Jack (00:00)
What are sprints, dude? They sound sweaty.
Hal Smith (00:05)
Yeah, yeah, it's a great question. ⁓ The biggest thing on the sprint side is how do we have really focused effort in a predetermined amount of time where we are going to define some type of success metric at the end, right? And how do we kind of work in these blocks ⁓ so we can consistently move forward, kind of keep climbing up the mountain, and we're confident that we actually have some direction and we're making some type of progress. I think without having this kind of sprint structure.
A lot of work might feel like you're kind of spinning your wheels and it might feel kind of syspian, like you're kind of rolling the boulder up the hill and it's just rolling back down and you don't really know what you're doing. So I think it's really important ⁓ to set up things in sprints, to have clear goals and to just have a clear timeline on when you need go back and measure if you were successful against those goals.
Jack (01:00)
Nice. Awesome. And we're talking about how sprints relate to new customer acquisition, right? And how you guys at H Street are using sprints as a way to scale new customer acquisition for your brands, right?
Hal Smith (01:15)
Yep, yep, 100%, yep.
Jack (01:18)
Nice, awesome. We've got a few people in the chat here. Someone says, hi Jack. Hey Josh. Awesome, let's jump into it. You said that you have a deck.
Hal Smith (01:28)
Yeah, so I'll go ahead and I can share my screen and we start walking through it. And then also to Jack, feel free to stop me at any point. There's like a number of slides here. I think we got like seven or seven to 10. ⁓ So I will, you know, I'll stop any time you pause me and then we should be able to, you know, answer any questions or anything like that. But let me just make sure I have the right presentation mode up.
Jack (01:53)
For sure, Before we jump into it, feel like I really like the idea of sprints, know, a concerted effort to get something done over a limited timeframe. saying it's going to take this much time and we're going to get this thing done to achieve this goal. ⁓
That's if I'm understanding how you guys are using sprints at H Street. But I like that concept, right? As opposed to this long ongoing task that's undefined, it's regularly repeated, you have a structure, right? And a goal, right?
Hal Smith (02:25)
Yep, no, 100%. And I will, me if this is gonna work here. I'll share my screen and just go kind of slide show here. Is this showing up for you? Okay, great. Yeah, so I kind of go into it. First of all, like our kind of principle as an agency is to scale up new customer acquisition, figure out like the system for scaling new customer acquisition, all the different kind of levers to it.
Jack (02:35)
Yep.
Hal Smith (02:52)
And then we're trying to solve for constraints to ⁓ basically help a business scale new customers. And so our key performance levers are ⁓ creative offers, media buying, and then landing page CRO. And the whole principle is we want to change the amount of new customers we're driving by orders of magnitude. Kind of central to this is it's not really just setting up and quote unquote managing ads.
It's having these 90 day sprints. ⁓ And then with these 90 day sprints, we're looking at key constraints that might be going on with your creative, with your offer strategy, how your ad accounts are structured. And then also too, how well your landing pages or product pages are converting people. And then there's another kind of piece to this too, which is the reporting and attribution infrastructure. Like how accurate is it? ⁓ Is it getting you good information? Is it directional? ⁓
All attribution is very difficult to ⁓ understand. It's difficult to get 100 % accuracy. But it's like, is this actually giving you ⁓ enough information to make good decisions on what is working and what's not? ⁓ One kind of key thing with this is this concept like theory of constraints, which is basically just a process of identifying key bottlenecks or constraints within any given system.
And then having a process to systematically identify that constraint, diagnose it, figure out how to launch a test to really understand the impact of that constraint and how to solve for it. Once you get enough information back, like how to actually solve for it, implement some type of solution, and then remove that constraint from the system. And the way our kind of philosophy or principles as an agency is growth advertising is just that systematic hunt for those constraints.
having a really good diagnostic process, and then removing those constraints. And that really is where growth comes from and what is technically optimization. Before I go on, do you have any questions on that?
Jack (04:59)
No, you know what, I think that makes sense. We've got a question from Ivan here, and it's pretty vague, but are you retargeting? I think this might be in relationship to the new customer's part of this strategy.
Hal Smith (05:13)
Yeah, so ⁓ sometimes we do retargeting. ⁓ really kind of, it's a.
I would say we like to add it in if we truly believe it's going to drive incremental net new conversions. And a lot of the way we set up retargeting, at least on let's say Meta or Google, is we actually try to exclude existing customers and try to retarget page visitors or people that take certain high intent actions. We do certainly have it as kind of our normal playbook. And it can apply here, although I won't say that retargeting is kind of a lever to solve for constraints.
I mean, it doesn't really help you solve for a constraint. It is more just like additive and it's something that we've just learned. It's like, hey, this will be part of our standard playbook in campaign structure within ad accounts.
Jack (06:02)
Right, yeah, I couldn't agree more. I think this is interesting because I think a lot of the time that...
brands or agencies or marketers talk about excluding existing customers. They're not talking about not doing any retargeting. You're going to show more than one ad to each of your new customers. And it's probably not the very first impression, the very first ⁓ ad that they see that's going to get them across the finish line. So you are retargeting. We're targeting in all of your retention marketing, your email and SMS. I think the focus on new customer acquisition is really about expanding that active customer base so that you
Hal Smith (06:36)
Yes.
Jack (06:36)
have
a larger pool of new customers to drive for Tungus Mortars from.
Hal Smith (06:42)
Yeah, exactly. It's, it's, you know, the one concept that Metta kind of likes to talk about and something they'll provide for you is like rolling breach, right? Like how many net new people you're delivering the ads to in any given timeframe, right? But yeah, the key thing for us on the new customer acquisition side of things is like, how are we delivering ads to new people ⁓ more and more? How are we expanding your pool of potential customers? And then, you know, when those people actually come to your website, like how are we doing a good job of
getting them to actually convert and purchase, right? How are we getting them through the different stages of the funnel, et cetera? And then more importantly, and this is kind of where the theory of constraints comes in, is where are the biggest bottlenecks and what's happening there and how do we really identify that, quantify it, and then solve for it?
Jack (07:34)
Megan says hello. Hello Megan.
Hal Smith (07:36)
Nice. The other kind of key thing here in this 90-day sprint process and why we chose 90 days comes from just a lot of reading and kind of thinking around scaling businesses, right? So there are a lot of quote unquote like business operating systems and there are books on these systems. The most I would say commonly known one is called EOS, which is the Entrepreneurial Operating System. This book, Traction, describes a lot of the different pieces to it.
But there are other kind of important ones, scaling up and rhythm. But basically, each of these scaling methodologies, they always really focus around a 90-day sprint. And the kind of common theme here is humans have a lot of trouble understanding how to set goals and really execute against those goals in timeframes that are longer than 90 days. So it's really difficult to do annual planning. It's also really difficult to do like five-year planning.
So the best kind of timeframe to do any type of work as a collective group with like a team is 90 days. And that is something that is a clear through line with all these different types of like business scaling books. ⁓ So we kind of took this ⁓ signal and then we applied it to what we do to help scale brands in terms of new customer acquisition. So we effectively set up all of our objectives in these 90 day goals when trying to solve for different constraints for these businesses.
Another kind of, sorry, any questions on that? Another key thing here is ⁓ why we chose these levers. And so this just comes with a lot of trial and error and also to my personal experience around scaling brands. I found that there's kind of, there really four key areas that if you make optimizations in these areas, they will have the
Jack (09:11)
No? Let's keep moving.
Hal Smith (09:35)
biggest most dramatic impact on the success of your business, ⁓ especially as it relates to new customer acquisition. And so the first one is your offer. The more clear, the more urgent, the more valuable that offer is to your potential customers, the more impactful it's going to be, right? And the better and better you get at really optimizing that offer, the better everything else is going to be. Your ads are going to perform better, campaigns are going to get more efficient.
⁓ You know people convert faster on the landing pages So if you can have you can be super clear on something that you want to present to a prospective customer You know as to why they should buy from you ⁓ If you get really clear on that you will be successful and a lot of brands kind of gloss over this they just ⁓ Think hey if I just kind of present my product and say buy now or if you know submit your email you get 10 % off That's enough and the reality is especially with how crowded D2C is today how distracted most
customers are today, you have to get super, super clear on your offer and you need to make it really compelling. ⁓ Another key lever here, and this is what most people talk about on D2C, Twitter and LinkedIn, is creative. You need to have good creative quality, you need to have good creative variety, and then you also need to have a strategy for how you're going to create a content pipeline for creative and how you're to test it on these ad platforms.
⁓ If you're limited by creative, if you're only launching one or two ads a month, you're really not going to move the needle too much for your brand, especially if you're heavily dependent on digital advertising. ⁓ The other key thing is how you're structuring your ad campaigns for scaling, for optimization, ⁓ in meta, in Google, TikTok, or Snapchat. There is a lot of strategy to how to structure these campaigns.
⁓ specific to each platform and if you're kind of missing that, if you're setting up things that really don't present you the opportunity to understand different performance levers within the ad accounts and how to kind tweak them to make better performance, you're really not going to do too good of a job scaling up new customers. And then the last piece is, you you could do all this stuff to make your product and brand really attractive to your customer, but once they hit your website, it's a different story. That website ultimately sells the product.
and converts your customers. And if they can't navigate your site with enough, with no friction, or if they can't find what they're looking for, if they don't have the product sizing or colors they want, if it's difficult and there's too many clicks to actually make that purchase, no amount of work you can do on the front end is actually going to improve sales. So you have to be really clear about, sorry,
Jack (12:23)
I couldn't agree more.
Hal Smith (12:28)
Jack, do you have a question?
Jack (12:30)
No, I was going to say, just, I couldn't agree more. I think that, you know, a lot of people attribute the performance of an ad, right?
to the ad, right? There's this term written on ad spend, ROAS, right? But what if somebody clicks on your very good ad and then they see a white screen for eight seconds because your website's super slow, all right? Your ROAS isn't going to be very good. And that extends to every experience that they can have on the website. When people are shopping online, a lot of the time, especially for low-wav, low-av products, right? For, you know, impulse buys with a less discerning customer, people are in like a zombie mode, right?
Hal Smith (12:44)
Thank you.
Yes.
So, keep going.
Jack (13:08)
it's for somebody who's walking around a grocery store with a cart, they're just trying to get it done. They're just trying to get their groceries in the cart and go home. They're not thinking. You kind of have to think of your customer website in a similar way. Not that they're totally incapable, but that you want to make that process of going from discovering the product from the ad to having it in their cart and checking out as friction with the customer.
Hal Smith (13:31)
Yeah, 100%. There are two kind of big mantras I have that I repeat all the time. one is ⁓ ads will sell your click, right? They'll get people to click and go to your website. But your website sells the product, right? And you really have to think, you know, you have to think, is my website being a good salesperson, right? And you really need to, everything you do on your website needs to be ⁓ optimized in a way that actually gets people to purchase. And that's all that matters, right? Your website is a salesperson.
website makes a sale. there's, and then that kind of leads to my second mantra, which is ⁓ customers are not stupid, but they're lazy. Okay. And so you really have to make it as frictionless as possible. You really need to realize that people, you know, zombie mode, yes, like they're, they're just trying to get through and make that purchase. And if you're creating any type of friction, if that page is taking too long to load, or if it's really confusing, if you haven't provided, you know, key answers to objections on that page, people are going to leave, right?
And it's really kind of fault of how you structure that website or the landing page to not make it a good salesperson.
Jack (14:40)
Right. We've got a question here from Megan. I know we weren't all the way at the end of this list, but she's asked, is this all relatable to e-commerce and durable goods?
Hal Smith (14:52)
⁓ Yes, 100%. Well, ⁓ no, are you saying is it all specifically only applicable to e-commerce or can this be applicable in other areas?
Jack (15:03)
That's a great question. Megan, if you're still in the chat, ⁓ maybe we can clarify that. But I would say that although this is specifically about e-commerce, I think it maps on to different ⁓ categories as well. I mean, at least what we're looking at right now.
Hal Smith (15:16)
Yes.
Yeah, it's, I mean, we've, like, I personally have worked in lots of different areas. Like, we as an agency specifically focus on e-commerce, but this really maps to any area. I mean, this is the kind customer acquisition, you know, pipeline, right? And these are the key levers that you really need to be focused on and thinking about, you know, all the time and how to optimize each to improve any type of new customer acquisition. This could be like lead generation for something.
completely different, but these are always going to matter. ⁓ And the last piece too on the pixel and event quality side of things. today with how these ad platforms work with this heavy push into AI and machine learning, there's less and less you really have control over as an advertiser. And there's more and more that Meta and Google will kind of self-learn about. So the more and more you're feeding it good high quality data at high frequency,
the better and better these algorithms are going to be at kind of delivering the right ads to the right people. And so you really have to think, you know, ⁓ is a quality in the quantity and speed of this data, you know, good enough to actually allow these algorithms to self-learn and to get better and better over time? Or is this just, know, ⁓ you know, shoot, I don't want to use a normal term, but if you're going to put a bunch of crap into this machine,
you're going to get a bunch of crap back. So you really have to think about how you're fueling this algorithm.
Jack (16:49)
We've got a little one more question from Megan here. Do you suggest a long form landing page to link my ads to? ⁓ Before I just want to give my two cents here to Megan. think that it depends on the brand, but probably longer than the default that you get from Shopify, right? Which is just like three sections. And you may also like, you know, the product section at the top, ⁓ especially if your product really has to differentiate itself in some way, right?
But yeah, I'd love to hear what you think.
Hal Smith (17:22)
Yeah, so I'm to give a pretty bad non-answer. Megan, I would say that you need to recognize that this is a ⁓ potential test opportunity, and you need to effectively test ⁓ it. I mean, this actually is a great question for what we're to talk about for a couple of these other slides. But the reality is I don't know what's best for your brand. And what's best for your brand is to identify all the different
possible high impact solutions that you have to test as it relates to landing pages. So that could be short form, could be long form, pure product page, it's a Shopify standard, whatever it might be. And then, you know, what's more important is you go test those and you determine which one is going to do best for you and then really help to solve for that constraint, right? ⁓ But it's saying like, hey, as a general rule, long form is better than short form isn't best. Like what's best to recognize that this is a potential constraint.
and to come up with a couple of options to test and then see which one ⁓ proves out to be the best performing one, which one has the highest conversion rates.
Awesome. So just kind of going into the logic of this too and why these levers are important is one is this is kind of the complete journey for acquiring a new customer, right? They first see the ad, they go back to make that purchase. The second thing is anywhere, if you improve ⁓ KPIs at any of these kind of levers, it's going to have big downstream impact. So if you can radically improve click through rates, view through rates, et cetera from your offer.
perspective, that's going to do a lot better job actually improving your conversion rates on your landing page. ⁓ The other key thing to note is with testing, these levers are going to give you the highest performance variance. And so what that means is I didn't have listed here testing button colors on your landing page. And the reason for that is like button color testing is pretty low variance. ⁓ It's not going to have this radical big impact on your performance.
of a new customer cat. So you really want to figure out what actually has the biggest swings in performance and how do I test for that first? How do I create a really high variance when I'm testing for? A lot of what people say on the creative testing side of things is big swings, right? You want to have very distinctly different things you're testing because if you have a winner on one, it will dramatically improve your account. Sometimes you're going to have lot of duds, but at least it's going to give you a much clearer idea of what the potential
⁓ limits are to how you can improve performance. ⁓ The other key thing here is for most D2C brands, can own ⁓ each of these levers with really just one person operating it. ⁓ With everything we're seeing today with these different tools and platforms that are constantly getting improved and enhanced with AI, one person can do landing page optimization with let's say Replo, but they can also build out ads very easily.
⁓ in let's say Canva and then they can also do the media buying. So one person can really optimize each of these levers. And then the other key thing is ⁓ each of these performance levers ⁓ have a pretty clear KPI that you can track and determine if your tester is successful or not.
⁓ Any questions on this?
Jack (20:58)
Doesn't look like we have any in the chat, but guys, drop your questions in the chat and I'll get them over to Hal.
Hal Smith (21:06)
Awesome. OK, so really just kind of going ⁓ to the last piece here before we go into the actual 90 days and why the 90 days is important. It's just like the diagnostic process and the testing process around solving for constraints. So the key thing is figure out the biggest bottleneck. It might just be you don't have enough creative volume. It might be that your landing pages take 10 seconds to load and they should only take like half a second. ⁓
You first need to identify that key bottleneck and where that and that metric that kind of signals that it's off track. You need to come up with some type of hypothesis as to why it's happening. You need to figure out a specific solution to test in that should ultimately remove that constraint and then run that test and then eventually measure the impact of it. And then it's successful, obviously implement it. And so what we do as an agency is we have a diagnostic process that goes across those key performance levers.
⁓ We then, you know, come up with solutions. Sorry, we have the, we do the diagnostic. We come up with some hypotheses and then we have a couple of solutions. With each solution, we run them through what we call the ICE matrix. So it's impact, certainty and ease, basically to figure out and how to prioritize certain solutions. ⁓ And then we put those into a SMART goal format that's kind of overlaid on 90 days. So SMART is specific, measurable.
⁓ achievable, relevant, and time bound. We just make sure that we have a clear way of actually executing on what we want to test or solve for. And then once we run it, we hit whatever time frame, let's say it's in 30 days or 60 days, we go back and reflect on it and we're like, hey, did this actually solve the problem we're trying to solve? Did this actually remove this constraint? ⁓ And so what's really important here is there are certain things that you can do within 30 days.
There's certain things that just take longer. Like typically for some reason we find landing page optimizations or building out new landing pages takes longer. So that might take 60 days. There are other things might take 90 days. That might be like a couple of rounds of UGC ads that we're testing, right? ⁓ But it's really important. And what we ultimately do is through each of these performance levers, we try to identify like one to three key constraints.
⁓ a clear metric to kind of measure how that constraint is creating a problem for us. Then we map out how to solve for it over 36 or 90 days. And then we just basically work across that roadmap for 90 days. And at the end of each of those time periods, we go back and look and say, hey, did we actually solve for this? And then an example here is what this kind of looks like is, know, tip, this is what we do as part of our normal onboarding, but we'll do
In 30 days, we'll kind of identify some very quick wins we can execute on across each of those performance levers. 60 days, we do a lot more kind of aggressive testing and really kind of push, you know, as much as we can against, you know, performance and seeing where we can scale. Then 90 days, we typically cut back on something and then we double down on what's working, what's not. And we, you know, from that kind of initial run, then we'll start building out more methodical 90 day sprints.
to kind of address each of those areas. And then some of the things that we have done here is kind of listed. So like on the offer side, it's like testing seasonal offers versus value stacking. Those might be just certain ways to kind of optimize that offer. ⁓ Let's say, let's skip over to like CRO. That might be like setting up a more simplified checkout, setting up like payment flexibility.
for purchasing, whatever it might be. But these are kind of small tactics that would be like, hey, we're going to test this and see if this helps us improve conversion rates. Yeah, and I say just too, this is kind of a simple example. And these are kind of, I would say, inflated numbers just to make the point obvious. But let's say these are several rounds of sprints.
You know, we might be working with a client this first row here is the baseline of what we're seeing across a couple of key KPIs We might then start doing some creative optimizations, you know have a 90-day sprint around creative That will kind of push down our CPMs. They'll improve our click through rates and then you can see here There's a pretty big improvement on that CAC, right? Maybe the next sprint we do is a lot of testing around creative and audience optimization That kind of pushes down our CPMs more it makes our campaigns more efficient
Meta, Google might favor these ads more, our click through rates start to ⁓ jump up a bit more, 50 % increase. Conversion rate also significantly improves because we're targeting a much more relevant, higher qualified audience. All of a that cap drops pretty significantly. So this is really what it looks like and I would say for sure we've seen, ⁓ I would say these kind of big gains, especially when launching a brand.
But this is really like how this becomes a methodical process and something where you can consistently kind of solve for these constraints and scale and see better and better performance month over month or basically over each of these 90 day time periods. So ⁓ that's all I got for you, Jack.
Jack (26:45)
Sorry, it sounds like we have a little bit of a delay there. That was awesome though,
Hal Smith (26:48)
No worries. ⁓
Yeah, for sure. Yeah, definitely happy to answer any questions. hope I, I don't know exactly how much time we want to dedicate to this, but I'm happy to answer any questions and kind of go into any detail.
Jack (27:02)
Sure guys, if anyone has any questions, feel free to drop them in the chat. I guess I have some questions around the new customers piece and ⁓ your mentality towards that.
And specifically how, you know, new customers and the sprint structure is helping you there. Right. Cause you can imagine if there's a brand that's listening right now, or there's an agency and their team. And they're thinking, you know what, when I run a retargeting campaign, I just see so much better performance. My CAC is lower, my CPA in the, in the account, not like a blinded CAC. My ROAS is higher, right. I'm showing ads to people who are more likely to purchase.
already engaged with the business, I'm going to see a higher rate of return. How, why should I be, you know, using sprints to acquire new customers, right? My performance is going to take a hit if I do that.
Hal Smith (27:58)
Yeah, I mean, there's a lot of threads to pull on in that question.
So number one thing is what are your business goals? And I always really try to get this figured out early on in, let's say, a client relationship. Because if their goal is maximizing revenue, that can be very different from maximizing profit. And if it can be incremental revenue growth, that can be different from overall revenue growth. And so it's kind of a big wind up here.
When you're looking at ad accounts and you're seeing the campaigns that are driving the best ROAS and you're doing retargeting, my kind of gripe here is that typically there are two things that are happening. One is your existing customers are always much more likely to purchase from you. So it's always gonna be more efficient to deliver ads to your existing customers because they're much more likely to be like, I know this brand, I like their product, I'll buy. And Meta will deliver, if you don't prevent it from doing it, it will deliver ads to existing customers. The other challenge is,
⁓ There's something called, ⁓ I think it's called the selection effect. But basically Meta, a lot of your existing customers are gonna come back and make a purchase anyway. Some of them might even be subscribers or just high repeat purchasers. And what happens is when you do retargeting and remarketing campaigns, a lot of times Meta will recognize that person will come make that purchase from you and deliver ads to them when that person is gonna make a purchase regardless.
Did I lose you?
⁓ you still here? Yeah.
Jack (29:44)
I am Jack. Yeah. I'm switching between
two different wifi's to see which one's the strongest.
Hal Smith (29:49)
Yeah, no worries. So what I like to do is I like to split focus, and this sounds bad, but I like to talk to a brand and be like, what do you want to do to drive revenue from existing customers? And what KPIs are we going to have? And how are we going to track that from a business perspective? And then what do you want to do to acquire new customers? And I'm going to split out campaigns based off of that and split out KPIs.
and performance metrics because what you're doing to acquire new customers and build incremental revenue is always going to look much, much worse. Your ROAS is always going to be lower. ⁓ And what a lot of brands do today, and this is really important in this DTC e-com space, is to recognize that ⁓ the most successful DTC e-com brands are ones that have high ⁓ repeat customer purchase rates and some type of trigger or subscription lever.
And so what you're really doing is you're playing the game to acquire customers at a potentially high CAC that might even be like break even or barely profitable, knowing that that person's going to have a high lifetime value. And you have to figure out that ratio. Typically people talk about like that, that three to one ratio. So your CAC is like basically a third of your expected LTV. ⁓ And so you really have to start thinking in those different directions and you have to recognize, and then it's like, if I'm going to go,
advertise towards existing customers and do a lot of remarketing. ⁓ Generally, your ROAS should be really high. Your expectations should be super high because you're going to go pay money to get customers who already purchased from you to go buy again. Frankly, I like to leave that to the retention teams, a retention agency, email marketers, whatever. It's like, hey, you guys go do the work to get these people. I don't think you should really be spending money on ads to target your existing customers, frankly.
Jack (31:39)
Right, you're not doing very much marketing. You're taking product to market. Don't get me wrong, I know that retention marketing is marketing. But if your investment in the growth of your business is primarily spent on people who've already shopped from you before, you're not really, I mean, growth might be a little bit tricky unless you have incredible word of mouth. You're just relying on different channels that might grow as a result of retargeting your existing customer base.
Yeah, it's an interesting one. think I really like that point on having an expectation because you might have a different expectation from new customer than return customer. You might also have a different expectation from different product categories or your best selling products as opposed to your worst selling products. You still have to sell your worst selling products. You still have to move that inventory. Right. So what is the expectation there? In my experience, a lot of agencies, a lot of marketers will over focus and they can be wrong. Best sellers are probably the best bait when you're fishing for new customers, right? Because they're the product that's proven in
Hal Smith (32:21)
Yes.
Jack (32:37)
market or new arrivals for exposure to a potential new bestseller. But there's still inventory that we've got to sell. Sometimes the least popular inventory is not so fun, but we've got to move it.
Hal Smith (32:52)
Well, that's what I do think. We like to focus on a hero product, right? We like to focus on hero product and then build kind of a hero offer around it. And I just know that's gonna do the best job of ⁓ acquiring a new customer. Yes, you still have to sell other products, other inventory. Some brands have like thousands of SKUs. ⁓ However, for what we focus on a lot is like, what's the most efficient way to get somebody to purchase from the...
like and leave it to somebody else to figure out how to upsell, cross sell, get them to come back, et cetera. But it's like, I think if you're really, really good and efficient in acquiring new people, that really requires you having one single kind of hero product or maybe a couple and then having a really strong offer that's built around that hero product.
Jack (33:39)
Hell, I got to tell you a couple of stories. This is so funny. I used to work with a cheese brand and they had, you know, this very popular cheese because all of the creative for it was incredible. You know, there's very ⁓ indulgent, you know, liquid cheese that when you see it in the videos and in the ad creative, it was just like sensational. Our CAC was so low on this cheese and, but all the comments on the ads were, you know, it was a mix of people who had already purchased it before or posting photos sometimes.
Hal Smith (33:45)
Okay.
Jack (34:09)
I remember there was this one comment that said, how do I get it to look like this? It turned up and it was like a brick of mayonnaise. Because it was like white cheese and it wasn't soft and gooey, you know, as it arrives, right? You have to heat it up. And it was really bad for the business. mean, a low cack for sure. Lots of, you know, they moved a lot of that cheese, but it's a low margin product that damaged their reputation. That didn't lead to people coming back. And then there was another product that had slightly higher cack, right?
Hal Smith (34:18)
Yeah.
Jack (34:38)
that was sensational, right? It was very good at getting people to come back, but it wasn't as like, I would say the town was smaller because it was a blue cheese. Not everyone likes blue cheese, but it was very, very good at getting those blue cheese lovers to come back to the brand. I have another story I want to tell you because I think this was funny. I used to work with this brand and they had like funny t-shirts, right? Like with slogans and stuff on it.
And one of those slogans kept getting banned by Meta. It was just way too fringe. It was way too raunchy. Right. And I remember back then, you know, it my job to have the conversation with Meta support, like, unblock this product. I want to, I want to sell this product. And then they just, released new products, right. All the time. And, you know, testing different designs. And now to this day, one of the new product releases makes up like 95 % of their sales. They're an entirely different business. They were like 150, 200 a month. Um, when
I was trying to get this product unrestricted. And now they're doing like plus 10 million a month. They're a huge brand. So it's just interesting to think about how that play on the offer front, but not just the promotional offer, but the product offer can be so game changing for the business.
Hal Smith (35:47)
Yeah, 100%. Yeah, think it's frankly like it's kind of everything. mean, that's I think fundamentally, especially like let's say e-comm if you're trying to come up with a potentially like distinctly different product. You know, that's why there's this whole concept around product market fit, right? Because if you do have a really good product offer that is really compelling, really interesting, really taps into some, you know, kind of strong ⁓ customer cohort.
I can really change your business. can be like, you these people are rabid, you know, supporters, you know, love this specific product, whatever it might be. But ⁓ if you can find that it's great. ⁓ And it really does a whole lot. And that's why I feel like a lot of econ brands, like they do kind of focus on, you know, say one or two hero products, because they know for sure that these are the things that, you know, get people in the door, they might buy.
a lot of other different skews eventually from that brand. But it's typically just one thing that gets them in the door. And I've seen this actually with app advertising as well. Typically, outside the gaming space, like with other types of apps, you really want to focus on one feature. And he had like a gateway feature. It gets people most interested in coming in and downloading the app and becoming a paid subscriber.
But typically what keeps people sticky and keeps them around for a long period of time and like long-term paid subscribers are typically other features in the app. It's not actually the one that you use to market and advertise for, which is pretty interesting.
Jack (37:21)
Yeah, I can agree more.
Yeah, that's why I like what you were saying before about big swings, you know, whether it's a big swing with your website, just completely redesigning your website, the amount of Shopify stores that are running the same theme that they just like bought a couple of years ago. And it's been working for them because they're doing something great with product or they're doing something great with their marketing.
But their website's never had a refresh. Take a big swing. Redo your website. Make it a website that you're genuinely proud of. And then maybe get some experts involved so it can be a website that performs well too. Or the same thing with product. I've seen, and I think a lot of the founders that I've worked with are really passionate and closely connected to the problem statement that they started their brand to solve. ⁓ I don't know, maybe it's a skincare product that deals with the skincare problem that they have. Or it's a supplement.
Hal Smith (37:49)
Yeah.
Jack (38:16)
that deals with a health problem that they have. ⁓ It can be hard when you're that person to look at shifts in what the market is asking for. And it could be just a small sidestep in the way that you talk about the benefits, but it's the same product or a small change and innovation in the product that unlocks a huge amount of market share.
Hal Smith (38:39)
Yeah, no, 100%. I think too, kind of going back to what you're saying on big swings or potentially doing a website update, what I found to be really important is you need to figure out your main KPI to track the performance of each of these areas, right? And so for, let's say these ad platforms, I mean, it's generally like your kind of main ad KPI, it's your CPMs, your click through rates, and then it's the kind of other key metrics like CAC and conversion rates.
⁓ But once you have an idea of kind of what this benchmark is in each of these levers, and so like landing pages is really like time on site, ⁓ know, conversion rate, ⁓ AOV. But what you want to do is basically set up a benchmark for your brand in each of these areas. And then you need to come and basically do some research or thinking around like how far off are these metrics from like what I should expect. And so like a typical range for a Shopify
store in terms of conversion rates is like two to 4%. If your Shopify store is doing less than 2%, you have a conversion rate problem. And so a lot of your effort, time, and energy should just be focused on getting your conversion rate higher. If you're running ads and your CPMs are through the roof, which happens a lot in certain industries, that's why benchmarking is important. Because I know for supplements, beauty,
those CPMs can be really high on Meta at least. It can be like hundred bucks or more. But at least have an idea of what the benchmark is for these metrics. And then if it's outside of that, outside of some normal standard deviation, then it's like, need to really focus here. So let's say your brand in your industry, your CPMs are typically like 50 bucks. And then in your ad account in Meta, your CPMs are 250 bucks. You gotta figure out there's something going on with your campaign structure.
or your creative quality, or your bidding, your budgets, et cetera. But you have to go and be like, all right, so where is this problem and how do I solve for this? But it's really important that you're not just randomly being like, I want to go update the website, or I'm going to go just do new types of ads on Meta. It's like, what are my key KPIs to measure performance? What's a normal benchmark expectation? How far off are my KPIs from this normal expectation? And then do I, you
if they're really far off, that's where I want to focus. If they're not, don't worry about it. Go focus on something else where it is really far off. ⁓ Does that make sense?
Jack (41:12)
Yeah, 100%. Yeah, for sure. We've got a couple of questions here. One from Megan again.
Megan, thank you for asking so many questions. What about when you launch a new product and previous customers haven't subscribed? Right. That's a great question. I'm not sure what you mean by subscribed, whether it's like subscribe to your email list, SMS, push notifications, or it's subscribed to like a subscription. But if they haven't subscribed to your email list, it might be worth paying attention to your pop-up. Right. I think this is one of the ways that compatibility in marketing teams is so important. Right. If you have a ad.
Hal Smith (41:44)
Mm.
Jack (41:50)
team, ad agency, an email agency, and then a web agency who are helping you out with these different parts of your go-to-market, someone needs to pay attention to the pop-up. Normally it's the email marketing team, because the pop-up might be in Klaviyo.
Maybe you don't want it to be in Klaviyo. Maybe you want the web team to pay attention to the pop-up. Maybe you want the acquisition team, because you need that congruency between what's being said in the ad and what's being offered when they land on the website. But man, the performance improvements I've seen from brands that just change their pop-ups a little bit, they have like a 2 % conversion rate on their pop-up from a new site visitor to a subscriber, and they move that to 4%. That's twice as many people subscribing to your email and SMS list. And if you can move it up to sometimes
think it's decent, like 10%. That's going to be game changing for the business. It's probably going to mean that you're to be able to spend more because you're getting more of those new site visitors across the finish line because you're able to hit them with retention marketing. Yeah, that's my two cents. What are you thinking now?
Hal Smith (42:50)
Yeah, so glad you mentioned pop-ups because I totally missed that and pop-ups are so powerful. ⁓ They can do so much for your brand. ⁓ Totally agree. I think one other interesting thing about pop-ups is you can collect really ⁓ insightful, like true signal from your customers or potential customers ⁓ with pop-ups. ⁓ You shouldn't always just treat pop-ups as just a way to give some promotion.
But you can think of ways to potentially add in a couple of quick questions in your pop-ups. So you're actually getting better information around, you know, who's hitting your site and who's actually ⁓ filling out the information on those pop-ups. The other thing that's interesting too about pop-ups, which I recently learned about, ⁓ is it's actually not best to just offer immediate discounts on your pop-ups because there's kind of a hidden cost to that where it's like that discount, if you didn't offer that immediately, it might not even... ⁓
it might not have even actually impacted someone's likelihood to convert. And everyone sees the same kind of discounting across all websites now. It's like 10 or 20 % of discounted. So it's actually, my recommendation is don't do any type of monetary discounting, but potentially offer some other value that's just not an immediate discount. And frankly, you might actually get like 10 to 20 % back in terms of revenue just by not offering that at the front end. And so what I mean by that is like you could offer,
you know, something to get into a giveaway or some type of sweepstakes. ⁓ Also some type of exclusive like membership perks that isn't necessarily just immediately like 10 to 20 % off because you're effectively taxing your own business by offering that on the pop up to everybody that hits your site. ⁓ And I don't think it's like a distinguishing factor anymore. So yeah.
Jack (44:33)
Riot.
And I think that tax can sometimes be way more expensive than a free gift. And the word free is sensational, right? Like a 20 % discount on a $200 product.
Hal Smith (44:42)
Yeah. I know. Yeah.
Jack (44:46)
You might as well give something for free, not a $200 product. But then you get to use the word free, and you get to treat people well. What I like in pop-ups recently is asking when it's their birthday. And pretty much everyone says, it's my birthday tomorrow, or like right now, give me the discount right away. And it just increases, mean, what do they call it? Micro opt-in, micro commitment, something like that. Just increases the likelihood that they're going to participate because they're gaming the system, right?
Hal Smith (45:17)
Yeah, no, 100%.
Jack (45:20)
All right, we've got a few more questions here. Megan said, is, who is it?
Hal Smith (45:22)
⁓ yeah, I I ⁓ would say
quickly to that to the previous question to like for if you're launching a new product and I would say always for sure go launch that product on your own audience So for sure always send it out to your you know Even if you don't have some customers like if you're still have like an email audience and text audience go launch it there first For sure you can do some remarketing to your existing customers with that new product
But I would also say too, I mean, effectively, you have to go take that to market. So you're to go launch that with prospecting campaigns as well.
Jack (45:58)
Yeah, I mean, if you have limited inventory on your new release, I know this doesn't apply to many brands, especially not the brands that are scaling, but you might as well release it to your existing customers first, right? Because it's going to be the most affordable channel. And if you can run out of stock with just your existing customers and never have to launch an ad, beautiful, right?
Hal Smith (46:03)
⁓
Yeah,
yeah.
Jack (46:21)
So we got a
few more questions here. We got one from Megan that says, I have a pop-up with an offer, but it's not converting. I've been thinking about gamifying it. Megan, how are you thinking about gamifying your pop-up? What are you thinking? What's the idea? We'll wait a second to see if Megan lets us know what our plan is. We got one from Josephette. I hope I'm saying that right.
What are clear indicators that our ad creative needs a refresh? We're on TikTok, meta, CTV, Pinterest. Does this vary by platform and how often should we be testing new creative to proactively avoid fatigue?
Hal Smith (47:03)
Yeah, this is a ⁓ great great question. It's a It depends on a lot of things one thing very simply the best way to look at fatigue is just ⁓ Frank frequency ⁓ So my kind of general rule is if your frequency is like six or higher you're fatiguing ⁓ Also to if you look at it over time and you actually get like penalized by meta
You'll see CPM start to creep up on ads that have been live for a really long period of time that have fatigued and you'll see click through rates start to drop off So what you need to look at is just over time like let's say a long enough period of time 30 60 days. What's your frequency? If it's over 6, yeah, you probably need a refresh and then also to like over the same time periods You need to look at the trend of CPMs and click through rates
click through rates start to really drop off, CPM start to really increase. Yeah, you're probably having a fatigue issue. But I would say too, if your ad is driving the KPI you want, so that target new customer acquisition costs or target ROAS, don't shut it off, just keep it going. Even if you have a super high frequency, if it's still driving the performance, don't worry about it. But the other thing too is ⁓ how frequently should you be testing new creative?
What we like to do is we'll have like a scale campaign, which is all of our top performing ads. We'll just have a pretty high number of them, like let's say like 10 to 15. Let's say Meta specifically, and we'll just kind of run those and we'll let them run until they blow up basically. And especially with Meta, like you can keep top performing ads running for very long period of time. I've had a top performer in one of our client accounts for like a year now.
And we're not too worried about it. we're running pretty high level amounts of spin on that ad. ⁓ In terms of testing though, like our testing cycle and frequency is either it's weekly or bi-weekly based on the rate of spin we have. And so really what you should be doing is consistently thinking of new angles you want to test ⁓ in two areas. One is like, how do you want to iterate off of top performers? And what kind of key things do you think are going to be the next best step in that iteration?
And then also too, you want to be dumping in some of those big swings. So like net new very distinctly different concepts. ⁓
So yeah, guess the follow up question there, I'm seeing it here now too. So either some clients will do weekly, and so we'll just run a test for that week with Creative, and then others will do it over like two weeks. And it really depends on the amount of spend you have and your KPI. So like all tests are really defined by the amount of ⁓ conversions you need to have to hit significance. ⁓ And so if some of our clients are a little bit lower spenders,
And so we need to stretch out that spend a little bit longer. Some of them are higher spenders, and we can get data back really quickly and get it within a week. Also, too, it depends on your KPI. Some of our clients' KPI is $100 in your customer acquisition cost. So that, takes a little bit longer. Whereas other ones are $15 CACs. And we learn answers to those tests really fast. So if it's high rate of spend, low CACs, super fast learning. Low rate of spend, high CACs, really slow learning.
Jack (50:22)
Right.
Right, I think the answer as well here, ⁓ Simone, is you can launch tests too fast. If you're making changes every single day, that can be too fast. And yeah, you can go too slow too, but you just want to make sure you're reaching significance before you're making huge changes. ⁓
Let's see, Simone, I'm gonna put your next question on the screen, but I'm gonna go back to ⁓ Megan and Lucy, because they've got a couple questions here too. How can I find benchmarks for my brand? So can I plug? I'm gonna plug. ⁓ Guys, you can find... Nice. Yeah, guys, you can use Lens. ⁓ We have an integration with Triple Well.
Hal Smith (51:08)
Yeah, plug away because I was about to plug for you, so go ahead. ⁓
Jack (51:17)
and 30,000 Triple Well integrated stores for benchmarks. So you can come into your Lens account in foreplay and see where you match up against cohort match brands. So brands that are in the same niche, the same AOV range and the same revenue range, right? On an annual basis. So yeah, right side is better, left side is worse.
For CPA for this brand, we can see that we're on the right side, despite the CPA being lower. That means that the CPA is better than the average, the benchmark right now. And these are benchmarks that, if you go on Google and you just Google benchmarks for brands and then you find a Google link, you open up a blog, that blog might have been written in 2022, 2024, and not be an accurate representation of actually what's happening in the market right now. Whereas these are moving benchmarks from over 30,000 Shopify stores.
So these should be pretty accurate to the sort of brand that's competing with you in your space.
Hal Smith (52:20)
Yeah, that's great. And I would say that this data is super powerful, especially from Triple Whale and also what you're aggregating, which you guys have on Lens because to your point, it's actually really, really difficult to find good benchmarking just by Googling. We have an advantage as an agency because we get to look across a ton of client accounts and we have internal benchmarking that we've set up as an agency. But it's a great question. It's actually extremely difficult. if you see, like there's like, top like, it's like
CPMs by channel and it's like some random kind of blog and it's like so inaccurate too So it really is important by the the brand category ⁓ Like what you're looking at and it's very like at least on meta CPMs can vary pretty significantly by different categories
Jack (53:08)
We've got another one from Simone here, getting very technical now, I love it. Okay thanks, do you usually do ABO or CBO? I think this is for testing.
Hal Smith (53:22)
Yeah, so I like CBO. What I like to do when I do testing is I like to, at least on Meta, so this is Meta, but one campaign CBO, one ad set broad, and then a bunch of variations of distinctly different ads. And then let Meta figure it out. What I've learned is best is not to try to control for it at all.
And don't like when people do ABO, they like set up these different ad sets and different budgets for each and kind of these bid optimization mechanics. Like I think that's like over engineering it. I think meta is incredibly good, fast and powerful at figuring out like the best ad to be allocating budget to. And I know this for sure, because it's like this concept is called a multi-armed bandit testing. But effectively what meta does is it's like the analogy is you walk into a casino.
and you have a bunch of slot machines, somebody like comes up to you and it's like, hey, one of these slot machines is rigged and they're like 10 slot machines. And so you have to figure out the optimal way of figuring out which slot machine is gonna give you the biggest payout. So what you do is like the kind of simple algorithm is you take 20 % of the money you're gonna bet. And so let's say you have a thousand bucks, you take 200 bucks and then you put it equally to all 10 slot machines. Whichever one gives you the best payback in that test period.
you take the remaining money, that 800 bucks, you dump it into it. And that gives you the highest likelihood of finding the slot machine that is rigged, quote unquote. Meta effectively does the same thing. And you can see this when you go launch a campaign. But effectively, you'll see all of these ads basically get like 10 to a couple hundred bucks to spend. And all of a sudden, like one or two ads will get like thousands of dollars or tens of thousands of dollars. so Meta is kind of doing that on steroids with a much more complex algorithm. But that's effectively the same concept.
And so I like to use CBO for creative testing and that very simple structure to give Meta all the leverage, as much leverage as possible to figure out winners.
Jack (55:21)
Yeah, I have a fun analogy for this too. Yours is way more technical than mine, but mine's just a competition. I'm not sure if you watch any sports, ⁓ but like if...
You only had one competitor in a sporting competition. It'd be very clear who would win, right? You'd putting all your chips on black. This ad better work for me. But if you have a tournament, right? And lots of different teams competing against each other, the losers lose and the winners win. And don't get me wrong, right? It could be the better of two teams on that given day, right? On that field, right? ⁓ But you are finding who is winning the, you know, a sequence of bouts.
Right, a sequence of different competitions and you're surviving what's good. And we have the same competition in everything that we do. You know, if you're good at thinking, you get rid of bad ideas and you keep good ideas. Right. And if you're good at acting, you act on the good ideas and you don't act on the bad ideas. Right. So it's a, it's the, I think, I think that you want to create that competition. If you have ABO, right. And you have lots of different campaigns and have lots of different ad sets, you have lots of different platforms, maybe that you're spending a little bit on.
sure you're diversifying into different locations. And that might be a really good way to find opportunities, but you're not putting these things into direct competition with each other. And on Meta, we're not doing manual marketing. We're not out on the street with pamphlets on our hand giving them to people who look like our customer. We're actually actively relying on every second basis on Meta to do that for us. And that's the reason that they're very good at it. That's the reason why so many brands spend so much on Meta.
a terrible idea to trust meta, it's going to handle that competition quite well.
Hal Smith (57:05)
Yep. No, I would say Meta is incredibly powerful at setting up that competition and figuring out winners really, really quickly. The other thing too is I will launch and throw away campaigns really quickly. So I'll have that simple setup just to test a big batch of ads. I'll figure out a winner, shut that campaign down, grab that winner and throw it into a scaling campaign. like,
I don't do as much media buying anymore, like a lot of my team handles this, but like if you ever look at an ad account that I've been in, it's just like tons and tons of dead campaigns. Cause I will quickly like launch these and test them at the campaign level, CBO, and then take the winners and keep pulling them out and then put them into this bigger, ⁓ better structured campaign that's still very simple. But it's just like all the winning ads are in that campaign.
all the kind of logic is put into it. And generally that campaign is incredibly consistent at higher higher levels of spend.
Jack (58:02)
Yeah, I wanna say one caution, Simone, Megan, Josabeth, is don't double and triple down on something that's broken, right? So I'm here saying, hey, maybe don't have like 600 ad sets with a $3 budget, but doubling down, tripling down on a bad campaign is probably not a winning method either, right?
Awesome. Guys, do we have any more questions before we wrap this up?
We do actually have one more maybe that we could answer before. We call it a day. Does Meta have an audience that can recognize return customers or do we have to filter out to match our audience?
Hal Smith (58:51)
⁓ So you can create that audience ⁓ using custom events. ⁓ We do it with ⁓ a server to server tracking solution. Like there are a lot of different ones. There's ⁓ Elibar, there's Blotout, there's Popsicle. There's also the Triple Whales, ⁓ Sonar, there's Northbeam Apex. However, you can create a custom event that's like existing customers or new customers and then pass it back. You can also just create a basic purchase event.
and that should technically be all existing customers. But you can create these custom events and then neither exclude them or target them explicitly. So it is pretty simple to set up. I like to do a server to server tracking because you can create a little bit more nuance in how you're doing it. So like we can set up custom events for like specific product purchases or if they became like a subscriber, et cetera. yeah, a lot of options there.
Jack (59:44)
Yeah, I'd say if you're going to go down this road of creating a new customer conversion events or conversion events for different product categories and types of customers, don't use Zapier. Save yourself the headache and yeah, go with it.
end-to-end service solution. One other thing, I think that this is a part of the auditing process for lot of marketers when they start working with brands. Define your audiences in your ad account, right, as soon as possible, and then do an account level breakdown of where you're spending, right, and where you're seeing your return as well. As long as you have your tracking in order, you should have fairly accurate numbers in your platform. And then maybe you'll find that, I don't know, 50 % of your budget is on existing customers, right? And yeah, that could be
worthwhile to take a look at.
Awesome. Is that a wrap, pal?
Hal Smith (1:00:35)
That's a wrap. I appreciate you having me on, Jack. I thought this was awesome. ⁓
Jack (1:00:38)
You too, man. Thanks for coming on. And thank you everyone who
joined. Appreciate you guys. Talk soon, guys.
Hal Smith (1:00:44)
personally.
Thanks guys.
