YT YouTube2Text

YouTube2Text → Transcripts → How I Scaled An App To $3M/Month With Meta Ads (Step by step)

How I Scaled An App To $3M/Month With Meta Ads (Step by step) — Transcript

by Lorenzo Pravatà · 6,757 words · 924 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00We scaled this up to $50 million in 10
0:03months and $3 million in one single
0:06month, all from a dad. And that's just
0:08the first purchase, not even counting
0:11subscription. In this video, I'll break
0:13down the exact funnel strategy and
0:15creative system that we use to scale
0:18apps up to $50 million a year. We manage
0:21ad accounts across multiple apps doing
0:24between 1 million and $50 million a
0:26year. So everything here is based on
0:28what we see working right now and not
0:31during. So in this video I'll cover the
0:34lessons from scaling an app with a quiz
0:36funnel to $3 million a month and how to
0:40build the perfect app funnel from the
0:42ads to the quiz, the unique mechanism,
0:44the upselles you can use and how to have
0:48a 180 days or longer LTV. You can use
0:52also different funnels. We have apps
0:53also with just install goals or free
0:57trials, but the quiz funnel is the one
0:59that we see working the best right now.
1:01Also some ads examples and the
1:04importance of the psychology behind the
1:06ads especially for app businesses, the
1:09creative strategy that we use and the
1:12ads that are working for us right now
1:14and how to scale it up to 8 figure. So
1:17this based on what we see working, we
1:19can't share the exact um business name
1:23exact ads. So I'll just take the
1:26learnings and our our strategy and show
1:29it to you. Here you can see also some of
1:31our ads for a client. Now our ads all
1:34have a different naming structure, but
1:36these are some from last year where we
1:38still the the owner still used my name.
1:41So you can see some of the results here
1:42and how much they generated. So in my
1:44opinion, the main levers you have to
1:46scale an app are one the offer. So you
1:49want to make sure that how you position
1:50your app, your solution, the audience is
1:54um like that's one of the most important
1:56things making sure it's you position
1:58that better than the competitors. Then
2:01you want to make sure that you can have
2:04very high especially LTV. Like if you
2:07can have a very high LTV that's going to
2:09allow you to scale this. The thing is if
2:12you have a high LTV but it takes you one
2:14year to get that you're spending more
2:17than what you're getting the first month
2:19then it's going to be a bit difficult to
2:21scale fast. So it's important ideally
2:23that you have a high average order value
2:25or at least you get a very high amount
2:27of money in the first month so you can
2:30get credit and spend more in that.
2:32Otherwise if you have a very high LTV in
2:34one year you're going to have to get
2:36some funding. Okay. something you have
2:37to consider. In general, you want to
2:39make sure you can pay a lot for that CPA
2:41on on meta. And yeah, you can try to
2:44spend less, but at certain point, you
2:46are going to have to spend a lot and
2:49have a very high CPA if you want to
2:51scale to $10 million, $50 million a
2:53year. Then the quiz, so you can have
2:55like separate like several type of um of
2:58funness. You don't have to have a quiz.
3:01So like we also have clients just have
3:03cost per style goals or they have a free
3:07trial with credit card. So you can have
3:09different funers. What we see working
3:11best right now for apps is a quiz.
3:14That's normally the fun that allow allow
3:17you to get the highest AOV and LTV. The
3:21other thing is doing apps. Like a lot of
3:24uh businesses don't use apps and um for
3:28like a lot of apps only have that app
3:30that app and that's all like they don't
3:32have anything else in the ecosystem. As
3:34you scale you're going to see that your
3:36CPA is also going to increase because
3:38because you get to more audience you
3:40have to convince them and you have a lot
3:42you start to have more competitors and
3:44so you want to have more upsets to
3:46increase the um your average order value
3:49when people buy. Then we're going to
3:51look at the creatives and uh the
3:53creatives are one of the most important
3:54part when you have the right funnel the
3:57right positioning and when it starts to
3:58work the creatives are going to be the
4:00main difference to really scale when
4:02like the quiz like the rest of the
4:04funnel is taking care
4:06it's already working and you need a
4:08process to create a lot of ads and a lot
4:09of winning app. So how apps are
4:12different compared to um to other type
4:15of businesses on beta. So for apps you
4:18don't have a physical product right? So
4:20it's very important to understand the
4:22psychology behind why people buy and uh
4:26you have to create the perceived
4:28transformation just through messaging
4:30and you have to make them visualize how
4:33they'll be in their future version after
4:36using the app. So you have to understand
4:38very well their emotions, their
4:40cognitive triggers, self perception. So
4:43the type of person they want to be. And
4:45that's why a lot of apps have a lot of
4:49different ads. They use psychological
4:51doors, let's say, like procrastination,
4:54ADHD, anxiety,
4:57um, dopamine, like all of these. You try
4:59to get something that triggers you to
5:03uh, to want to understand more about
5:06your problem and then um, introduce
5:09their unique unique mechanism. And yeah,
5:12in general I'm making ads for apps like
5:14for most apps. Um, like obviously like
5:17there are some apps that are different
5:19like games and u there are different
5:22niche but especially for the ones that
5:24are more direct response they really
5:26like that really test how good you are
5:29with direct response because you don't
5:31have a physical product you just have to
5:32work with the psychology. So as the
5:35messaging is very important you need to
5:36focus on understanding the right angle
5:38to use the right format. So how you are
5:41communicating that message, what type of
5:43ad you you use and then also the right
5:46funnel. So normally depending on
5:48different buyer personas, we see that
5:51different formats can work well. And so
5:53you have those like formats work well
5:55for um for Gen Z for like younger
5:59audiences like memes, POVs or like um
6:04also like some ads with fast edits. Then
6:07you have some uh like in general for
6:09problem aware audiences uh especially
6:12older direct response formats can work
6:14well. I'll show you some examples later.
6:16And um in general for older audiences so
6:20at least 45 years old or 50 60 years old
6:24or even older just video sales letters
6:27work pretty well. And for younger
6:29audiences, uh, I'll show you some
6:32examples also of just identity content.
6:35Um, that can work pretty well and just
6:38leveraging the the identity that a
6:40person has. And then I see too many
6:42strategies that are working for us. So
6:44one is try to sell the app. the less
6:47direct response and less saturated the
6:49app is, the easier it is to just sell it
6:52and just have a video 30 seconds to 1
6:55minute and you just explain why they
6:57execute and why they need it or you just
6:59you just explain the problem and the
7:01solution or you compare it to different
7:04uh alternatives. Then the other option
7:06is to sell the mechanism. You can have
7:09like some clickbait videos or images
7:12that just sent to the quiz. That's also
7:14an option or you cannot sell the
7:16mechanism. So basically you have a video
7:18maybe still like 30 seconds, 1 minute
7:21even multiple minutes like even three
7:24four minutes video but you just talk
7:26about the mechanism. So you just educate
7:28them about why uh they have that problem
7:32and how to solve it with this mechanism
7:34but you don't talk about just the the
7:36solution the the product let's say the
7:38app. you just talk about the mechanism
7:40and then on the uh landing page with the
7:44quiz after a quiz on the sales page or
7:47based on the like what they answered or
7:49with the VSSL can be at the beginning or
7:51it can be after the quiz um you can
7:54especially after a quiz you can
7:55introduce solution why because basically
7:57you are telling them that there is a way
8:00to solve that problem there is a
8:02mechanism there is a way to solve it and
8:05you don't give the exact solution until
8:08you know more about them because then in
8:10the quiz you get more information about
8:11them and then you can actually uh show
8:14the uh the the solution and sell it.
8:17Then something we see working pretty
8:18well there are like some of the formats
8:21that we do for example the fake podcast
8:24straight interviews authority figure ads
8:27uh ski conversations so there are a lot
8:29of formats that we do that are working
8:31pretty well now for for apps and these
8:33are based on what we see working on
8:36organic in general and I'll show you
8:37some examples later so another thing
8:39that is interesting for apps is that
8:42they require a lot of buyer personas uh
8:45and a lot of angles So in general like
8:48we see that before you could find like a
8:50very big winner. You can still do it for
8:53some niches but now with Andromeda we
8:56see more and more having like smaller
9:00ads that smaller winners. So instead of
9:03of getting one winner expense 1 million
9:05in one month maybe you got uh 10 ads 20
9:08ads uh 30 ads
9:11that together spend $1 million. And uh
9:14you want to understand well your buyer
9:16personas and the angles for them. For
9:19example, for um uh here's an example for
9:22an app where they have the buyer persona
9:24over our thinker stress student over one
9:27mom and they have like several different
9:29type of buyer personas they are talking
9:31to and u you basically want to start
9:35treating all of them as different uh
9:38almost as different businesses. As you
9:40scale and you basically have a business
9:42makes um even like let's say $1 million
9:46and you have maybe uh three buyer
9:48personas that are making like 300k each
9:52for you then you want to start treating
9:55them as different businesses and having
9:57separate concepts dedicated to them. And
9:59you can also use separate landing pages
10:01for them. That's something we use a lot
10:03in other niches. But for apps, what we
10:05see working better is having separate
10:07quiz uh separate quizzes and um that's
10:11why a quiz is much better than just
10:12sending to download the app because
10:14normally you can actually do some
10:16marketing. You can actually customize it
10:18to different buy persona. We normally do
10:20a sort of creative strategy map and we
10:22do that with all businesses and with
10:24apps we see it's especially important
10:27where you have one by persona and uh you
10:29have different awareness levels. For
10:31each awareness level, you have different
10:33angles like this. And then for each
10:35angle, you have different form. So then
10:37you start to see, okay, this angle work
10:39well. Let's double down on this and
10:41let's just do a lot of ads on this
10:43angle. And for this V persona, this
10:45angle and with this format, but then you
10:46see this angle didn't work well. Why
10:48didn't this work well? So maybe this is
10:51not the right angle. Like this is not
10:53big enough or like we're not good at
10:55communicating it. There are better
10:57competitors on this. So maybe we don't
10:59want to test it anymore. Or maybe we say
11:01actually I think this is a very good
11:03angle. I think the format we used is not
11:06the best format and then you can try to
11:08test uh something else and see what
11:10work. What this allows you is to have a
11:13very clear map what works, what doesn't
11:16work and what you still need to test.
11:17You can even do a similar map for
11:19competitors. I would actually recommend
11:21it and you see what the competitors are
11:24are going for and you can try to beat
11:26them at their own game or try to just
11:28target different type personas as you
11:31scale. You normally go want to go more
11:33unaware and the more unaware you go the
11:35bigger scale you have. It's not always
11:37possible to go completely unaware. For
11:39example, in weight loss going unaware
11:41it's going to be very difficult and it
11:44makes sense if you really make billions.
11:46Otherwise, a lot of times you just want
11:48to go problem aware or even solution
11:50aware. But in general, we see that we
11:52are moving more to the unaware audience
11:55as we scale. What happens as you scale
11:58is that you start to have something like
11:59this. You don't have anymore a map like
12:02this. You have several ones like this.
12:04Similar to what I was saying before
12:05where basically you treat each buyer
12:07persona as a different business
12:09basically. And instead of saying okay
12:11now for this buy persona I'll do uh 10
12:14concepts a week or 10 concepts a month
12:16and I distribute them between um these
12:19awareness status or these angles um
12:22these formats instead you sort of say
12:24okay I'll do 10 for this 10 for this 10
12:26for this and 10 for this and so it's a
12:29much higher scale but it makes sense
12:31because as you scale you also need to
12:33increase your input and do more. So in
12:35general when when you do that you don't
12:38also like you don't need to innovate too
12:41much in terms of okay let's to like
12:44let's customize a lot the app for this
12:46bio persona and then for this and for
12:47this like you can do that but a lot of
12:49times you just have to to add new
12:51mechanism. So you have a mechanism that
12:54is perfect for one by persona and
12:56another for another by persona. So yeah,
12:58you can add new features, but what
13:00matters the most is making sure that you
13:03are uh explaining them what's the right
13:05like what's the right way to solve their
13:07problem and it's customized to them. And
13:09the quiz is very good for that because
13:11you just have to create some a couple of
13:13sales pages with different explanations
13:15based on the main profiles and main buy
13:18personas that you find. And um yeah for
13:20example uh for relatio um actually yeah
13:24I think the example here was relatio uh
13:27they have something like dopamine detox
13:30plan pro procrastination triggers test
13:33uh like anxiety management pathway. So
13:35like they have all these things that are
13:38different based on the buyer persona and
13:41the trigger they use to get you to uh to
13:44the quiz and what you answer on the
13:46quiz. So basically you have only one
13:49product that can solve different
13:50problems but you're able to market it to
13:52different buyer personas and in general
13:55like you want people to see your ads do
13:58the quiz and see the solution and say oh
14:02wow this will actually help me to become
14:05the person I want to be. That's what you
14:07are trying to do right I'll show you a
14:09couple of examples. Unfortunately I
14:10can't show you the examples and the data
14:13at the same time of our clients. I'll
14:16show you some examples of some ads we
14:19did for a couple of app clients. But
14:22yeah, here I'll just uh go through a
14:24couple of examples. I'll take some
14:25statics first just because it's a bit
14:27easier to to show them. So uh this um I
14:31think this relation the um the app and
14:35they use here a mechanismdriven ad and
14:39they have dopamine detox angle and they
14:42basically introduce a sciencebased
14:45mechanism the dopamine disreg
14:48disregulation and they want to basically
14:51tie it to everyday symptoms overthinking
14:53lack of motivation anxiety destruction
14:56then they use a quiz funant and They
14:59make sure that once you see the sales
15:01page after a quiz that the sales page is
15:03customized on what you answer and people
15:06say okay this app understands and in
15:09general this market is skeptical and uh
15:12we see that when you do ads that are
15:14very broad they don't convert well. So
15:16you have to understand the buyer
15:18persona. So you do that in the ad and
15:20with the quiz and the customer sales
15:22page in this case like basically what
15:24they're doing is they are talking about
15:27the mechanism they're not talking about
15:29the solution on the app for now and also
15:32like that's something works pretty well
15:33they talk about the real cause and they
15:35give you a justification so it's not
15:38lack of discipline it's not you it's
15:40your dopamine baseline so it's like it's
15:43not you it's this and the mechanism is
15:46basically um reframing the product from
15:50just an app to actually like a custom
15:52plan that works for you. This is another
15:54example of an ad about the
15:56procrastination and you can see here
15:58this is how the brain of procrastinator
16:00functions compared to high achiever. You
16:02see like classic they scale a lot on Tik
16:05Tok too. So you see um these type of ads
16:07just some music and text. So you see
16:10like all the things that are common they
16:12see here and then they say the first
16:14step is discovering your unique
16:16procrastination diet and yeah they just
16:18talk quickly about the quiz and then
16:20they send you to the quiz. So uh
16:23basically they target problem aware
16:24users and so you have to understand why
16:26the awareness level as we were saying
16:28before where uh you want to understand
16:30the awareness level you're speaking to
16:31and sophistication too and this case
16:34it's they understood it's a problem
16:36where user that procrastinates but
16:39doesn't know why. So basically the ad
16:41educates of why that happens, the guild
16:45cycles, the trigger loops and um and
16:48then they basically tell you to do the
16:49quiz because the quiz will reveal you
16:51your procrastination type and then they
16:53also talk about solution but it's uh
16:56especially like using curiosity to bring
16:59you to the quiz. These are some other
17:01examples of like this um this app uh I
17:05think it's called Leven and they are
17:07scaling a lot on on meta and Tik Tok and
17:10in this case you have some ads are
17:12pretty different but something that is
17:14common in general is that for apps we
17:16see that some like viral formats it can
17:19be videos can be images normally videos
17:21scale more but also images that are
17:24similar to what you see on organic and
17:26just um like for example in this case
17:28they use the sense of identity
17:30So uh in this case uh it's the
17:33discipline, strength, focus and you see
17:36like this um this type of angle. So
17:39don't be a nice guy, be a king. And uh
17:42in general um for these type of ads, it
17:45try to leverage more the identity
17:47transformation, who they want to be and
17:50not and not science uh not really
17:53science. especially for like for young
17:55audiences, young men that respond well
17:59to aspiration and identity instead of a
18:02long mechanism of how that works, the
18:05science behind it and everything
18:06everything else. And these are um are
18:08very similar to what works on organic
18:11and you can just take ideas from organic
18:13and try them on paid. So this is a
18:15funnel for um for corive. I actually
18:19took this one from funnel of the week. I
18:21highly recommend it if you want to see
18:23some funness. They break down the
18:25funness very well and instead of having
18:28to break it down again, I took all they
18:30did that is a great breakdown. So
18:33basically um is this very similar to a
18:36lot of apps we work with and as I can
18:38show I can show you an example of one of
18:41our clients. I'll show you this. So you
18:43see an ad like this, right? You see an
18:45ad like this on Instagram,
18:48Facebook, Tik Tok. Then you go to a quiz
18:51page and you have um quiz page like this
18:54on your phone where you can you can
18:56start to answer questions. Something
18:58that you'll notice is that be between
19:00questions you'll have social proof and
19:02in this case two questions then they
19:04show this then you continue then main
19:07goal in this case this uh basically to
19:10help you make more money. So um it ask
19:13you the income source yeah work schedule
19:18challenges and yeah do our best to help
19:20you. So you see there is some
19:22communication in the middle another
19:23question then yeah stabilities within
19:25reach here. Yeah something is common too
19:28like in some quizzes you'll see that
19:30here it looks like they say they say the
19:32actual questions sometimes you have
19:34maybe the bar but is more closer to the
19:37end than what it actually is. And same
19:39for VSSL videos. It goes faster and then
19:42actually either restarts or it starts to
19:44go slower. And that allows you to
19:46basically think that you are closer to
19:48the end and you just have a couple of
19:50questions or a couple of seconds to
19:51watch and then actually there is more
19:54but like that allows allows these
19:57companies to retain you for more time.
19:59Okay. So here you see something like
20:01this. And I mean here they don't you
20:03have like a lot of social proof but
20:06normally in the quiz you're going to see
20:08a lot of social proof. In this case,
20:11yeah, they also ask you about the time
20:13you you can invest on this and then they
20:15ask you the email with our clients.
20:17Something common to is to use in the
20:19quiz a lot of social proof. So you have
20:22several parts between a question and the
20:24other where you show a lot of people
20:26have been able to achieve this goal with
20:28this app and you can show like some uh
20:31testimonial quotes and also like a short
20:34video in the middle between one question
20:36and the other and that allows you to
20:38just get people um and also ideally
20:41people similar to their situation. So
20:43closer to the end you show some people
20:44that are more similar to them and
20:46achieve their end goal. And uh this in
20:49the quiz between one question the other
20:51helps to convert people better. Here
20:53they ask about the email and then yeah
20:55your um your growth challenge is is
20:59ready. Yeah. And then you go to the
21:01sales page. The sales page is customized
21:04for um each buyer persona. At the
21:07beginning if you have smaller app you
21:09probably have just a couple of options
21:11for the sales page. And then as you
21:13scale so much, you start to have more
21:15and more different sales pages. And
21:18anyway, like those are very similar. So
21:20there are just a couple of
21:21customization. Normally you have a timer
21:23like this, 10 minutes or something like
21:24this for the discount. So you basically
21:27have urgency. The urgency makes sense
21:28for them. You can see here like those
21:31scores and then it's very common to have
21:33something like this. So you see here
21:35it's nothing, right? And actually like
21:37in this case it's not that much. uh 61
21:411939.
21:42A lot of times you can see these like
21:44very low because it's per day, but then
21:47it's actually like three three months
21:49minimum. And so uh basically you have to
21:52spend at least $50,
21:55$100, something like that. So it's very
21:56common they try to make you spend more.
21:59Um also it's common to have like the
22:01free trial and then they charge you just
22:03after the free trial, but they charge
22:05you for a year or for three months or
22:07something like that. money back
22:08guarantee. Money back guarantee is
22:10normally pretty important. Also like to
22:12push them to actually use the app
22:14because maybe you say okay the money
22:15back guarantee is 30 days, 60 days, 90
22:18days, but um you can connect it to the
22:21fact that they have to use it for that
22:23time. Yeah. Then you check out and
22:25something very interesting here is the
22:27upselles too. So you have all these
22:29upselles this and here is $44 more
22:33upsell two post screen. um here. Okay.
22:37Didn't have another upset here, but you
22:39can see it's already two upsets here.
22:41Misconception about this is that you
22:44maybe think that you have to have most
22:48people buying the apps. It doesn't have
22:50to be like that. So, it depends on the
22:52app you have and the product. So, you
22:55have two main options. you either have
22:58some absesses with normal prices maybe
23:00up to $100 and uh decent percentage of
23:04people will buy it and that increases
23:06your AOV and uh even if just 10% of
23:09people buy it or you can have a even
23:11like very expensive upselles and a very
23:15small percentage of people will buy but
23:17that will still increase the AOV a lot.
23:20Normally you see up like normally you
23:23you can have like an upsell and downell
23:25option. So like one is a bit more
23:26expensive like $100 or something like
23:29this. Then you have down cell that is
23:30less expensive. Uh so you can you can
23:33experiment with that. Something to
23:34consider is that most apps in the space
23:38that are actually scaling to eight
23:40figures or more they are profitable
23:42because they have a very high LTV and a
23:45lot of times very high AOV too. So they
23:47can reinvest the money as soon as they
23:49get. So, you should have a lifetime
23:51value of $100 or more. That that's very
23:53important because that will allow you to
23:55put more money on it. You can like we
23:58can see CPA is also $10, $15 for example
24:01for a free trial with credit card or
24:03even a subscription of $20. That's
24:06possible. Normally, as you scale to 100K
24:09a month on MA, 500K a month, 1 million a
24:12month, that starts to become more
24:14difficult. And so that's why you should
24:17try to work at the same time on
24:18improving your LD. Then you can have
24:20different funnels as mentioned before.
24:23Um sometimes you can't have a quiz
24:25funnel with purchase just because you
24:28have to bring people in the app and then
24:29some people will buy uh later but it
24:32doesn't make sense to push people to buy
24:34at the beginning for some apps. But the
24:36ones that we see scaling the most,
24:38scaling very fast without funding, just
24:40bootstrap,
24:42those are normally some apps where they
24:45have either the option to buy something
24:48at the beginning or a free trial and
24:50then they charge. So a quiz funnel is
24:52the funnel where we see working the best
24:54right now. So like the apps we have that
24:56are spending a million dollars a month
24:58or more, they all have a quiz and um
25:01actually we have one exception but
25:03besides that there they all have a quiz
25:06funnel and the quiz also allows you to
25:09uh make sure you can use a unique
25:11mechanism and differentiate from others
25:13and uh so when they see the sales page
25:15they see okay this is the perfect
25:17solution for me and that's why they buy
25:18from you even though it's more expensive
25:21maybe than other solutions. Then the
25:23using the upsets is very important. We
25:25see sometimes you can scale but if you
25:27add some um upsets then you can actually
25:30scale. So that happens to some clients
25:32they couldn't scale and then um they
25:35started to see how to add upsess and
25:38what to do to increase the LTV and the
25:41over value that allowed them to actually
25:43scale. Then for the LTV, ideally use the
25:47180 days guarantee or something like
25:49that. Everywhere you just try to push
25:52people to not asking for the guarantee
25:55as the start if they don't like
25:57something but to actually practice um
25:59like the especially for some type of
26:01apps where they have to use it to
26:03actually use it and and mention how it's
26:06important that they use it for a long
26:08time and then they see the result. So
26:09you try to have that communication so
26:12people actually get better results and
26:14they are happier and they talk better
26:16about your app. In terms of creative
26:18strategy, we see one thing is very
26:20important is differentiating from
26:22competitive and uh when the business
26:25doesn't have a unique a unique mechanism
26:28um then we try to help them to create
26:31this unique mechanism and talk about
26:34that instead of the features because
26:35that's why people buy from you and buy
26:39from uh meet ad instead of just
26:41searching another app and the psychology
26:43as I was saying before it's very
26:45important and we have to understand the
26:47psychology behind why people buy and
26:50speaking exactly to them. Then
26:52understanding also the identities of
26:54people, how they see themselves. And
26:56especially for apps, we see for a lot of
26:58apps, just understanding that and how
27:00they want to be, how they want to be
27:02perceived and how they want to become,
27:05use that and then you use that in your
27:06ads. And then also uh for a lot of ads,
27:10you can understand the angle, you can
27:11understand the buyer persona and
27:12everything. If you use a format that
27:14feels like an ad normal normally it
27:17won't work or it won't work as well as
27:19it could. So you want to use some
27:21formats that are uh similar to what
27:25people see on organic what just makes
27:27sense for them what actually speaks to
27:29them and as I said high LTV that's very
27:32important so you can pay higher CPA than
27:35competitors and that will help you to
27:37make um will make everything easier for
27:40you. Then we often test for a lot of our
27:42clients five 10 constants a week per per
27:45uh um for for the app and per buy
27:48persona let's say and with three to five
27:50hooks each and then we increase volume
27:53as we scale. So it depends. Uh we
27:55normally don't do crazy volume. Uh so
27:58maybe um for some clients it can be up
28:0010 concerts a week. Nothing crazy. Um
28:03and then maybe we add some iterations or
28:06other things. But there are I mean there
28:09are two ways to scale. So normally as
28:12you scale you see a correlation between
28:15finding winning ads and just scaling uh
28:17your business, scaling your app. what
28:19you want to do is do more ads, better
28:22ads and also diverse ads like diverse
28:24creative. So often times do just doing
28:26more is not enough because you need uh
28:29just you also need different type of uh
28:31better personas to target but the at the
28:34end at the end of the day at the end of
28:36the day the goal you have is to find
28:38more winning ads. So and for winning ads
28:41I mean ads actually take most of the
28:43spend and become the highest spending
28:45ad. How you do that it depends. So
28:48ideally you want to understand what's
28:49your bottleneck to find more real
28:51winning ads. It can be that you are not
28:53doing enough ads or it can be that the
28:56win rate is pretty low or that you're
28:57only targeting the same audiences and
29:00you need to expand to other audiences.
29:02So normally if you have a very low
29:04winner rate, you want to prove that
29:06first and you want to try to have a
29:07slightly better win rate. So it's not
29:09going to be super high when you test a
29:11lot of ads and you're spending a million
29:13dollars a month while when you spend
29:15less it should be higher. While if your
29:17winner rate is fine, you just want to do
29:18more ads and continue to explore new
29:21audiences and then you just yeah just
29:23want to scale by I was like something
29:25that you'll see is some apps just do a
29:27like do maybe uh those 10 30 50 concepts
29:32a month that are high quality and that's
29:35more similar to our method. You also
29:37have some apps that scale a lot with
29:39just thousands of AI ads that can also
29:42work, but you have to be very good at
29:44doing very high quality ads. And uh it's
29:47it's difficult to keep the quality high.
29:49At the end of the day, whatever works,
29:51it's fine. And um yeah, we just focus
29:54more on quality because you see we see
29:56those ads are more stable. They can
29:58scale more and uh just like we found a
30:01system to to make them work. But we also
30:04see there are some businesses that just
30:07go on the quantity and they are able to
30:09um they're able to make it work too. And
30:12in general, yeah, then we continue to
30:13scale with different buyer personas,
30:15different angles and different format.
30:17So I'll show you a couple of examples
30:18just for seconds that you can see in
30:20general like from the other videos, the
30:22stuff we do. I'll show you a couple of
30:24examples for some of the of our clients
30:27and the type of ads we see working and
30:30we are doubling down on. for example,
30:33podcast ads, street interviews, some
30:35skit conversations. I'll show you a
30:37couple of examples.
30:38>> They say it takes about 800 hours to
30:40learn Spanish, but there are methods to
30:43cut down to like 135.
30:45>> Okay, so like those free language apps,
30:47>> apps with flashcards,
30:48>> they have stuff like this, like podcast
30:51clips. These are working super well for
30:53us.
30:54>> Hey, just make a wish.
30:56>> I wish to speak French the fastest way
30:59possible. Oh, you should try to learn
31:01the hack, man.
31:02>> So, you have stuff like this, just a
31:04skid conversation.
31:05>> This is David. [bell]
31:06>> He's a successful professional who
31:08thinks he knows his net worth.
31:09>> I've been tracking it for years. It's
31:11about $580,000.
31:12>> He's off by $92,000 and missing out on
31:15an easy additional 2,000. That's because
31:17>> So, here you have these type of formats
31:19also common organic that is skits where
31:21basically uh the person thinks they're
31:23doing everything right and they're Mr.
31:25Perfect and then actually uh they're
31:27doing everything wrong. So we are
31:29educating on that and like in a funny
31:31way and then presenting the app
31:34>> Canva for real. I mean why are you still
31:35using that?
31:36>> So you have like stuff like this too
31:38where it's podcast clip but yeah just
31:40more focused on on the uh competitors
31:44and why we're better competitors.
31:45>> This is Sarah. She's a successful
31:48entrepreneur who knows how hard
31:49>> yeah formats like this. I mean this is
31:51similar to that skit but you have
31:53someone on on the laptop and u yeah so
31:57this we have similar formats also like
31:59as um as expert um ads they're talking
32:03to someone on a call eight years of
32:06cross word puzzles gives you the same
32:08cognitive boost as just three months of
32:10language learning the brain needs active
32:12language
32:13>> this similar to the podcast ads but in
32:15this case it's an authority figure in a
32:18podcast setup so we just uh put
32:20something here like mentioning who this
32:22person is like I could say role and uh
32:24why they are an expert and uh we see
32:27that working pretty well too with also
32:29spoke person format I don't have an
32:31example here but that is working pretty
32:33well where you have like someone like an
32:35expert an expert that is talking in a
32:38room about something
32:39>> hey Chris I saw your request workers
32:42compet
32:42>> yeah so like here you have stuff like
32:43this or just like on a call
32:46>> hi quick question do you invest yes I do
32:48>> how do you deal with nowadays market
32:50fluctuations like do you have a
32:51financial adviser
32:52>> like certain interviews like this
32:54>> want to become bilingual in 3 weeks
32:55here's how memorizing words quizzes and
32:57flashcards
32:58>> or you have some uh yeah some formats
33:01like this are more simple but also
33:02similar to what works in organic I mean
33:04even like like some didn't add them as
33:07examples but even stuff like this works
33:10well like with just a music and um and
33:13no like without any voice uh just just
33:18music here and uh the video that works
33:20pretty well too, similar what to what is
33:22seen on on on Tik Tok. And yeah, in
33:25general now um to scale to $5 million a
33:28month with a couple of our clients, main
33:31focus is uh just continue to find new
33:34winners every month. Like ideally even
33:37more, but at least new winners every
33:39month. Um and doubling down on what
33:42works. At the same time, we want to
33:44expand to different bio personas,
33:46different angles, different formats
33:48because a certain point you can do ads
33:50better, but you're just getting most of
33:52the audience. You just have to to get to
33:55bigger audience or like different
33:57audience. Then uh just improving the
33:59creative process, the landing uh landing
34:02page. Sometimes we send to a landing
34:04page and then to a quiz, sometimes
34:06directly to a quiz. So proving that and
34:08the conversion rate uh between the
34:10landing the quiz or just between the uh
34:13quiz and the sales page, proving the
34:14sales page, the upsells and also the
34:17congruence between the creatives and all
34:20of these because they have to be
34:22similar. They have to be customized per
34:24buyer persona. Um the good thing is with
34:26the quiz you don't really need to do
34:28that that much because you can just
34:30enter the quiz and then the quiz um
34:33allow allows you to have a personalized
34:36sales page. You just want to have
34:37different variables of sales pages based
34:39on what they answer. Yeah, just focusing
34:41on increasing the conversion rate, the
34:43average value, the LTV is what we see it
34:46helps to then have also higher to to
34:49allow you to get a higher CPA on meta
34:52and still the profit and uh and on the
34:54other side you just work on meta to be
34:56able to spend more by keeping the same
34:58KPIs and the same CPA. Uh then unique
35:01mechanism is also important to add new
35:03unique mechanisms because one can get
35:05stated or because you are expanding to
35:07different bio personas and you have to
35:09have new unique mechanisms for them or
35:12also like for people in different
35:14awareness level and that's to be the
35:15generic one for the ads but also the one
35:17that you use in sales page and then
35:19ideally we try to add some bars to entry
35:22I mean that's not really on us it's more
35:24on the um on the uh business side but
35:28like some unique positioning that is
35:30more difficult to replicate. Just
35:33improving the app so it's better than
35:34competitors and the competitors can
35:36really come in and do a similar app and
35:38just working also with influencers for
35:40example authority figures so you just
35:44get more recognized and you have more
35:45social proof or just other things that
35:47can add barriers to entry. So I think I
35:50I hope that was helpful and you got some
35:52insight from this. If you have any
35:54questions, feel free to add a comment
35:55here or even better like send me a
35:57message on X and uh um I'll I'll take a
36:01look and if it's interesting, I'll just
36:03do another content based on that too.
36:06And if you have an app or any other
36:09business that is spending on meta and
36:11you spend more than 50k a month, then uh
36:13I have a link in description. You can
36:15book a call with me and yeah, hope that
36:17was valuable. I'll see you in the next

About this transcript

This page contains the full transcript of How I Scaled An App To $3M/Month With Meta Ads (Step by step) by Lorenzo Pravatà, generated from the public captions YouTube serves with the video. The transcript has 6,757 words across 924 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 choose a caption track available on the video.

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, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.