YT YouTube2Text

YouTube2Text → Transcripts → Give Me 33 Mins, I'll Give You 10,000 Hours of Ecom Ads Knowledge

Give Me 33 Mins, I'll Give You 10,000 Hours of Ecom Ads Knowledge — Transcript

by Sergio C · 6,654 words · 1,053 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00A women's apparel brand started working
0:01with us a few years ago. Within the
0:03first 30 days, we were able to drive
0:05more traffic, more clicks, [music]
0:07and plenty of add-to-carts, but the
0:09purchases were still falling off. The
0:11ads were doing their job, even better
0:13than before, but something outside of
0:15the ad account was killing each sale.
0:18And the founder was protecting a policy
0:20that was causing that leak. And I'll
0:22show you the checkout decision that we
0:24made in a minute, but whether you're a
0:25D2C founder, head of growth, or an
0:28in-house marketer already spending real
0:30money on acquisition, you probably
0:31already know how to launch a campaign.
0:33The harder question is whether you can
0:35tell which part of the business you
0:37should actually focus on. At ADVT Media,
0:39we've managed paid acquisition for more
0:41than 600 e-commerce brands now over the
0:44last 8 years, and driven over 125
0:47million in client sales from our
0:49campaigns alone. In this video, I'm
0:51going to give you the lessons that
0:52changed how I think about creatives,
0:54campaign structure, attribution,
0:56economics, and scale. Some of these
0:58things apply to $500 a day budget, and
1:00others apply to those at $50,000 a day.
1:02Every one of them changes how you read
1:04the account before you spend the next
1:06dollar. So, number one, the highest ROAS
1:09campaign may actually be the least
1:11scalable. Now, let's start with
1:12something that I think a lot of
1:14e-commerce brands don't even think
1:15about, and that's how they look at
1:17return on ad spend. If I asked you
1:18whether you'd spend $1,000 a day at a
1:20five times ROAS or $10,000 a day at a
1:24three times ROAS, which would you
1:25choose? Because I think a lot of people
1:27immediately gravitate towards the 5x.
1:30It's a higher return, so naturally they
1:32assume it's the better option. But the
1:34first campaign is generating $5,000 in
1:36revenue, while the second one is
1:38generating $30,000 of revenue. Now,
1:41obviously we need to account for
1:43margins, acquisition costs, and actual
1:44profitability before making a decision,
1:46but the point I'm making is that the
1:48highest ROAS doesn't automatically mean
1:50it's your best opportunity for growth,
1:53or best for the business. And this
1:54becomes even more important when you
1:56understand the difference between
1:58creating demand and capturing demand.
2:00New customers versus limited low-hanging
2:03fruit. Let's say you're running two
2:05campaigns. One is generating five times
2:07return on ad spend, and the other is
2:08generating a 1.5. Natural reaction for
2:11all of the people I talk with is to shut
2:13off that 1.5 campaign and put more money
2:15into the 5x. And trust me, I've gotten
2:18the AI brief from founders who yell at
2:19us to do this for them. But, what if the
2:221.5 campaign is actually introducing
2:24thousands of completely new people to
2:26your brand while the five times campaign
2:28is primarily reaching those engaged
2:31users just at a later point? Well, that
2:33changes things a bit because the lower
2:35ROAS campaign could actually be
2:37contributing and fueling the performance
2:39of the higher ROAS campaign. And a lot
2:41of the times it is. I've seen this
2:43happen with brands where existing
2:45campaigns look incredible on paper, but
2:47when you actually break down who those
2:49campaigns are reaching, a lot of that
2:52spend is going towards people who are
2:54already familiar with the business.
2:57Engaged users and existing customers.
2:59You can look at your audience segment
3:01breakdown here and see for yourself.
3:03Meanwhile, your new customer acquisition
3:05campaigns or top-of-funnel campaigns
3:07might have a lower return or reported
3:10return, but they're actually doing the
3:11work of bringing in the new people
3:13[music] into the ecosystem. But, they're
3:16actually doing the work of bringing
3:17people into the ecosystem, into those
3:20other campaigns. And if you shut off
3:21those bad campaigns, the ones with the
3:231.5 or the top-of-funnel, simply because
3:26their ROAS is lower, you could
3:27eventually start hurting the campaigns
3:29that appear to performing so well. They
3:32exist because of the other ones. I'm not
3:34saying you should continue spending
3:36money on an unprofitable campaign just
3:38because it's reaching new people. We
3:39still want those campaigns to be
3:41healthy, but you have to understand the
3:43ecosystem the consumers are in and
3:45whether that spend is actually valuable.
3:48But, I think the bigger lesson here is
3:49that you need to understand the role
3:51that each campaign plays in growing the
3:52business. Not only campaigns, but also
3:55platforms. How do Google branded
3:57campaigns help support the awareness you
3:59drive from Meta? Some assets create
4:01demand while others capture it, and some
4:04are responsible for retargeting people
4:06who are already interested. You can't
4:08judge all of them by the exact same
4:10metric without understanding what
4:12they're actually contributing. And that
4:14leads into something else I see
4:16constantly, which is brands spending all
4:18of their time looking at the advertising
4:20dashboard when the actual problem is
4:22happening somewhere else entirely. Which
4:24leads me to number two, better ads can
4:27actually make a broken funnel more
4:29expensive. [music] Remember that women's
4:31apparel brand I mentioned in the
4:32beginning? We were driving a lot of
4:34traffic, getting more clicks, and seeing
4:36people add products to their cart. Our
4:38process is really good at cutting the
4:39fat and making your system more
4:41efficient. But for whatever reason, that
4:43activity wasn't translating into the
4:45number of purchases we wanted. And I
4:47think a typical response from a lot of
4:49advertisers would be to start changing
4:51the creatives, testing new targeting, or
4:53rebuilding the whole entire campaigns.
4:55But when you think about it, why would
4:57we immediately assume the ads are the
4:59problem when they're already getting
5:01people interested enough to add to cart?
5:03Changing the ads would just change the
5:05flow of users to the site. It wouldn't
5:07address where the constraint is right
5:09now. So, instead, we started looking
5:12further down the funnel. Remember the
5:14river of water analogy I had in another
5:16video? Well, we did exactly that. And
5:18what we identified was the $12 shipping
5:21charge that the customers were
5:22encountering during a checkout for a $30
5:25product. The founder didn't want to
5:26offer free shipping because they were
5:28concerned about protecting their
5:30margins. Which I completely understand.
5:33But what was happening was that they
5:34were protecting the margin on each
5:36individual order while losing a
5:39significant number of potential
5:41purchases. And once that shipping
5:43friction was removed, even though the
5:45margin per order decreased, we were able
5:48to actually generate considerably more
5:50purchases. I had to get on a call with
5:52them and literally lay out, "Do you want
5:5410 sales per day and $300 of revenue
5:56with perfect margins?" Or would you
5:58rather have $100 sales per day and
6:00$3,000 of revenue with no extra ad
6:03spend, by the way, and a little less
6:05margin on each sale. And I think that's
6:07such a good example of why you have to
6:09look beyond the ad account. The problem
6:11wasn't that people weren't interested in
6:13the product or the ads or they weren't
6:15getting to the website. It wasn't that
6:17we needed to completely change our
6:18strategy, especially on the meta side.
6:20The problem was that the customers were
6:22getting all the way to checkout and then
6:23abandoning the purchase. If we had
6:26focused on changing the ads, generating
6:28more traffic, we would have just
6:29continued sending more people into the
6:31same problem. This is why I always look
6:33at the full customer journey when we're
6:35diagnosing performance. Look at the
6:37whole river of water. Start at the top.
6:39If your CPMs are unusually high, we need
6:42to understand why it's so expensive to
6:44reach people, whether we show a good ad
6:46or not. If your click-through rate is
6:48low, we might have a creative problem
6:50because people aren't interested enough
6:51to even click through, even if they are
6:53seeing it on the feed. But if people are
6:56clicking and landing on the website and
6:58then they're not adding anything to
6:59their cart, we need to start looking at
7:01the product page, [music] the offer, the
7:03price point, or how the product is being
7:05communicated in the description or the
7:07copy. Does it explain the value? How are
7:10the reviews? You have to look at
7:11everything on the product page. Think of
7:13yourself as the consumer. And if people
7:15are adding to cart, but they aren't
7:16completing the purchase, then I'm
7:18looking at everything after the cart,
7:19the checkout process, the checkout
7:21friction, the shipping cost, cart upsell
7:23apps, making the cart maybe too noisy,
7:25payment options, and anything else that
7:27could prevent them from buying and
7:29completing that next step. And it
7:31doesn't stop at the first We also want
7:33to understand average order value,
7:35customer lifetime value, and how
7:37frequently people are repeating
7:39purchases. All of those things affect
7:42how much we can afford to spend on
7:43acquiring customers. So, let's say you
7:45have 500 people adding products to their
7:47cart, but only 20 of them are
7:49purchasing. You could launch another 20
7:51creatives and potentially get 550 people
7:54to add to their cart and 21 sales, but
7:56what have you really accomplished there?
7:58If the same number of customers continue
8:00abandoning the checkout, you're making
8:02the problem more expensive, and that's
8:04not something that spending more money
8:06can fix. Sure, you'll have more sales
8:08per day and it'll look better maybe on
8:10paper, but they'll never actually be at
8:13an efficient rate that you want. So, the
8:15lesson isn't that every brand needs to
8:17offer free shipping, it's that you need
8:19to identify the biggest constraint in
8:20your business before deciding where the
8:23next dollar should go. If it's the
8:24shipping rate, try free shipping for 3
8:26days or try baking in the shipping price
8:28into the product cost. Did that help or
8:30what else can be done? Sometimes the
8:32most valuable thing you can do for your
8:34advertising performance has absolutely
8:36nothing to do with the advertisements
8:37themselves. And number three, platform
8:40delivery can make bad data look like a
8:42winning test. Now, let's talk about
8:44creative testing because I think this is
8:46one of the most misunderstood parts of
8:48running Meta ads. Everybody talks about
8:50the importance of testing creatives and
8:52I completely agree. I love testing for
8:54the opportunity to find something that
8:56works better than what we already have,
8:58but I also think a lot of advertisers
9:00believe they're testing creatives when
9:02in reality they're just letting Meta
9:03decide what creatives receive the
9:05budget. And here's what I mean. Let's
9:07say you launch 10 creatives inside of
9:09one campaign. And after a few days, you
9:11look at the results. One creative has
9:13spent $1,000, another spent 200, and the
9:16other eight have barely spent anything.
9:17The first creative generated a few
9:19purchases, so you decide that that's the
9:21winner. You shut everything else off and
9:23you move on. But, how do you know that
9:25those other eight lower on the list
9:27[music] wouldn't have performed better
9:29if they received a fair opportunity? All
9:31you really know is that Meta decided to
9:33allocate the majority of your budget
9:34towards one or two of the ads and people
9:36are on Reddit and YouTube comments
9:38saying, "Yeah, because the AI is so
9:40smart, we have to listen to it." And
9:41these same people haven't been able to
9:43scale past $200 a day. And there's a
9:45reason. And look, I'm not suggesting
9:47that Meta's algorithm doesn't know what
9:49it's doing. There's a reason it's
9:50prioritizing certain ads based on the
9:52signal it receives, but there's a
9:54difference between allowing an algorithm
9:56to optimize delivery and actually
9:58conducting a controlled creative test to
10:00understand what resonates with your
10:02market. If one creative receives 50,000
10:05impressions and another receives 400, I
10:07I think that you and I can confidently
10:10conclude that the second creative was a
10:12loser. We just simply don't have enough
10:14information. I'm going to be honest,
10:16usually there's a winning ad somewhere
10:18down on that eight ad list that changes
10:20everything for the business. If you left
10:22it up to meta and if you trust the
10:24algorithm and if you think you know the
10:26strategy, it'll never get the spend it
10:27needs and this is why our approach to
10:29testing at ADV Team Media is very
10:31intentional. We want to isolate the
10:33variables. We want to understand exactly
10:35what's responsible for performance that
10:37we're seeing. [music] It's not some
10:38magic service that gets you a
10:40nine-figure business in 30 days, but
10:42it's clarity and action on building a
10:44structure and plan for your advertising
10:46to get the consistent results for the
10:49long term. So, if we're testing
10:50creatives, I don't want to
10:51simultaneously change the targeting
10:53landing page general copy because then
10:55if something performs, we don't really
10:57know which variable was responsible. I
10:59want to keep those things as consistent
11:01as possible and give each creative a
11:04meaningful opportunity to perform. And
11:07when we do these tests, we're not just
11:08looking at purchases either. Obviously,
11:10purchases are what we're ultimately
11:11trying to generate, but there's a lot of
11:13useful information that we can collect
11:15before someone actually buys. We talked
11:17about that river of water. We can look
11:19at the CPM, the CPC, the link
11:21click-through rate, the add-to-carts,
11:22initiate checkouts, how those metrics
11:24relate to one another, the purchases,
11:25the AOV, everything. So, for example,
11:28one creative might generate really
11:29inexpensive clicks, but none of those
11:31people are adding products to their
11:32carts. You're just getting a lot of
11:34them. Another creative might have a
11:35higher CPC, higher CPM, but the people
11:37clicking much more likely to initiate
11:39checkout or purchase. To me, that second
11:41creative would be more valuable even
11:43though the initial traffic is expensive.
11:45So, depending on the account, we might
11:46use something like 1,000 to 3,000
11:48impressions as the initial checkpoint
11:51when comparing creatives and we can use
11:53automated rules or impression limits to
11:55prevent one ad from taking all of the
11:56delivery before we learn anything about
11:58the others. And that's not some magic
12:00number that applies to every business.
12:02If you're dealing with extremely high
12:04CPMs or a small budget, you need to
12:06adjust your testing accordingly. The
12:08objective with all of this is to collect
12:10enough comparable evidence, comparable
12:12is the keyword, to make a informed
12:14decision, not follow some arbitrary
12:17testing rule. And I personally believe
12:19that's the distinction. Just because
12:20Meta didn't spend more money on a
12:22creative doesn't automatically mean it
12:23was a bad creative. I've seen this play
12:26out hundreds of times. I just know. And
12:28if you understand that, you start
12:29approaching creative testing very
12:31differently. And there's something else
12:33equally important, and that's what you
12:34actually do with the information once
12:36you have identified the winners, which
12:38brings me to number four. The winner is
12:40less valuable than the reason it won.
12:42And I think one of the biggest missed
12:44opportunities in creative testing is
12:46that brands spend so much time trying to
12:48identify winning advertisements, but
12:50very little time understanding why those
12:52ads are winners in the first place.
12:54Let's say you launch 20 new creatives
12:56and three of them perform really well.
12:58Most advertisers will take those three,
13:00put more money behind them, shut off the
13:01losers, and then go and produce another
13:0320 creatives. And that's fine to an
13:05extent. I think you're leaving an
13:06enormous amount of valuable information
13:08on the table if you don't take the time
13:10to compare those winners and understand
13:12what they have in common. For example,
13:14maybe all three winning creatives are
13:16addressing the same objection that the
13:18customers have before purchasing, or
13:20maybe they all demonstrate the product
13:22within the first few seconds, rather
13:23than spending time introducing the brand
13:25or having the brand logo pop up. Or
13:27perhaps they're leading with a
13:28particular problem the customers
13:30experiencing instead of just talking
13:32about the product's features. Well,
13:33those things, all three of them or more,
13:36I want to identify because now we have
13:38learned something about the customer
13:39that we can use moving forward. And of
13:41course, mirroring success is always the
13:43fastest way to it again. And once we
13:45understand that, we can build an
13:47entirely new round of creatives around
13:49the same insights. We can test the
13:50different hooks, different formats,
13:51different creators, different product
13:53demonstrations, and different ways of
13:55communicating that message that won. But
13:58we're no longer just throwing random
13:59concepts at the wall and hoping that
14:01something sticks. All of this is about
14:03what resonates with the market, build on
14:05that. And that's really how I think
14:06creative testing should work. Every
14:08round of testing should teach you
14:09something that helps inform the next
14:11round. And over time, you should develop
14:13a much better understanding of why
14:16customers are purchasing your products,
14:18which messages resonate with them, and
14:20how you can continue communicating those
14:22messages in different ways. Because an
14:25individual winning creative might
14:27eventually stop performing. They They
14:28always do. But the insight is what made
14:30that creative successful, and that can
14:32help you produce five or 10 more
14:34winners. And to me, that's a lot more
14:36valuable than just finding one good ad.
14:38Number five, one product may need
14:40different ads for different stages. So,
14:42another thing that I think people will
14:44overlook is that a creative can perform
14:46extremely well for one audience and
14:48poorly for another, even when you're
14:50advertising the same product. And a lot
14:51of the times, it comes down to where the
14:53customer is in their purchasing journey.
14:55A retargeting ad might work for the
14:56low-hanging fruit, but not for a new
14:58customer who needs to learn about your
14:59product. And we've worked with a
15:01pickleball brand, and one thing we
15:02noticed was that there's always several
15:04different ways we can advertise the same
15:06paddle. We could show someone actually
15:07playing pickleball, demonstrate the
15:09paddle in action, and etc., etc. We can
15:11create a more technical breakdown
15:13explaining the paddle's construction,
15:15the materials, the performance
15:16characteristics, hit stronger, get more
15:19agile, whatever. We can produce content
15:21helping the customers understand what
15:22paddle was best suited for their skill
15:24level. There's so many different ways to
15:26show this paddle, why it's important to
15:28you, and which one's the most effective
15:29to use for paid ads. And what we found
15:31was the gameplay style creatives worked
15:33actually really well at the top of
15:35funnel when we were introducing people
15:37to the brand. Maybe the movement was
15:38engaging, who knows? And they kind of
15:40make sense because someone who's
15:42scrolling through Instagram never heard
15:44of this company probably isn't looking
15:45for a detailed technical explanation of
15:48the paddle and the materials. Like, why
15:49would they care? But they might stop
15:51scrolling when they see an interesting
15:53gameplay video that catches their
15:54attention, especially if they're in that
15:57world and they can kind of see skill
15:59where it is. Now, once that same person
16:01has already discovered the brand,
16:02visited the website, started considering
16:04the product, and more technical
16:05information become a lot more relevant.
16:07They're further along in the buying
16:09process, and they're looking for the
16:10information that helps them make a smart
16:13purchasing decision. In this particular
16:15account, those technical breakdowns
16:16worked better in retargeting, not for
16:19top of funnel. So, imagine if we only
16:20tested those technical videos against
16:23cold audiences, saw that they weren't
16:25performing as well as the gameplay
16:26creatives, and decided to stop running
16:28them forever, right? And a lot of people
16:30put everything in one campaign, and
16:31etc., etc. It just doesn't work like
16:33this. We could have eliminated something
16:34valuable simply because we were showing
16:36it to the wrong person, or maybe just at
16:38the wrong time. This is why I don't
16:40think the question should always be
16:41which creative performed the best
16:43overall. I also want to understand what
16:45creative performed the best for which
16:47audience at a certain [music] stage of
16:49the funnel. Sometimes a creative isn't
16:51inherently good or bad, it's just
16:53positioned in the wrong context. And if
16:55you're getting traffic and add to carts,
16:57and you still can't tell what constraint
16:59is holding back the business, that's the
17:00work that we do at ADVT Media. We look
17:03at the campaigns, the creatives, the
17:04website funnel, the economics together,
17:07then show you where the next dollar has
17:08the most leverage. Go to the link
17:10description below to book a call, and
17:12we'll talk soon. But, number six, small
17:15and large accounts should not follow the
17:17same playbook. Let's talk about scale,
17:19because the advice that helps a $100,000
17:22a month account can waste a smaller
17:23brand's entire budget. And
17:25unfortunately, the strategies you hear
17:27on YouTube is exactly that happening. It
17:30might work for them, but not for you. I
17:32think one of the worst things smaller
17:34e-commerce brands can do is blindly copy
17:37the strategies that they're seeing from
17:38advertisers managing significantly
17:40larger budgets. And I see this all the
17:42time. Someone watches a video on how a
17:44brand spending $50,000 a day structures
17:47its campaigns, how they produce 10 ads a
17:50week, and just toss them into the
17:51campaigns, and suddenly they want to
17:53implement that same strategy in an
17:55account that's spending 30 or $100 a
17:57day. But, those two things are
17:59completely different situations, and
18:01they require different approaches,
18:02different structures. If you're spending
18:04under $1,000 a day, you might be able to
18:06scale your business with three to five
18:08really strong creatives. You don't need
18:10to produce 50 new creatives a month, and
18:12you probably don't need 15 different
18:14campaigns competing for the same limited
18:16budget. Your priority should be
18:18identifying what's working,
18:20understanding why it's working, and
18:21allocating your budget towards the
18:23things that have the most leverage or
18:25the highest probability of producing
18:27results. Now, compare that to a business
18:29operating at $50,000 a day. At that
18:31level, you're putting a significantly
18:33larger amount of spend behind your
18:34creatives, and you're potentially going
18:36to need a much faster process for
18:37introducing new concepts because each
18:39individual ad is absorbing considerably
18:41more budget. You have more data, you
18:44have more opportunities to test, and
18:46ideally, more resources of available to
18:48support that production. Because even if
18:50you toss in 10 new ads a week or a day,
18:53you have enough budget for each ad to
18:55get the data required to let the
18:56flywheel spin. You don't have that at
18:58$30 a day. And the same thing applies to
19:00campaign structure and automation. I'm
19:02not against broad targeting and
19:03automating campaigns. I think those
19:05tools can be incredibly valuable and we
19:07use them. But, I also think you need to
19:09understand how much control you're
19:10giving up and whether that actually
19:11makes sense for the account that you're
19:13currently managing. If you're spending a
19:15few hundred dollars a day, you don't
19:16necessarily have all of the budget five
19:18or 10 different strategies run
19:20simultaneously while you wait to see
19:22which one works. You also have to be
19:24careful about scaling something too
19:26quickly just because it generated a few
19:27purchases. I get on calls all the time
19:29where they find a winning ad, and then
19:31they want to scale at a 1.5 return. It
19:33just doesn't make sense. We've all seen
19:35creatives that have a incredible first
19:37day or first week and then completely
19:39fall apart once you put more money
19:40behind them. A good initial result is
19:42encouraging. I actually think it's
19:44really good proof of concept that your
19:45market is there and they want your
19:46product, but that doesn't automatically
19:48mean the creative has enough evidence
19:50behind it to support aggressive scaling.
19:53The soft metrics also have to matter.
19:55So, I think the biggest takeaway here is
19:57you need to match your strategy to the
19:59actual scale of the business. Don't
20:01complicate things just because you see
20:03larger brands doing something
20:04differently. It's for them. A simpler
20:06strategy that's appropriate for your
20:08budget and supported by the data will
20:10often be far more valuable than trying
20:12to imitate a much larger advertising
20:15operation, especially in a different
20:16industry. Number seven, sometimes the
20:18best ad optimization happens after the
20:21click. I know after the click, I keep
20:23saying this, but this is another concept
20:25I think every e-commerce founder should
20:27really, really, really understand. And
20:28that's the fact that ROAS is an output
20:31of several different inputs. It's not
20:33the same independent number that the
20:35advertising platform controls on its
20:37own. Your cost per click matters, your
20:39conversion rate matters, your average
20:41order value matters, and your customer
20:43lifetime value matters when determining
20:45how much you can afford to spend
20:46acquiring the customer. So, let me give
20:48you a really simple example. Let's say
20:50you're spending a $1,000 a day on
20:51advertising and generating 100 purchases
20:54with an average order value of $30. That
20:56is $3,000 in revenue or a three times
20:59return on ad spend. Now, imagine we
21:01don't change anything about the
21:02advertising. We're spending the exact
21:04same amount, getting the exact same
21:05amount of traffic, and generating the
21:07same number of purchases. But instead of
21:09a $30 average order value, we increase
21:12that to 45. Bundle, an upsell, whatever.
21:15We're now generating [music] $4,500 in
21:17revenue, which means our ROAS increased
21:19from a three to a 4.5 without making a
21:21single change to the ads. And if we can
21:23increase average order value to 70, now
21:26we're generating $7,000 in revenue from
21:28the exact same advertising budget.
21:30Obviously, you need to make sure those
21:32additional sales dollars are translating
21:33into healthy margins, but I believe this
21:35illustrates how much opportunity can
21:37exist outside of the advertising
21:39platform. We actually had an example
21:41with this baby bedding brand. They were
21:42selling products like crib sheets,
21:44pillows, blankets, crib skirts, and we
21:46noticed that the customers were
21:47frequently purchasing multiple
21:49complementary products [music] together
21:51on their own, manually. But they were
21:53having to go through the website and
21:55assemble these things themselves. So,
21:57rather than just continuing to advertise
21:59each product individually, we identified
22:01an opportunity to package those into a
22:03complete bundle. Instead of only
22:05offering something like a $70 crib
22:07sheet, we could now present a complete
22:09welcome home bundle around $200. And
22:12what's interesting about that is we're
22:13not necessarily asking Meta to find us
22:16[music] cheaper customers or generate
22:17more traffic or test this creative and
22:19this concept and this angle. We're
22:21creating an opportunity to generate
22:23significantly more revenue from the
22:24customers we're already acquiring while
22:27also making it convenient for them to
22:29purchase everything that they need. And
22:30so, again, it's not so complex. It's not
22:33so I have to learn Meta ads. It's we
22:35have to learn the customer. And I think
22:36this is where founders need to start
22:38thinking about advertising as part of
22:40the broader business rather than
22:42something that operates independently.
22:44If you can improve your average order
22:45value, introduce complimentary products,
22:49create a better bundle, or build
22:51stronger post-purchase experience, those
22:53things [music] can completely change the
22:54economics of acquiring customers. And I
22:57don't think discounting should
22:58automatically be the answer, either.
23:00It's very easy to offer 20 or 30% off
23:03and see your conversion rate improve.
23:05But, what happens to your margins? Are
23:07you actually making more money or are
23:09you just generating more revenue at a
23:11lower profit? Those are questions we
23:13need to be asking. Sometimes the best
23:15thing you can do for your advertising
23:17isn't to improve your cost per click or
23:19your CTR. It's to improve the amount of
23:21value the business generates from each
23:23customer because once those economics
23:25improve, you potentially create much
23:27more room to acquire customers
23:29profitably at [music] scale. And don't
23:31take this as you need a high AOV to make
23:33advertising work. It just makes it work
23:35a lot easier. Number eight, platform
23:37attribution isn't the same as
23:38incrementality. The next lesson is
23:40something I think separates people who
23:42understand how to operate Ads Manager
23:44from people who really understand how
23:46advertising impacts a business. And
23:48that's the difference between
23:49attribution and incrementality. Let's
23:52say someone sees your product right on
23:53Instagram for the first time. They've
23:55never heard of your company. It's
23:56usually how it goes, but your ad catches
23:58their attention for a second or two.
24:00They watch the video, maybe visit the
24:02website, then continue scrolling without
24:04purchasing. Very common, happens all the
24:06time. A few days later, they remember
24:08your product. They search for your brand
24:09on Google, click the branded search ad,
24:12and then complete the purchase. Or,
24:14sometimes what happens is they go from
24:15Google, sign up for the newsletter,
24:17leave, click on the email, now go back
24:19to the website and purchase. Now,
24:20depending on the attribution settings,
24:22Google or the email or whatever might be
24:24reporting that conversion. And Meta may
24:26also be taking some credit or maybe not
24:29if it's outside of the 7-day attribution
24:30window, but here's the important
24:32question. How would that customer have
24:33purchased from your business would have
24:35never seen the meta ad in the first
24:37place and that's something you can't
24:38necessarily determine just by looking at
24:40the platform's dashboard or even Shopify
24:42sometimes will say that that's a direct
24:44purchase or direct customer because the
24:45difference between that platform
24:47receiving credit for a purchase and a
24:48platform actually creating incremental
24:50demand, that distinction is incredibly
24:52important when you're deciding where to
24:54allocate your advertising budget. The
24:56email sale or the Google sale never
24:58would have happened without the initial
25:00touch point. I've seen situations where
25:02a brand's meta advertising is
25:04introducing a significant number of new
25:06people to its products, but some of
25:07those people ultimately complete their
25:09purchases through Google and you can
25:10have similar situations with Amazon
25:13where someone discovers a product on
25:14Instagram but decides to purchase it on
25:16Amazon later. Now, we have to be careful
25:17here because just because branded search
25:19volume or Amazon sales increased while
25:21you're running ads doesn't automatically
25:23prove that meta caused those sales. You
25:25need the stronger evidence potentially
25:27including incrementality testing before
25:30making that claim, but it's absolutely
25:32possible for advertising on one platform
25:34to influence buying behavior that
25:36eventually gets recorded somewhere else.
25:37We have people on the team internally
25:39here that when they see ads they never
25:40buy on the websites. They always check
25:42on Amazon first. Your customers are
25:43probably doing the same thing a certain
25:45percentage of them and this is why I'm
25:46such a big believer in looking at
25:48blended performance and understanding
25:50what's happening across the entire
25:51business. If we're increasing
25:53advertising spend on meta, what's
25:54happening to total revenue? What's
25:56happening to new customer acquisition?
25:58How are emails and subscriptions being
26:00affected? Are we generating more
26:02contribution profit? Is the business
26:03actually growing in a healthy way on
26:05other channels as well because
26:07ultimately that's what matters. It's an
26:08ecosystem. It's not meta versus Google
26:10versus email and I don't want to make
26:12decisions exclusively based on whether
26:14each platform reporting has the highest
26:16ROAS. It's once you get to a certain
26:18point ROAS is kind of silly. I want to
26:20understand what is contributing to
26:22growth of the entire business. Once you
26:24start looking at advertising that way
26:26you start to understand that the
26:27platform dashboard is just one piece of
26:29the puzzle. It's useful, but it doesn't
26:31tell the entire story, which leads me to
26:33number nine. Flashy platform updates are
26:36not a strategy. Unfortunately, hate to
26:38break it to you.
26:40All the YouTube videos are fake, but and
26:41I think having that broader
26:42understanding is especially important
26:44today because every time Meta announces
26:47some new update, it seems like the
26:49entire advertising industry is they just
26:51start trying to convince everybody that
26:52they need to change their strategy
26:54completely. Whether it's Andromeda, AI
26:56generated creatives, or some new auto
26:58mated campaign, whatever feature or
27:00whatever happens to be getting attention
27:02that month, everybody suddenly has an
27:03opinion about the new best way to run
27:05Facebook ads. Now, I want to be clear.
27:07These updates obviously matter. The way
27:09Meta delivers ads changes and as
27:11advertisers, we need to absolutely
27:12understand what's happening and how
27:14those changes might affect performance,
27:15right? We always want to be open to AB
27:17test, but I think people get so caught
27:18up in these updates that they forget
27:20about the things that actually drive
27:21results. If your product page is
27:23converting at a terrible rate, hate to
27:25break it to you, but Andromeda isn't
27:26magically going to fix that. If your
27:28average order value is too low to
27:29support profitable customer acquisition,
27:31producing another 50 AI creatives isn't
27:33necessarily going to solve your problem
27:35either. And if customers are abandoning
27:36your checkout because of your shipping
27:38costs are too high, completely
27:39restructuring your campaigns probably
27:42isn't going to make much of a
27:43difference. And this is where I think a
27:45lot of agencies and media buyers, they
27:46just get distracted almost. They'll
27:48spend hours talking about the latest
27:50algorithm changes or which campaign
27:52structure Meta supposedly prefers, but
27:54they can't actually explain why the
27:56business that they're advertising for
27:58isn't profitable. To me, that's a much
28:01bigger problem than not understanding
28:03the latest platform update. When you
28:04work with an agency who just says, "Give
28:06it 3 more months. Give it more time. Let
28:08the campaigns learn. It'll figure it
28:09out." Isn't that a bigger problem than
28:11actually understanding what the next
28:12step should be? And the fundamentals of
28:14advertising haven't disappeared, whether
28:16Andromeda came out or not. You still
28:18need to understand your customer,
28:19communicate why your product is
28:20valuable, create advertisements that
28:22resonate with your market, send people
28:23to a website that converts, and make
28:24sure the economics support your
28:26acquisition cost. Like, where is
28:27Andromeda in there? And think that's
28:29where the focus should be. We absolutely
28:31We to take advantage of improvements in
28:33advertising technology, but those tools
28:35should support the fundamentals, support
28:38the strategy. They shouldn't become the
28:40strategy because when things change,
28:42your strategy dies, basically. At the
28:45end of the day, no one really cares how
28:46familiar you are with the latest updates
28:48if you can't actually grow the business.
28:49What matters is whether you can identify
28:51the problems preventing the growth,
28:53understand which opportunities are worth
28:55pursuing, [music] and make informed
28:56decisions about where to invest the next
28:59dollar. 10, your best practice is still
29:01only a hypothesis. And the last lesson I
29:04want to share is probably one of the
29:06most important things I've learned from
29:08managing advertising across so many
29:10different e-commerce brands. Just
29:11because something worked extremely well
29:13in one advertising account, doesn't mean
29:15it's going to work in another, even for
29:17the same industry, even for the same
29:18products. I think one of the biggest
29:20problems with advertising advice is that
29:22everybody wants to speak in absolutes.
29:24You'll hear people say that broad
29:26targeting is always better than interest
29:28targeting, that look-alike audiences
29:29don't work anymore, that UGC is the only
29:32creative format you should be testing,
29:33or that static images are dead and it's
29:362026 and yada yada. And I just don't
29:39agree with those kind of statements
29:40because I've seen too many accounts
29:42where where the exact opposite is true.
29:44I've seen comments that say, "You're
29:45still doing interest in post-Andromeda?"
29:48Well, I have right now interest
29:50targeting campaigns that outperform my
29:51broad targeting at a six times return.
29:53I've seen look-alike audiences
29:54performing incredibly well. I've seen
29:56simple static images outperform
29:57expensive UGC videos that brands
29:59invested a lot of time and money into
30:01producing. I've also seen broad
30:03targeting and UGC outperform everything
30:05else completely in different accounts.
30:07So, which one is right? Well, it depends
30:09on the business, and I know that sounds
30:11like a boring answer, but that's the
30:12reality of advertising. This is why the
30:14AB tests are so important because every
30:16brand has different products, different
30:17customers, different price points,
30:19different levels of awareness, different
30:20ads, different landing page, different
30:22colors, different amounts of money
30:23available to spend on acquisition. Like,
30:25if the conditions change, should we
30:27expect the same results? This is why we
30:30have to figure out what's good for you,
30:31and that's why I think the role of an
30:32experienced advertiser isn't to come
30:34into every account and immediately
30:35[music] implement the exact same
30:37strategy. Experience should give you a
30:39better understanding of what to test
30:41first and what opportunities have a
30:43higher probability of producing results
30:46that you want. At the end of the day,
30:47even those are hypotheses because
30:50they're not guarantees. This is
30:51something I try to emphasize with our
30:53testing process at EDVT Media. I don't
30:55want to assume a creative's going to
30:56work just because we've seen that format
30:58perform well before. I don't want to
31:00assume that broad targeting is
31:01automatically going to be the answer
31:03just because it worked for the client
31:05before. I want us to develop a
31:06hypothesis, test it, look at the data,
31:09what does your customer say, and then
31:10use what we learned to make the next
31:12decision. We build the engine for you.
31:15We don't just hope that a strategy works
31:17and then we're spinning our wheels and
31:19not really sure where to go next when it
31:20doesn't. And I think this is where a lot
31:22of advertisers also become their own
31:23worst enemies. If you become too
31:25attached to a certain strategy and you
31:27listen too much to everybody else,
31:29you'll eventually start looking for
31:30evidence that supports what you already
31:33believe rather than objectively
31:34evaluating what's actually happening.
31:36You don't want to test interest because
31:38you think broad's going to work. You'll
31:39keep trying to make your favorite
31:40campaign structure work and wonder why
31:42it's not. You'll dismiss a creative
31:44format because someone told you it was
31:45outdated even though the data might be
31:47telling you something completely
31:48different. The whole process of testing
31:50is to eliminate the guesswork as much as
31:52possible. And the more you test, the
31:54more you learn about the customers
31:55you're actually trying to reach. You'll
31:57start understanding what messages work
31:59with them, which offers motivate them,
32:01which creatives generate purchasing
32:03behavior and just attention which parts
32:05of the funnel need improvement. When you
32:07combine all of that with the
32:08understanding of the business economics,
32:10that's where I think advertising becomes
32:12really powerful because the goal isn't
32:14to build the most complicated ad account
32:15or find a creative that generates a
32:17great return for a few days. The goal is
32:19to build a system or an engine, we like
32:21to call it internally, where you
32:22consistently understand why it's
32:24working, what needs to happen next to
32:26continue growing. And I think that's
32:28really the biggest lesson from
32:29everything we've covered. If you want to
32:31become better at advertising, you need
32:33to stop looking at it as just launching
32:35campaigns and optimizing ads. You need
32:37to fully understand how advertising
32:40connects to the entire business. From
32:42the first time someone discovers your
32:43product all the way through purchase and
32:46beyond. Because once you understand
32:47that, your decisions become much more
32:49intentional. You're no longer just
32:51reacting to whatever the dashboard tells
32:53you or chasing whatever new tactic
32:55everybody is talking about on YouTube.
32:57You're actually identifying the biggest
32:59opportunities to improve your
33:00performance and putting your resources
33:03towards [music] the things that matter
33:05most. And if you want to apply this
33:06judgment to the message inside of the
33:08ad, watch how to make Facebook ads so
33:10good they can't stop themselves from
33:12buying next. I built the full buying
33:14decision around one cleansing balm
33:16example from the customer's problem to
33:18the hook, product, proof, offer and
33:20landing page. Click the video on the
33:22screen and I'll walk you through

About this transcript

This page contains the full transcript of Give Me 33 Mins, I'll Give You 10,000 Hours of Ecom Ads Knowledge by Sergio C, generated from the public captions YouTube serves with the video. The transcript has 6,654 words across 1,053 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.