the guy minimizes the risks
49:37because you, actually, you paid—look,
49:39you pay your editor, you pay, you've
49:41done your hooks, you agree with me—
49:42but then, delivering 15 different
49:44personas didn't cost you any more. It's
49:47not exponential, quite the contrary.
49:49The script is done, everything is done.
49:51>> You just have to change the guy's face
49:52and maybe change his environment, you
49:54see. But the script and the hook are
49:56the same. Well, I think that would be a
49:58winner tomorrow.
49:59>> Yeah, yeah, for sure. No.
50:01>> It fits, it fits, it fits perfectly
50:02with how the market is evolving anyway,
50:04because we're entering an era where AI
50:06will be so realistic that shooting a
50:07video will be exactly the same as using
50:09AI.
50:09>> Yeah, absolutely.
50:10>> We're talking about six months from now
50:11. We're not talking about ten years
50:12from now. You know?
50:13>> In six months, videos will be the same
50:15whether they're human-made or AI-made.
50:17It's
50:17>> actually, for me, almost like a
50:18creative agency. The real value it will
50:22provide is well, the brief, but
50:26everything surrounding the brief so
50:27that AI generations are done perfectly.
50:30That means, okay, we've found a new
50:31concept, we're going to test this
50:33character, this environment, we'll put
50:34in exact photos of the environment.
50:36We'll put in exact photos of the
50:38persona. We'll provide this deliverable
50:39to clients or we'll generate the videos
50:40ourselves. In fact, this client
50:42deliverable—well, more than ever, a
50:44creative strategy team, a creative
50:46strategy agency, will be about creative
50:48strategy rather than production. Behind
50:50that, production, to me, will gradually
50:52take a back seat. Right now it's not
50:54100%possible because the market still
50:56needs a human touch, and in any case,
50:57we'll always need humans in the sense
50:59that it's important to test things to
51:01constantly see the differences. But we
51:03are entering a world where your job
51:06will be 100%creative, 100%about
51:08implementing new ideas. And for...
51:11Because in fact, AI, down to the
51:13millimeter, captures the idea your
51:15creative strategist had—what you were
51:17saying earlier, the production that the
51:19strategist had in mind—the strategist
51:21can do exactly what they want, down to
51:23the pixel.
51:24>> That's true. Yeah.
51:26>> And that is a huge change in e-commerce
51:28, and that's where creative strategists
51:29will really become the emperors, in a
51:31way, of e-commerce, because everything
51:33they say can be applied down to the
51:35millimeter and tested the day after the
51:36brainstorming. I mean, you see, it's
51:39wild. You really need to manage to
51:41align yourself with this because very
51:42few agencies will take this approach
51:44100%. But if it's done well, a brief is
51:47worth its weight in gold. A brief—
51:48it's the brief and everything, the
51:50whole concept with good prompting for
51:51the AI, and so on. Good, proper
51:53instructions. This is going to be worth
51:55a whole lot, and people are going to
51:56want to buy it.
51:57>> Yeah, that's clear. I'm going to add
51:58some associates there. No, because
52:00honestly, in this case, I'll take, I'll
52:02take 10%. Well, me too then.
52:04>> Starting out, if there are no dividends
52:06, shake on it for now guys, I haven't
52:08said yes yet.
52:09>> And if there are no dividends, it’s
52:1120,000 minimum each.
52:14>> We'll see. No, but the truth is, the
52:16nuance is that today, for me, it’s
52:18not quite up to par considering the
52:20energy you have to put in to get a
52:22decent result. So we don't bother with
52:24it, really. That's how you fall behind.
52:26The day it is up to par, you'll have
52:28too much to catch up on.
52:29>> Actually, I'll give you something that
52:31Alex is doing for us today, you see. At
52:33Fincut. I don't have time to dig into
52:36AI anymore, even though I used to, you
52:38know. But on the other hand, we know we
52:40have Alex who is focused on that, you
52:42see. Indeed, it will take you a week
52:45and it's sorted. The thing is, it took
52:48you a week, but everything after that
52:49is sorted. We had issues with our
52:51creatives; we were struggling to
52:53provide them. Alex dug into it, and now
52:55we're able to churn out 1,000, maybe
52:57even 2,000. If you really push it, you
52:58can get up to 5,000 creatives a week.
53:00It makes no sense. But today, he's put
53:02out something that meant he had to
53:03isolate himself to get it done. But
53:05once it's done, it's done. You see,
53:07typically how to plug in all the
53:09different AIs that will allow you to
53:11make all your new iterations. I mean,
53:13you know, your whole process, what will
53:15it take? Come on, a solid week, maybe
53:1710 days. But once it's done, that's it,
53:19you can deliver your three creatives
53:22with a human and the 15 iterations of
53:24different personas. Why? Because you've
53:27understood that the process is a 360-
53:29degree thing, that you need images so
53:31the AI can handle it well, and so on.
53:33But you absolutely have to—I mean,
53:34for me, I think you absolutely have to
53:35get into it because everything is
53:37moving so fast. Yeah, that's true.
53:38>> And we need more—I mean, we are being
53:39called upon much more. It means that
53:41any idea we drop or any business we
53:43launch, someone is going to come along
53:45and copy it, and the problem is that AI
53:46will allow them to copy it a thousand
53:48times over, you see.
53:49>> Well, actually, we need to be even more
53:51agile regarding this, and we really
53:52need to have people who can support us
53:54with that part as well. Yeah,
53:56absolutely. Your role as a co-founder
53:58is really to step out of your business
54:00a bit so you understand exactly where
54:02you're headed and how to be constantly,
54:04constantly aware of that. That's kind
54:06of what happened at Fincut, you see, I
54:08100%stepped out of operations myself.
54:09>> To try and see a bit more of what's
54:11happening in the market, because
54:12actually, a team that's in the weeds
54:14operationally, even if you try to dig
54:15and so on or take the time, you'll
54:17never really be able to put 100%of your
54:18focus on it. And you, I think your
54:21focus needs to be on what's happening
54:22tomorrow. I'm going to talk to the
54:24merchant. What are they telling me?
54:25What do I need to understand about the
54:28coming market so that I can maximize
54:31the direction of the agency? And yeah,
54:34that's something super important
54:35because if you're a classic agency,
54:36you're going to die. Clear.
54:37>> Just as we analyze our clients'
54:39friction and feedback, you need to have
54:40the same thing on the other side. And
54:42there, in a certain way, what we're
54:43telling you is that actually
54:46>> and that's what I tell everyone, a
54:48concept can work. It all depends on the
54:51persona if we hit the right persona.
54:54But the concept itself isn't
54:56necessarily bad,
54:58>> it just might not have worked with that
55:00persona, and actually, what we expect
55:02is to be able to take this concept that
55:04we thought of, that we found useful,
55:05and multiply it by 15 because we have
55:07these 15 personas. So there, in a way,
55:11you're hearing our friction on the
55:13client side, you see.
55:16>> You need to be able to respond to that,
55:18and you shouldn't view the people you
55:19recruit as, well, as heroic figures,
55:21you see. Actually, you just need to
55:25bring people on so that you can isolate
55:27yourself and find new solutions. You
55:29see, Alex stepped up, he said: "Okay,
55:31I've got this, there's Jean-Vincent,
55:32there's Romain, Valentin, everything is
55:34handled, I mean everything is
55:36functional." But, on the other hand, he
55:37clearly sees that we, with our heads
55:39down in the grind, there are things we
55:40hardly do anymore. And actually, it's
55:42Alex who comes to feed all these things
55:44back to us, and to
55:45>> get out of the business to, in a way,
55:48refuel it.
55:48>> So for you, it's the same. Just because
55:50you brought on an extra person to do
55:52your work doesn't mean that this person
55:54has to make you money. No, it's up to
55:56you to make sure you find the new ideas
55:58so your agency gets ahead of the others
56:00.
56:00>> Clearly. That's that's true. But look,
56:03how much time did it take you before
56:04you were able to step out of the
56:06operations? You see, there is still a
56:08period of time, and actually, the fear
56:10I have right now. Yeah, there's that
56:13already, and actually, I'm lucky that
56:15none of my clients, I mean you see when
56:17a client joins us, they're happy and
56:18they stay, find yourself a right-hand
56:20person, you
56:21>> Yeah, actually, a right-hand person,
56:22you don't just place them and save time
56:23. That's it, first you place them, then
56:25you cultivate them, then you also help
56:27them understand how each of the members
56:28already present in the team works. So
56:30you don't get out of operations just
56:31like that, but I agree with you and I
56:33think it's never you never
56:35>> Yeah. And the first step for now is
56:36delegating the operational parts to
56:38each of the teams.
56:39>> Yeah, but start the other way around.
56:40>> You would put the right-hand person in
56:41first. Start by thinking, "Damn,
56:43they're going to free up my time, I'll
56:44be able to do this," because right now,
56:46even in what you're saying, sorry,
56:47we're being a bit blunt, it's that
56:49you're just saying, "Yes, I need to
56:50hire them." That doesn't work like that
56:52. No, think about what it's going to
56:53bring you. You'll see that suddenly,
56:55when you redo your list starting with
56:57what the result of hiring someone is,
56:59in the end, you say, "Okay, I can
57:00onboard them in a week, I need to do
57:02this." And suddenly, you've solved your
57:03problems.
57:04>> Yeah. Yeah.
57:05>> OK. If you head in all these right
57:06directions, I really think you have the
57:07potential to become a stratospheric
57:08agency.
57:09>> Yeah, you're going to have to take them
57:10anyway. You have the passion. Yeah, you
57:12have, you have the passion, you have,
57:14you have the entourage, you have
57:15everything you need to truly be
57:16constantly ahead of the market. And
57:18precisely, you have the agility, okay,
57:19because you’re not a huge corporation
57:21; you have the agility of a small
57:22agency compared to your big competitors
57:24who are massive and for whom changing
57:26everything overnight is actually
57:27impossible, it’ll take months or
57:28years. And so, you still have that
57:30advantage of being able to be very
57:31agile. So, if you manage to jump on the
57:33AI train quickly and really offer a new
57:35proposal to the market, I'm sure there
57:36are plenty of...well, I have one in
57:38mind, we'll talk about it, but there
57:40are plenty of offers on the market that
57:41you can put in place, which are
57:43extremely winning, in a way, and that
57:44everyone will want to snatch up. But
57:46once again for me, I think the concept
57:48is going to be the most important thing
57:49, a concept extremely detailed down to
57:51the millimeter,
57:52>> it's endgame for the for the
57:54>> for e-merchants. We realize even
57:55ourselves at the level where we try to
57:57detail, or rather automate, production
57:59>> to a certain extent, we realize that it
58:01is always the starting point, the
58:02concept, that must be as precise as
58:03possible. Today, when we make a concept
58:05, we have so many parameters that we
58:07put in, that we put in
58:08>> in, how should I say, that we must
58:10analyze or that we must detail so that
58:11already when it gets to the client,
58:13it's clear enough. And in fact, it's
58:16just that the day we say, well, the
58:17client validated it, it goes to AI,
58:18well then you need even more context.
58:20That's what we were saying earlier, I
58:21think you have to realize how much AI
58:23needs context to produce something
58:24completely.
58:25>> That's why you need a very well-made,
58:27very well-thought-out, very
58:28well-scripted concept, every angle,
58:30every thing, because everything you did
58:32there, when you give it to the AI, your
58:34work is done.
58:36>> Actually, when I say, when we tell you
58:37it's really millimeter-perfect, it
58:39means everything must be perfectly
58:40described in the scene. The movement of
58:42the hair, everything, really, it’s
58:44100%of the thing, must be perfectly
58:46described so that it’s like a...well,
58:48the AI almost has nothing left to do.
58:51And actually, we know how he talks, we
58:52know the guy's emotion. I mean, even
58:55with ElevenLabs V4, you can make voices
58:57that are insane, everything is super
58:59credible. It’s about how you’ll
59:01manage to transcribe the emotion
59:02exactly.
59:03>> that your creator had and how much
59:05detail you manage to put into it, and
59:07that is
59:08>> And I think, you see, the best way to
59:10do it, the way I see it, for the pods:
59:11you make a parallel pod, you keep yours
59:13, you don’t shake up your processes
59:14too much, you don’t try to find a
59:15sale at the same time, and you make a
59:17pure IA pod where you put only the best
59:18in it, and their goal is to manage to
59:20find the process themselves because you
59:21don’t have it in-house.
59:23>> Well, that’s my job right now,
59:25creating that kind of pod. I’ll do an
59:27AI one first and then a real one, but
59:29for me, that’s game over after that.
59:32>> Yeah, totally. Actually, it’s the
59:34vision, it’s telling yourself
59:35you’re already 6 months ahead because
59:37it’s going to move fast. Even I, who
59:38produce video today, am capable of
59:41spotting AI. But sometimes, I get
59:42surprised at the beginning of a video.
59:44It’s happening more and more, right?
59:45>> Yeah. And so, it’s just that soon
59:47we’ll be able to see everything.
59:49>> Yes. But imagine how much you doubt.
59:51>> So that’s my
59:52>> imagine Jacqueline in
59:53Meurthe-et-Moselle,
59:55>> imagine Karim in Alsace.
59:57>> Yeah. Yeah, that’s clear. No, but
59:59there’s also this thing of being in a
1:00:02bubble; since we’re video people, we
1:00:04notice it, but a tiny grain of sand is
1:00:07nothing, you see.
1:00:08>> People don’t notice anymore. You see
1:00:10your aunt sending you a video of a cat
1:00:12working out, and she wonders if it’s
1:00:13real. You see,
1:00:15>> last time, she told me, "That cat is
1:00:16pretty strong though."
1:00:18>> No, but AI is already good enough to
1:00:21make sales
1:00:22>> just as well as classic UGC. So
1:00:24>> that's true
1:00:25>> and there are even some who make UGC
1:00:26that, well,
1:00:27>> I was recently sent some AI UGC videos
1:00:30>> French
1:00:31>> no, it’s US, stratospheric. There’s
1:00:35that too, every time I get fooled,
1:00:36it’s from the US, but it’s a matter
1:00:38of the image because even now with V4,
1:00:40the French voice is, it’s settled.
1:00:42>> OK, no problem.
1:00:44>> Well, at that point, I think the danger
1:00:45is falling into that creative agency
1:00:47arrogance of saying, "I do things the
1:00:48human way," so actually, um...
1:00:51>> That's where people get lost; they’ll
1:00:52just keep selling creatives for 500
1:00:54bucks and so on.
1:00:55>> In fact, the concept will be able to
1:00:56sell for a high price—for me, you can
1:00:58make a concept expensive. It’s a bit
1:01:00like on YouTube; I know there are
1:01:01agencies that specialize in the concept
1:01:03, in the creation of YouTube video
1:01:04concepts.
1:01:05>> That’s super interesting. Actually,
1:01:07the idea—what a human can really
1:01:09bring to the table—is what idea they
1:01:10manage to pull out of their brain, or
1:01:11with the help of AI, obviously. But how
1:01:14does the human manage to create this
1:01:15thing and then, go ahead, save it?
1:01:17Because a created concept—you can
1:01:19scale a mega-concept, because once your
1:01:21concept is created and you do it
1:01:22entirely in-house, then you automate it
1:01:25, you automate it with all your
1:01:27personas and all your stuff. It’s too
1:01:28simple. Yeah, totally.
1:01:29>> But the concept will be worth a lot. To
1:01:31me, it’s going to be the, uh, digital
1:01:34gold.
1:01:34>> That’s true.
1:01:35>> Whoever has the best concepts.
1:01:36>> I’m going to be printing gold here.
1:01:37I’m going to have to work on my
1:01:38concepts now. No,
1:01:39>> that’s already what they call
1:01:42Treasury bonds. There you go, the
1:01:44concept will be like a Treasury bond.
1:01:45>> OK.
1:01:46>> No, in any case, this is an interesting
1:01:47conversation.
1:01:48>> I’d actually like to ask the question
1:01:50, perhaps to bring value to every stage
1:01:52of an e-commerce merchant's growth. You
1:01:54, who experienced the zero stage at the
1:01:56beginning, how do you manage your
1:01:57creative production? What makes an
1:01:59e-commerce store go from 0 to 10, or
1:02:00let’s say 0 to 1?
1:02:02>> First of all, I arrived in an era that
1:02:04is quite different from today’s
1:02:06e-commerce, but for a beginner, I think
1:02:10AI is one of the worst things that
1:02:12could have happened to the market.
1:02:15Actually, it comes back to what we were
1:02:17saying earlier, but a beginner who uses
1:02:19AI—a beginner who unfortunately uses
1:02:21AI poorly, let's say—is a beginner
1:02:22who is going to have a lot, a lot of
1:02:24problems. To me, AI is something great
1:02:27for advanced users, but ultimately, I
1:02:29think it's more of a burden for a
1:02:31beginner because, precisely, the best
1:02:33leverage when going from 0 to 10 is you
1:02:35as a founder: you're going to make your
1:02:37own creatives, you're going to dig,
1:02:39you're going to create your own
1:02:41concepts. And actually, it's super
1:02:44simple to do a "zero to ten," for
1:02:46example: you take your product, look at
1:02:48all the personas your product
1:02:49corresponds to, look at all the
1:02:51marketing angles per persona, all the
1:02:53pain points for each persona, and you
1:02:55just go and create three to five
1:02:56creatives per pain point, per persona,
1:02:58per marketing angle, and just go create
1:03:00through pure logic of mass on all that,
1:03:02by doing it yourself. You can get help,
1:03:05you can use tools obviously, but to do
1:03:07it yourself, you're really going to sit
1:03:09at a table and take a sheet of paper.
1:03:11Okay, I have this product, persona 1,
1:03:13persona 4, persona 5, pain point 1,
1:03:15pain point 2, marketing angle, it's
1:03:18really just normal, you see. There's no
1:03:20need to overcomplicate everything, and
1:03:22that's what you're going to focus on.
1:03:24You're going to ask your loved ones to
1:03:25make videos for you. You're going to
1:03:26ask your cousins. You're going to ask
1:03:28whoever. If you have concepts you want
1:03:30to test with AI. Obviously, you can do
1:03:32it with AI because, obviously, it saves
1:03:33time. But you remain the human who
1:03:35manages 100%of what you do, and you're
1:03:37going to duplicate that in volume
1:03:38across different personas and different
1:03:40marketing angles. To me, e-commerce
1:03:42doesn't change. E-commerce is always,
1:03:44and will always be, a business of
1:03:45fundamentals. AI will propel you, it
1:03:48will help you, etc., but it remains
1:03:49about the fundamentals. That means what
1:03:52worked 10 years ago is what works now,
1:03:53and what will work in 10 years. The
1:03:55fundamentals of e-commerce, you sit
1:03:57down, it's a story of personas,
1:03:58marketing angles, and pain points.
1:04:00E-commerce isn't anything more than
1:04:01that.
1:04:02>> OK. So, concretely, the steps for you,
1:04:04for the guy who wants to go from 0 to
1:04:0510k, is to understand his market. So
1:04:08the persona isn't just about AI, but
1:04:09about asking the right questions,
1:04:11looking at comments, doing the research
1:04:12, and understanding it well enough to
1:04:14find relevant marketing angles, imagine
1:04:16the creative, like a script, and then
1:04:17move on to production yourself. You
1:04:20don't sell something you haven't
1:04:21understood, right?
1:04:21>> Exactly. That’s why, in our club,
1:04:24>> there's Part 1, which covers the
1:04:26fundamentals. You learn the classic
1:04:27e-commerce fundamentals without AI
1:04:29being mentioned once. OK. And then in
1:04:32Part 2, or in a separate section, we
1:04:33explain how to automate things with AI
1:04:35afterwards. But the most important
1:04:37thing is that people understand the
1:04:39basics first before worrying about AI.
1:04:40With AI, we’ll explain how to do it,
1:04:42we’ll show you everything, but first
1:04:43come and understand the fundamentals
1:04:45and go step by step, especially if
1:04:46you're a beginner.
1:04:47>> OK, let's continue with the e-commerce
1:04:49evolution, because after going from,
1:04:50say, 10k to 50k, I think once you’ve
1:04:52hit 10k, you’ve found your
1:04:53fundamentals. Now it's about pushing,
1:04:55it's a matter of volume.
1:04:56>> Then diversity.
1:04:57>> Yeah. Diversity.
1:04:58>> Diversity of a team that will actually
1:05:01be able to
1:05:02>> structuring, then.
1:05:02>> That’s your goal once you've passed
1:05:0410k: it's really about structuring your
1:05:07brand to generate maximum volume and,
1:05:09in fact, being able to constantly
1:05:10deliver. Meaning you’ll start hiring
1:05:13different agencies and putting people
1:05:15in-house to constantly ship volume. But
1:05:17from 10 to 50k, or even 100k, it's just
1:05:19a story of volume. A story of volume in
1:05:22terms of personas, volume in terms of
1:05:23creatives, in terms of concepts,
1:05:24whatever you want, but it's just volume
1:05:26. It's how you manage to do everything
1:05:28you were doing alone, times 100, times
1:05:291000, and that's how you quickly reach
1:05:31100k.
1:05:32>> Nothing more to it. And then
1:05:33>> even if you had to start over, would
1:05:35you hire an agency on the side?
1:05:37>> Yeah, of course.
1:05:37>> And you would still structure your
1:05:39setup in-house?
1:05:39>> Yeah, but for me, agencies are a
1:05:41win-win.
1:05:42>> I mean, you make 10k, you have some
1:05:44money, well, go ahead and give 10 or 20
1:05:45grand to an agency that will handle all
1:05:47your creatives, save you time on
1:05:48concepts you wouldn't necessarily have
1:05:50thought of, and that you can then
1:05:51duplicate in-house or even execute
1:05:53yourself.
1:05:54>> Yeah. And agencies are often better
1:05:55than you are, it has to be said, you
1:05:56know.
1:05:56>> Yes. And besides, it allows you to
1:05:58learn; if you hire a contractor, ask
1:05:59them questions, take advantage of
1:06:01having them there to say "how do you do
1:06:02that, why do you do that?" And that is
1:06:04what you should expect from your agency
1:06:06.
1:06:06>> A good point, because unfortunately 95%
1:06:08of people in e-commerce rely solely on
1:06:10agencies; they let the agencies do the
1:06:12work and don't educate themselves at
1:06:13all. That's precisely where it's a bit
1:06:15more advanced, because he really
1:06:17educates himself, whereas people just
1:06:18trust agencies; nothing happens, they
1:06:20understand nothing, they just send it
1:06:22to the agencies and it gets the job
1:06:23done. Since it creates volume, since
1:06:25it's somewhat qualitative, etcetera,
1:06:27well, it works. But as a result, you
1:06:28don't really understand your business.
1:06:29It's a bit like delegating to agencies,
1:06:31>> it's essentially letting an outsider
1:06:32run your business without you
1:06:34understanding it, and that is a problem
1:06:36.
1:06:36>> That's clear. But I think for Médi
1:06:38it's different because since he has
1:06:39decades of experience, he didn't really
1:06:41try to understand us as much as he—
1:06:43wait, did he say dinosaur?
1:06:44>> You should put a background in like
1:06:45that, with dinosaurs and everything.
1:06:47>> Look at his great, great age there.
1:06:49>> His great age, I didn't say that. Oh,
1:06:52here we go, here we go. It's
1:06:54>> Médi started, he started in art
1:06:56direction. I was learning addition, you
1:06:58know.
1:06:59>> But why is your thing different with
1:07:02>> because, because since he arrived with
1:07:04his experience, he didn't try to
1:07:05understand what we were doing as much
1:07:07as he tried to help us improve. He
1:07:09already had insights on what we could
1:07:10do better. And so I think at this stage
1:07:12, uh, well, that's how he arrived. Yeah
1:07:13, it's simply that. He brought us
1:07:15insights on how we can better structure
1:07:17things, think better, and what we can
1:07:19do differently from other agencies, for
1:07:21example, you see.
1:07:22>> So, uh, so that was good for that. But
1:07:24there you go, if you don't have a "
1:07:25dinosaur" like Médi in your teams, uh,
1:07:28take advantage of your contractors to
1:07:30ask them questions. That's why it's
1:07:32lucky you're among us. That way, I can,
1:07:33I can roast him. You see, you protect
1:07:35me.
1:07:36>> I like, I like Alex a lot. No, but
1:07:41really, you always have to try to
1:07:42understand what the agencies are doing.
1:07:43>> Always understand
1:07:45>> well, how it really works, whether
1:07:46you're a beginner, advanced,
1:07:47intermediate, whatever you are, that's
1:07:48it.
1:07:49>> And now, the next level, and I think
1:07:51this is the most important part for our
1:07:53discussion, is to ask: you talk about
1:07:55brand, you talk about vision, what
1:07:57makes an e-commerce business transition
1:07:59into a brand? What even is a brand to
1:08:00you?
1:08:02>> I think the best person to talk about
1:08:03this would be our
1:08:06>> Actually, for me, a brand is when you
1:08:08are able to sell without running ads.
1:08:11That's a first step. And for me, a
1:08:14brand is an identity that you manage to
1:08:16broadcast repeatedly without stopping
1:08:18so that it becomes a brand. It’s the
1:08:22simple repetition of a thing, a value,
1:08:25a product, something you can reiterate
1:08:28perpetually, all the time, you see? And
1:08:32that’s when you become a brand.
1:08:34>> Yeah. It's really about detaching
1:08:35yourself from the "spend one, make two"
1:08:37model.
1:08:38>> Yeah,
1:08:38>> that’s the big problem. And actually,
1:08:40that is our challenge every day. It’s
1:08:42>> little by little reducing the
1:08:44percentage of ad spend relative to
1:08:46revenue, and when you start reaching 25
1:08:48%or 20%in ads compared to the revenue
1:08:50generated, that’s when you know
1:08:51you’re starting to become a, a brand,
1:08:54sorry.
1:08:54>> And it really means detaching yourself,
1:08:56meaning your brand lives and
1:08:57word-of-mouth is working really well.
1:09:00Uh, people think of you directly when
1:09:02they think of a product. uh, being
1:09:03truly assimilated by people. OK. Like,
1:09:06for example, I'll take Uniqlo; I need
1:09:08new clothes, I go to Uniqlo—it’s
1:09:09logical for people. And there you have
1:09:12something that works well.
1:09:13>> For them, it's the repetition of
1:09:15quality.
1:09:15>> Yeah.
1:09:15>> OK.
1:09:16>> You see, from time to time there might
1:09:17be some cuts, but it's the repetition
1:09:19of quality. And it's hammered home so
1:09:20much, it's repeated so much that if
1:09:22tomorrow you want to buy at a
1:09:23reasonable price, knowing the fit is
1:09:25good and the quality is good, I’ll go
1:09:26to Uniqlo. In fact, they simply
1:09:28repeated their process indefinitely so
1:09:30it would stick in people's minds. So,
1:09:32you recognize Uniqlo just like that.
1:09:34>> So, in fact, it’s an association game
1:09:36.
1:09:38>> Your brand is associated with a value,
1:09:39an idea; we took the angle of being
1:09:41associated with a fit. You need the
1:09:44best fit on the market. Finally,
1:09:45that’s what we try to hammer home,
1:09:46that’s what we try to make people
1:09:48feel.
1:09:48>> OK. And so, regarding your definition
1:09:49of a brand, would you agree with Amy,
1:09:51or would you have a nuance? Yeah, I
1:09:52would, I actually relatively agree with
1:09:55Amy, honestly. To me, that’s exactly
1:09:57what a brand is: a company that manages
1:09:59to survive without real paid marketing
1:10:01and without constantly having to push
1:10:03hard just to make money. Uh, a brand is
1:10:06, well, this might sound a bit silly,
1:10:08and it’s not necessarily true, but
1:10:10it’s something you’re proud to see,
1:10:12proud to wear, proud to buy.
1:10:14>> It works.
1:10:15>> Like, you know, it's
1:10:17>> For example, Supreme, you know, when
1:10:18you’re young, you love going to buy
1:10:19Supreme, you’re like, "Ah, damn, you
1:10:20see?" And trying to create an emotion
1:10:22for the consumer when they make a
1:10:23purchase, that’s when you know
1:10:24you’re becoming a brand, but a brand
1:10:26is everything, you see. A brand is how
1:10:28much you satisfy your customers. A
1:10:30brand is how much you’re recognized
1:10:32on the street. A brand is to what
1:10:34extent, you know, I think Fincut will
1:10:35become a brand one day too, and it’s
1:10:37starting little by little when a guy
1:10:39with a well-cut t-shirt on the street,
1:10:40people actually think it’s a Fincut.
1:10:43>> And for us, we don’t have a logo, so
1:10:45that could hurt us in the short term.
1:10:48>> But in the long term, it could really
1:10:50serve us in the sense that, if the guy
1:10:51just has a well-cut t-shirt and he
1:10:52thinks it’s Fincut, we’ve won. And,
1:10:54in fact, it’s happened several times
1:10:55where I saw guys saying, "Oh, do you
1:10:57think that's a Fincut?" Actually,
1:10:58it’s great, meaning if we manage to
1:10:59associate ourselves with every well-cut
1:11:00t-shirt in the world.
1:11:01>> OK.
1:11:02>> Man, you know.
1:11:04>> So, your vision for moving Fincut from
1:11:05an e-commerce site looking for
1:11:07profitability to a brand was to ask
1:11:09yourself, what is the thing we stand
1:11:10for? It's the fit, let's go with that.
1:11:12So now today, you think about your
1:11:14products, new launches, and so on,
1:11:16always with that approach in mind.
1:11:18>> And a fit can be whatever fit you want.
1:11:20For example, we launched the Boxy.
1:11:21>> And it’s going to be the best Boxy
1:11:23fit possible. OK.
1:11:23>> Actually, we want to offer the best
1:11:25fits on the market at Fincut, no matter
1:11:27the type of fit you like.
1:11:29>> OK.
1:11:29>> So, the best pants fit. If you like it
1:11:31oversized. I personally like oversized
1:11:33fits, the best oversized fit on the
1:11:34market with the best materials, and so
1:11:35on.
1:11:36>> Actually, we just want to differentiate
1:11:37ourselves for that.
1:11:38>> And also, what I want to say is that we
1:11:41often forget a bit, because we often
1:11:43think when we say brand and branding,
1:11:45you know, it's always the color and all
1:11:47that; but responding to an issue in
1:11:49under 48 hours via customer support is
1:11:51branding.
1:11:53>> OK. All those are also real touchpoints
1:11:55,
1:11:55>> touchpoints for the client, for the
1:11:57consumer.
1:11:57>> Yeah,
1:11:58>> that is the brand.
1:11:59>> I'll give you an example because I
1:12:01talked about it recently. I am of
1:12:04Tunisian origin. I take Tunisair to go
1:12:06back home. You see, Tunisair is a brand
1:12:09that is identifiable. It's that white,
1:12:12that red, it's the flag, you see. Hm.
1:12:15>> But on the other hand, the support, the
1:12:18customer service, the customer
1:12:20experience is so lousy that my
1:12:22association with the Tunisair brand is
1:12:25delays,
1:12:26>> it's no refunds. For me, all of that is
1:12:28branding. And often people ask me, "Yes
1:12:30, but what would you do? What color?"
1:12:31>> First, offer the best customer
1:12:33experience, then repeat something
1:12:36regularly, one of your values, and
1:12:38that’s how you build your brand.
1:12:41>> The official definition of a brand is
1:12:42that a brand is a distinctive sign.
1:12:45Word, logo, slogan, sound, or visual
1:12:47that allows one to identify and
1:12:49differentiate a company's products or
1:12:51services from those of competitors.
1:12:54It's really that, which takes into
1:12:55account the commitment and all the
1:12:56points of...
1:12:57>> Supreme is about belonging.
1:12:59>> And exclusivity only reinforces that
1:13:02sense of belonging. They work on a
1:13:05sense of belonging.
1:13:06>> OK.
1:13:06>> And it's repeated, repeated, repeated,
1:13:08repeated, and so on.
1:13:09>> And what would you recommend then?
1:13:11Let's say an e-merchant, you know, who
1:13:12started in pirate mode, pure
1:13:14profitability. He had a product that
1:13:15worked well, he pushed the volume.
1:13:17Today, he says to himself, "Well, the
1:13:18thing is actually so serious that I
1:13:19want to establish it."
1:13:20>> Yeah.
1:13:20>> What are the questions to ask oneself,
1:13:21the steps to put in place to say, "I'm
1:13:23taking it from an e-com to a brand?"
1:13:24>> Well, that's exactly what happened with
1:13:26Finkout and honestly, with many of the
1:13:28brands from people in the V Club, you
1:13:30see that you have a product that works
1:13:32well. Actually, it's important to start
1:13:33like that. For me, start like a pirate,
1:13:35in dropshipping, breaking the rules a
1:13:37bit to see what happens in the market
1:13:38and how it reacts. And as soon as it's
1:13:40good, well, actually start to get
1:13:41professional. And the first steps are
1:13:43exactly what we said, well, it's
1:13:44optimizing your delivery times,
1:13:46improving your product. For me, the
1:13:47most important thing in e-commerce by
1:13:49far, no matter what else you do in your
1:13:51company, is the product, number one.
1:13:52The product is the thing.
1:13:53>> Actually, I would say it's the customer
1:13:54experience. The product is the
1:13:55translation of that customer experience
1:13:57.
1:13:57>> Yeah. But even beyond the customer
1:13:58experience, if you have a bit of a
1:14:00disastrous customer experience and you
1:14:01have a good product, it will still save
1:14:03your customer experience.
1:14:04>> OK?
1:14:05>> Because the product is the end point,
1:14:07the final thing. People buy your
1:14:09product and if the product solves your
1:14:11problem well, if the product delivers
1:14:12on what you bought, you've won. So, for
1:14:16me, the most important thing to go from
1:14:17a small dropshipper to a future brand
1:14:19is: work on your product, then work on
1:14:21your customer experience, make sure
1:14:23everything is fluid, everything is good
1:14:25, fast customer service response, fast
1:14:27delivery, anyway, everything you can
1:14:28improve regarding that, make sure it's
1:14:30done well, and also your exposure in
1:14:32the outside market. Is your Trustpilot
1:14:34good? Are the reviews good? Is the
1:14:37message people are saying about your
1:14:39brand positive or negative? The NPS,
1:14:41all that. In short, lots of little
1:14:43things to put in place. And then,
1:14:44there's marketing. Marketing is how you
1:14:46are going to become, you are going to
1:14:48create marketing that can engage people
1:14:50rather than just being there to sell.
1:14:53Meaning, you are going to stop only
1:14:54doing bottom-of-funnel where, in
1:14:55dropshipping, you only do
1:14:57bottom-of-funnel to hammer out sales,
1:14:58etc. No, now you are going to start
1:15:00thinking about powerful top-of-funnel
1:15:01and middle-of-funnel: how you are going
1:15:03to bring people into a story, how you
1:15:05are going to bring people into an
1:15:06adventure, and then you are going to
1:15:08start pushing this marketing more and
1:15:09more with an art director, people who
1:15:11take the brand image to try to enhance
1:15:12it. Then, you are going to get
1:15:15influencers to start building trust
1:15:16around the project. Actually, it's all
1:15:19these points that you're going to put
1:15:20in place little by little that will
1:15:22make you transition from a brand—well
1:15:23, a dropshipping site that's just a
1:15:25site for product X—
1:15:27>> to a brand that truly becomes, well,
1:15:28the brand for that product.
1:15:30>> Okay,
1:15:30>> that’s the difference, really. It's
1:15:32really about gradually breaking
1:15:33everything down and coming in to
1:15:34consolidate your brand's credibility.
1:15:36>> Yeah, there’s no single project, no
1:15:37single answer. Unfortunately, you have
1:15:38to pursue them all simultaneously.
1:15:40There are a lot of things, but to me,
1:15:41it seems quite logical. It's product
1:15:43first, then customer experience, and
1:15:45then, really, a level of belonging to
1:15:47the brand, ultimately. Okay.
1:15:48>> Project.
1:15:49>> I have a, I have a thing regarding this
1:15:51part where, if tomorrow you're asked to
1:15:53walk, to walk physically, you know how
1:15:55to walk. It's one step after another.
1:15:56Do you agree with me?
1:15:57>> Yeah.
1:15:58>> When you're a dropshipper, the idea is
1:15:59to move forward. So you're walking.
1:16:01Okay.
1:16:02>> When you start becoming an e-merchant,
1:16:04it's about understanding how the market
1:16:07works that transmits the information.
1:16:09It's the brain that sends the signal to
1:16:11the leg, to the nerve that sends it to
1:16:13the leg, which makes you move forward.
1:16:15It's a forward momentum.
1:16:17>> And when you're going to become a brand
1:16:19, it's about asking yourself, what is
1:16:21the best nerve that can make this work,
1:16:23make me run, make me advance? Do you
1:16:25get the point? It’s about digging a
1:16:28little deeper and taking something you
1:16:30do instinctively—because when you
1:16:31start making sales instinctively,
1:16:33you'll say, "Okay, it works like this,"
1:16:35but you don't really have the rule for
1:16:37why it worked. Actually, that's it.
1:16:39>> I didn't understand. Actually, I
1:16:41understood the idea, but for me, the
1:16:42way you presented it, I feel that the
1:16:44way I perceive it in my ChatGPT head,
1:16:46you see, is about aiming for efficiency
1:16:47. It's about saying, what's the
1:16:49shortest path to reach my goal?
1:16:50>> Yeah.
1:16:50>> And how does that help build a brand?
1:16:53Actually, you're going to understand
1:16:55why you have an idea, why you have a
1:16:57perception of what you're doing.
1:16:59>> In fact, you can also take the feedback
1:17:01>> that you have around your brand, which
1:17:03makes it so that this exact model is
1:17:06what allows you to be
1:17:07ultra-identifiable. Oh, right, okay.
1:17:09You see,
1:17:10>> sometimes the accident can come from
1:17:11elsewhere, and that's when you
1:17:12understand.
1:17:13>> Actually, what I understood is that you
1:17:15think: "Well, in the end, your shortest
1:17:17path means: what worked the best?" And
1:17:18so, that's what you can be associated
1:17:20with, and it's from finding this
1:17:21shortest path that you create your
1:17:22brand. That means, ultimately, what is
1:17:24your value proposition, you know.
1:17:25>> That's it.
1:17:25>> But sometimes you need to dissect it to
1:17:27understand it. Yeah,
1:17:28>> because sometimes it's so instinctive,
1:17:30my product works because it's that. But
1:17:31actually, when you start to dissect,
1:17:33dissect, dissect,
1:17:35>> you actually discover that it's for
1:17:36other reasons people identify you; it's
1:17:38not just the color of your logo.
1:17:41Sometimes it's really the quality of
1:17:42the fabric fiber of the thing that will
1:17:43make you known. That's why I talk, you
1:17:45know, about walking versus knowing how
1:17:47the human body functions to make you
1:17:49walk. It's the same result, but the
1:17:52exercise of understanding is much
1:17:54sharper, you see.
1:17:57>> OK. Interesting. Alright guys, I
1:17:59understand better now.
1:18:00>> Well friends, listen, it was great
1:18:02chatting with you. We've done a little
1:18:04over an hour of podcast regarding
1:18:06creative strategy. Frankly, it was very
1:18:08, very interesting. If you want to work
1:18:10with R1, obviously, don't hesitate. We
1:18:12provide that information right below,
1:18:14and don't hesitate to subscribe to the
1:18:16channel. Don't hesitate to leave a
1:18:18little like, share the videos, don't
1:18:20hesitate to subscribe to Am's Instagram
1:18:21right there, to my Instagram, to
1:18:23Ederan's Instagram. Don't hesitate to
1:18:25show some support, for real, truly. All
1:18:27that is most important for us is the
1:18:28love and the strength you'll give us in
1:18:30return because that's what allows us to
1:18:32evolve, that's what allows us to grow.
1:18:34Don't hesitate to also tell us what you
1:18:36thought of this podcast, how we could
1:18:38improve them, and if you want more in
1:18:40the future. And, and that's it. In any
1:18:42case, it's a great pleasure to have had
1:18:43you. Thank you gentlemen, thank you for
1:18:45your time.
1:18:45>> Very cool.
1:18:46>> And we'll see you in the next video.
1:18:48Ciao! Bye everyone.

About this transcript

This page contains the full transcript of Nos STRATÉGIES d'ADS chez Fincut : Podcast avec Hamdi et Erwan (CEO - Agence OhMyGads) by Alex Shane , generated from the public captions YouTube serves with the video. The transcript has 17,722 words across 2,755 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.