Limited Supply
Limited Supply

Build a Q4 Landing Page With AI

3h ago32:536,505 words

The Podafi briefing · AI analysis

AI tools accelerate landing page creation for ad angle testing

5 min brief · 4 takeaways · Based on this episode’s transcript

The 30-second brief

TL;DR
  • 01AI coding tools build landing pages in minutes, bypassing multi-week agency delays that invalidate ad angles.
  • 02Untrained AI generates poor pages; operators should input specific brand modules and customer research for quality.
  • 03Meta ad feedback loops identify winning angles, which then dictate the specific landing page structure.

The big picture

Traditional agency landing pages take weeks, causing ad angles to expire before launch. Operators can use AI coding tools to build conversion-focused pages in minutes. This approach requires training the AI with specific brand modules and customer research to avoid generating low-quality outputs.

Useful for: Direct-to-consumer performance marketers seeking to accelerate landing page creation and improve cold traffic conversion.

Synthesis of podcast conversations. Speaker claims are not independently verified.

01Conversion copy

Answer five questions before checkout

Nick Sharma says landing pages should answer five questions: what is the product, why should the consumer care, how fast is shipping, how does it improve daily life, and how does it compare to alternatives.

Why it matters. Addressing these specific consumer doubts may reduce friction by proactively answering objections before the buyer reaches the checkout module.

Your next move · Podafi’s suggestion

Audit your current top landing page to ensure it explicitly answers these five questions in the primary copy.

The catch. Shipping speed claims should be accurate; promising next-day delivery during Q4 peak volumes without operational backing will damage trust.

Transcript evidence

From the transcript

“Question number one, what are you selling? Explain it to me, like a five year old, who doesn't know anything”
Read the source transcript ↗

From the transcript

“how fast do I get it is a really big lever to increase conversion on the page”
Read the source transcript ↗

Evidence summary · paraphrased

Sharma outlines a five-question framework for landing pages, emphasizing that copy should explain the product simply, justify the consumer benefit, clarify shipping times, highlight life improvements, and compare against alternatives.

02AI creative

Train AI with specific page modules

Sharma warns that untrained AI generates low-quality pages. He recommends feeding AI specific modules like a five-reason listicle, a comparison table against non-direct competitors, and visual demonstrations to match brand standards.

Why it matters. Providing structural constraints may prevent the AI from generating generic layouts, helping the output align with direct-to-consumer conversion practices and brand identity.

Your next move · Podafi’s suggestion

Create a text document listing your required page modules and feed it to your AI coding tool as a foundational prompt before generating new pages.

The catch. AI tools still require manual setup for tracking pixels and measurement partners to ensure data accuracy before publishing.

Transcript evidence

From the transcript

“adding the right modules and takes your page to the next level”
Read the source transcript ↗

Evidence summary · paraphrased

Sharma notes that basic AI prompts create generic pages, so he inputs specific modules like sideways-scrolling reason lists and comparison tables to elevate the design and conversion potential.

03Acquisition research

Use ad data to find page angles

Sharma suggests analyzing Meta ad metrics like cost-per-add-to-cart to find inefficient spend, then building challenger landing pages for those specific angles. He also recommends tallying customer review mentions to find top product benefits.

Why it matters. Aligning landing page copy with proven ad angles or actual customer language increases relevance, which may improve conversion rates and lower overall acquisition costs.

Your next move · Podafi’s suggestion

Export your top customer reviews and tally the most frequently mentioned product benefits to use as the primary headline for your next landing page test.

The catch. Review data reflects post-purchase sentiment, which may not perfectly align with the pre-purchase motivations driving your cold traffic ad clicks.

Transcript evidence

From the transcript

“open your ad account, figure out what, where are you spending the most amount of money and where is it the most inefficient”
Read the source transcript ↗

Evidence summary · paraphrased

Sharma advises finding the weakest performing ad creative to build a challenger page, or tallying early product reviews to identify the most mentioned benefits for a new listicle page.

04Conversion optimization

Dedicated landing pages lift focused conversions

Sharma noted a hair care brand saw a 10% conversion rate on a dedicated landing page compared to 2% to 3.5% on a standard collection page. The landing page removed outbound link clicks to create a focused experience.

Why it matters. Removing navigation distractions keeps buyers focused on a single offer, which can increase return on ad spend when scaling paid traffic to specific audiences.

Your next move · Podafi’s suggestion

Identify your highest-volume collection page and build a dedicated landing page for its top-selling bundle, removing global navigation links to test the conversion lift.

The catch. This tactic works best for specific bundles or promotions. General browsing shoppers might bounce if they cannot easily navigate to other products.

Transcript evidence

Evidence summary · paraphrased

Sharma shared an anecdote about a hair care brand testing a bundle on a focused landing page versus a collection page, noting a jump from roughly 3% to 10% conversion.

From listening to doing

Take one idea into the week

Suggested experiments, not proven results. Choose what fits your brand.

Test AI-built challenger landing page

  1. 01Identify the highest spend, lowest converting Meta ad angle from your recent campaign data.
  2. 02Use an AI coding tool to build a challenger landing page addressing the five-question framework for that specific angle.
  3. 03Run the original ad creative to the new challenger page while sending a control group to the existing product page.

Measure: Compare cost-per-add-to-cart and landing page conversion rate between the challenger page and the control product page.

Guardrail: Stop the test if the challenger page increases cost-per-add-to-cart significantly after reaching a statistically relevant number of add-to-cart events, indicating the new angle is underperforming.

Context & limitations
  • The speaker's 45-minute page build time assumes prior familiarity with AI coding tools and existing brand research; beginners will require significantly more time.
  • AI-generated pages require manual implementation of tracking pixels and behavioral analytics tools to measure incrementality accurately.
  • The speaker notes AI outputs are only as good as inputs; using this method for a homepage rather than a specific landing page may yield poor results.

Listen to the conversation

0:000:00
Original episode description

Most AI landing pages look like slop. Here's how to build ones that actually convert. This week, Nik goes all in on landing pages with a presentation from the recent Ecom Founders Q4 Summit. He's been building pages with AI for a while now, and he's turned his whole process into a skill you can copy and use for free. Nik also shares example pages he built for Caraway and Love Wellness, with no...

Transcript

EN

Welcome back to Limited Supply, the podcast where we get deep into the tactic...

side of e-commerce, digital marketing, and building consumer brands.

“I'm your host Nick Sharma, I've spent the last nine years building, scaling, and investing”

in brands, and through this show in my weekly newsletter at nip.co/email, I'm here to share everything I've learned. The wins, the losses, the experiments, the tactics, and the insights, also you can unlock your next $100,000 in revenue. Today's episode is a good one, but before we dive in, let me tell you about our chosen

sponsor for this week's episode. If you're using AI to plan your ad-spend triple whales moby, can make the budget changes for you in your connected ad accounts with your approval. Brands like Kitch use Triple Whale, book a demo at triplewhale.com/limitedsupply. That's T-R-I-P-L-E-W-H-A-L-E.com/limitedsupply.

Welcome back to another episode of Limited Supply.

I'm your host Nick Sharma, and today we are -- we're basically going all in on landing pages.

And I've talked a lot about landing pages. I've talked a lot about AI. I've talked a little bit about my process of building landing pages with AI. And today, well, every day with AI, right, innovation happens. And one of those things is that you don't have to keep doing everything manually.

“You can bundle things into skills which I've talked about before, but what I've been able”

to do now is basically take a skill for how I build landing pages and turn it into a full skill, right? Make it look like it's not AI. Make it something that is very commerce, consumer conversion friendly, and then give it to you.

So, at the end of this episode, you're going to see the link on there. You can see page examples I've made for Carrowase Coffee Maker, for some love wellness stuff. You know, these are just different pages I popped up because I wanted to see for a brand. I have no context on.

I'm not working with. I don't have any other files, any other memory. Can I still build really good pages? And the answer is yes. So, take a listen to this episode.

We talked deeply about landing pages and then we go right into the skill and how to leverage it. Enjoy the episode. And let me know if you have any questions just DM me on Twitter at Mr. Sharma. How you guys feeling?

We're almost at the end here. Thanks for bearing with us.

I'm going to show you guys how to build, how I've basically been building landing pages

with AI. It's one of the things I feel like I figured out that a lot of others haven't yet. So my goal is going to be show you that AI is artificial intelligence, not just anonymous Indians. And then second, teach you guys how to build landing pages because I'm the one Indian who

doesn't know how to code. So this has been super helpful for me. So I'm going to take you guys through a bunch of slides. What I'm going to do also is I'll just send anybody the deck if they want it. So maybe what we'll do is after the conference, when we send an email out, we'll include

the deck. Click a picture of everything because there's going to be a lot. It's like 45 slides and I got like 20, 20 something minutes here.

“So quick background, why am I talking to you about landing pages?”

Well, I've built a lot of funnels and a lot of landing pages for a lot of brands. Most of these brands are focused on acquiring customers through meta, mainly cold prospecting traffic. And that's where I feel I kind of do my best work as cold prospecting traffic. So you know, a couple of years ago, I had this agency called Hooks and the idea was that

we could help brands figure out new hooks and angles and wedges into finding new pockets of customer audiences. And the way we would do that is we'd have this pretty, you know, four to six week process. It's what it ended up being I thought we could do it in two weeks, but then you've got

all the back and forth and the revisions and all this button looks too rounded and our buttons are rectangle and this section should go there and we should put our founder story up at the top and whatever it is. But I started to see that chat, GPT was getting a little bit, you know, smarter and smarter and this is again two years ago.

So just basic chat is all that existed. Now I saw this come back because what I was thinking was, all right, AI, you're going to trade it, train it like an untrained puppy. So you got to teach it your 10,000 hours and then it can start to, you know, learn and replicate from that.

And the first output I got was something like this, which was essentially a wireframe that was built using, you know, plain text characters inside chat GPT. And then I said, oh, shit, I got to shut down hooks, this is not going to last. This is going to take over. So you know, a page like this, which I'll just scroll through a quick, you know, there's

a full, beautiful page, you got animation, you've got, you know, basically everything you could need. A full page like this used to take three to six weeks to make. Your first week, you buy the page, then you set up a kickoff call, then the agency does some level of research and strategy, which is usually kind of horseshit.

And then they do the wireframe, the copy, the UI, then they dev it. Oh, but then the client wants changes after dev. So, you know, you end up at a six to eight week process.

The problem is by the time you actually get the page to go and run, the whole thing is irrelevant.

Right. The angle you wanted to test, the persona you were going after is now no longer what you're

Looking to go after.

But if you know me, you've always heard me say that landing pages are where you kind of

treat your customers like their Kim Kardashian walk in the red carpet, right? You're the assistant to the customer who's Kim on the red carpet. And it's your job to answer everything she possibly needs to know as she's walking, you know, without her prompting you. And that's really what I think of a landing page as you want to figure out how to basically

“onboard people to your brand in the best way.”

The best messaging to bring them in, the best product assortment, the best variants, you know, if you sell flavors or scented products, you know how important that is to nail on the first purchase. Now, you've probably also seen a lot of slop landing pages, a lot of slop creative. This is stuff that's basically untrained AI models where you're forcing it to say, hey,

take my product to URL and turn it into a landing page. And you get something like this, right? We just had Lauren up on stage. If you checked out her site or if you know the brand love wellness, you know, this is definitely not her branding and not how she would, you know, push the brand out herself.

So I got to think, all right, how do you work on, how do you basically do this in the way that is right?

And this is the workflow I've always used, right?

So I love to use meta because meta gives you the shortest, quickest feedback loop to understand what works. You can immediately look at CPM, click through rate, you know, cost-perlanded view, cost-pad the cart and really just know, are you hitting the right people, are they liking what you're

“saying, are they, you know, do they like the product you're selling and how you're selling it?”

Then once you figure out what actually works, you go ahead and build behind it. So, you know, if you can launch 50 to 100 pieces of creative a week and realize that maybe 12 of them really have legs from a messaging or an angle perspective, now you can go ahead and build these pages behind it. So as a ton of pages, you can build, right?

There's these evergreen kind of customer acquisition pages, you've got a listicle, you've got your bestsellers bundle, you've got the founder letter, you've got the Google ads, specific LP, which might be more of a collection's page with some education, you've got a social proof page, then you've got all your Q4 promotional pages, right? So you've got your Black Friday pages, maybe it's a preview of your holiday sales, you've

got a cyber Monday collections pages, essentially you're trying to figure out how to make the traffic more efficient, how do you minimize the number of clicks as somebody comes in? Gift guides, and then you've got brands who are using landing pages for purely for performance marketing. And what they're doing is pretty much what I just described, which is use Facebook, figure

out angles that work, as soon as they work, go ahead and build a full funnel behind them. If they work really well, going get whitelisted creators and just keep adding to that, right? Once you find an angle or a persona, you just start throwing more eggs into that basket. And it doesn't mean you stop testing other angles and personas, you're still doing that, but whatever is working, just keep feeding it until it stops working.

So the reason why we do this is because the journey to check out is at the top, and the things I didn't include is when people leave your site, they go Google search something, they come back, they leave again, they get an email as a follow up, because you use some card event or you know, site abandonment tool, they leave again, they go back on TikTok, they see your ad, but really the goal is inferior to just consolidate this whole user journey and figure

out how to make it more efficient.

“So now I think about, you know, now we're talking about building these pages, right?”

So one of the things AI misses is this five question rule that I have. Question number one, what are you selling? Explain it to me, like a five year old, who doesn't know anything, right?

I always think, how do I explain this to my mom, so my mom's not going to take the

time to go research a bunch of shit, she wants to know exactly what it is and how it's going to benefit her life, which is a second question, why should I care, how does this benefit me as a consumer coming to your page? One of the reasons the angle specific page is work so well is because number two is very dependent on the angle of traffic that came to the page.

Number three is really focused around shipping is a big one in Q4, how fast do I get it, right? If I order today, does it ship today, does it come tomorrow, is there two day shipping, does that mean it ships tomorrow or ships Monday, how fast do I get it is a really big lever to increase conversion on the page? The fourth one, how will this make my life better?

What am I doing now that could be improved? How is this going to improve daily behaviors I already have? And then the last one is, how does this compare to other things on the market? Slash, the other one I ask is, you know, if I'm going to order this, how do I not look stupid in front of my girlfriend, boyfriend, wife, husband?

And I always think you want to answer that for the person before they buy because that's always something in their head. So all right, now I'm going to go through like 30 modules of pages that I highly recommend. This is all based on, you know, pages I've run. So we've got up front the promotion bar and the header and nav, you know, I think the header

and nav and the promotion bar, that stack is I call that ocean front real estate on your site, right? It's above the fold.

People land there.

They see it every time. People have really big navigations, some people have really efficient ones. I'm more fan of being efficient, but for some brands depending on your catalog size, you might want bigger. You've got your hero and your benefit strip, you know, again, just reiterating.

You've got your product gallery. You'll notice here the product gallery heavily focuses on using that real estate, right? They've got iconography and text overlays throughout the gallery and then even the how it works.

So kind of making it very obvious when you get this, here's how you open it, here's what

comes in the box, here's how it works. You've got the how to use, you've got the ingredients, you've got the problem solution. The reasons list, I'm always a big fan of adding a five reasons why, even if it's not a big listicle, adding a five reasons why that scroll sideways is huge. Comparers in table, if you don't already run the comparison table, you don't have to go

direct brand to brand. You can go brand to something else. So for example, you know, I worked at Hintwater. We wouldn't compete against other water companies. We would say, here's Hintwater against diet soda against vitamin water against, you know,

basically another sugar bomb of a drink. Demonstration.

“If your product is demonstrable visually, this is the best thing you can do, show that show”

how the product works and really just use it on camera, then you've got, of course, you're testing certifications and experts. So experts, I consider like thought leaders, but more industry thought leaders. And then the testing and certifications goes a long way because, you know, it's been brought up here actually a few times, but there's a lot of dropshippers that see what works and

then they try to pile on and take advantage of your education. The more that you can separate yourself with education, certification, experts, any sort of testing results, clinical results, clinical trials, you know, the more you stand out and you really become more protected. Press in authority.

Social proof is I think the biggest part of this and, you know, these are both examples of social proof, then answering any remaining doubts.

So I always think, okay, if somebody's on the fence, what are they contemplating before

they go ahead and make a purchase? And this is kind of what you can answer through the module. So in this case, we've got it through reviews and UGC. Somebody talking about how they overcame a thought in their review or in their UGC. And of course, delivery returns guarantees FAQ, footer, you know, talking about the offer,

you've got the purchase module, the inclusions, I'm going to send this all to you guys so you haven't. Okay, so we've got all the modules now, right? And these are a lot of these modules are ones we didn't see on that AI slot planning page. So right away, just adding the right modules and takes your page to the next level.

So if we were to think, you know, untrained dog as AI, I got to put my 10,000 hours in or my process, maybe even my modules or ideas, a big part of this is actually doing the research on your own to try to figure out what are the angles you want to go for. So reading Reddit, reading TikTok comments, reading your ad comments, reading customer service complaints, anything within other forums or communities, that's all prime valuable

alpha that you can use to decide what you build in your page. Now if I were to go the long way of building a page, this is my process. Number one, start with the product person should be say, persona, product persona and problem. So what is this person trying to solve, right?

And if we look at the new beautiful caraway coffee maker, well, their biggest angle is really going plastic free and, you know, cleaner coffee. Then there's evidence. So how do you show somebody that you have the evidence to say what you're saying?

The third one is think about the past alternatives.

So this isn't on you as a brand.

“In fact, this has really nothing to do with you, but you have to be empathetic to your”

customer and understand what have they tried before. What have they tried to do to solve this same problem before that you are now introducing in a different way. Because they've already tried to solve it, they have perceptions in their mind about what you are selling.

So how do you get around that in an effective way? Your angle or your punchline. So what is the angle or your wedge into the door and how are you going to stand out from a competitor? And then the last one is the page sections, all those sections I just showed you.

Now if that's too complicated, which a lot of times it is, you know, sometimes you just need to rip something fast, then you just go through your reviews and you make a tally of what are people talking about the most, right? So I went through some of their early reviews, they were talking about the brew path materials, the fact that it's glass and not plastic, the coffee quality, the simple morning is the

easy clinic cleanup, et cetera. And I basically just put a tally next to each thing mentioned. And by the end, I've got a list of most mentioned things, to least mention things. I take the top five and I turn it into a listicle, right?

“Now, if you want to get even easier, right?”

If that's still too hard, it takes too much time. Really open your ad account, figure out what, where are you spending the most amount of money and where is it the most inefficient, and say, that's where I'm going to start. Just build a challenger page to whatever you're running traffic to now and start there, right?

If you're spending a lot of money, even a 5, 10% increase makes you a lot more money and adds

A lot to the bottom line.

A landing page is supposed to be your brand's seal team six, supposed to go out there,

get the customers in, basically grab them, bring them in and push them in to check out.

Now my toolkit for building these AI landing pages is actually very simple. I don't use too many of the fancy kind of AI SaaS tools. I pretty much use codecs to build all the pages. I use GitHub to host all my files. I use Versel to display and publish those files. I use HQ as my own internal memory system

across my team, which has everything from the do's and don'ts, you know, what kind of messaging and language do you say organic or natural or all natural and what's allowed, what's not allowed? I use a mic that pretty much sits up right here all day because I prefer to voice my inputs into codecs versus type them, right?

If you type them, you're only actually getting 10% of what you want to say, because you're a price slow typeer. And then I use Grockbot to do ad creative, which is shocking because I didn't expect Grockbot to do ad creative, but use Grockbot with Higgs field as a connector, you get really phenomenal ad creative as an output.

You can spend an hour working through your ad budget with AI and still have to make every change yourself.

“Triple wells Moby uses your business data to recommend exactly what changes you should”

make and then can go and make those changes for you with your approval. You can plan before any money moves. Book a demo at triplewail.com/limitedsupply that's TRI, PLE, WHALE.com/limitedsupply. So now I wanted to do this as a live example with love wellness because we had Lauren here and you guys were all here, but what I realized is the internet is not the fastest

and you do need fast internet with AI. So I sped it up a little bit, but I still want to show you the final result.

The whole process probably took me 45 minutes in total to make the first page.

The other two pages I'll show you took about 15 minutes because I had all the learnings done. I'd already done the research. I knew what angles I was going after and now I was basically repurposing the information I gathered.

So I'm not a consumer of love wellness myself, but I wanted to go out and see what are real use cases that people have and that we can leverage in our own kind of formulation of the page. So I found this Reddit thread about buy-by-blades, one of their most popular products. And somebody was talking about essentially how she can now fit back into her jeans again,

which I thought that's an interesting angle because if you lead with fit back in your jeans versus D-blades, there's a lot more people that are down to fit back in their jeans than buy a D-blades supplement.

And basically across through this process ended up making a few different revisions as

you do.

“That's not going to be perfect, right? That's why I use the microphone to just stream”

my feedback and make those changes. And then right before you go live, I do what's called the rehearsal dinner. Now the rehearsal dinner is, you know, for all my white friends in the room. This is something that people do right before you're wedding. I didn't know this because I'm not white.

But see, make sure you got your pixels, right? Google, Meta, TikTok, Snap. If there's anything else you run, you include that. You make sure you've got your measurement partners in there because you want to understand the incrementality of the page.

You make sure you've got your behavior partner and there I recommend Microsoft clarity. It's a free heat map, a scroll depth, user session recording tool. And they also have an API. So if you're using codecs, right, you can then in two weeks say, go look at that traffic and make another version based on everything that happened on that version of the page.

And that's pretty much my tech stack setup. And in addition to the tools I mentioned earlier. And then you just pair it with static ads. That's like the easiest way in my opinion to get landing pages up. If you don't know, if you're not going after the third option, which was again, find the

weakest performing creative and just make a new page for that, then you can add additional static on top. And again, once you have version one of the page with your brand, everything else is much easier to build. So I'm going to just quickly show you.

Or actually before I get to that. So I thought, OK, if anybody goes home and tries to do this, it's probably going to look

“like shit because it's going to look like slop, right?”

My stuff has a bunch of learnings built into it. So how can you guys get that? Well, I'm going to give you a link in a second, where you can just scan the QR code. And it's just my landing page skill for free. So you hit the copy button, you paste it in your own cloud code or codecs.

And for the rest of your life, you can make landing pages like this. That's really easy, like I said, copy, throw your link in and it's good to go. This is where you can get that link, highly recommend stealing this. Don't sell it, please.

The password on the page is Q4, let me get out of the way, password is just Q4.

You put your URL, you say use next landing page skill, and 15, 20 minutes later, you've

got a landing page ready to go. So we saw the slop, right? I'm going to show you what the final one looked like. This was what the final one looked like.

“You have pretty much everything you can think of wanting in a supplement brand page, right?”

It's even mobile friendly. It went ahead and said, oh, people know that they're going to zoom in on the thing. So on mobile, make it easy for people to zoom in on the nutritional panel. You've got your social proof. You've got your benefits.

You've got your UGC. You can create text overlay images that come in. And now that I've got this page made, I wanted to make a five reasons why page instantly made. Ready to go.

You can add different elements. I've got this side scrolling CTA that sticks with the user as they go down. If they're on mobile, this is doing the same thing. And then I thought, let me just make some add tutorials. Now that I've got love wellness as whole brand system, research, wedges, kits, everything,

right? And so now you've got add tutorials that you used to have to pay publishers to get up and running. You can do these for free. The only verdict is just a domain I bought a while ago, but you can do this on any domain.

You just go to go daddy by the domain, tell code X you own it, and it'll connect everything for you. So that's pretty much all I got. Anybody got any questions? Oh, we got a mic right here.

Hey, thank you so much. So the skill that we have to build is single landing page.

“Would you also apply the skill to, let's say, a brand's home page?”

Yeah. You know, my biggest thing is your outputs are only as good as your inputs. So this skill will do a much better job building you a landing page.

I would use it as your kind of first step to really like build your brand kit.

That was actually one thing I mentioned under my tools was brand kit. And that's usually where people go wrong in the sense that, you know, when they can't figure out why their ad creative doesn't look good and it looks like slop or it's not getting the product imagery or your messaging is wrong. It's because the brand kit hasn't been established.

And what that is, is basically your tone of voice, your brand style guide, your assets, you know, kind of all the things visually or or messaging wise or brand wise that an agency or a creative partner would need to know in order to execute work. If you build that intentionally, meaning you spend some time, you say, cloud code, go through all my newsletters in the last two years, create me an email messaging or tone

of voice style guide, boom, that's one file. Do the same thing with paid social, do the same thing with customer service, right? Now you're creating these foundational files that essentially live inside your brand kit. And you know, you do a little bit of that and you can absolutely then use that landing page

skill to probably build you a version of a home page. Okay. Yeah. Thanks. Next question.

Are you using GitHub to host all the imagery used in these landing page or what's the process there, like we use smug mug as like an internal dam, dropbox, do we damn like what's the process on that?

“You're saying, do I use GitHub to host the page or do I host something else?”

Like the imagery, the assets you would use the bills that page. Oh, a jury videos. I want to say, yeah, I want to say it's all hosted in GitHub. I couldn't tell you to be honest.

I basically tell codecs, make sure this page doesn't go down once it goes up.

I believe it's all pushed into a GitHub folder and then verse out launches it and displays it publicly. I'm pretty sure. Again, I'm not the tech guy. I know I'm Indian.

Question. How do you make the decision to go landing page versus general home page for your brand? So I would say, so the questions, how do you make the decision to build a landing page versus just essentially rebuild your home page or optimize your home page? So the way I think about it is landing pages are where you want to focus specific subset

of traffic, right? Your home page kind of speaks one way about your brand to everybody. Your landing page can speak 50 different ways to 50 different audiences. So it's really a function of scaling spend or, you know, in Q4, a lot of people do like promo specific pages, you know, giving Tuesday page, small business Saturday page,

back Friday page, cyber Monday collections page. But, you know, if you don't have that much spend, you're mostly sending traffic to the home page, definitely focus on your home page as your landing page. As you scale spend, you're probably going to want to figure out like, all right, moms actually react better to this messaging.

So let's focus on this. We could create bundles for moms and their, you know, their families. This audience is going to want something completely different.

You know, they're maybe hosting events.

So they want popcorn for something different, right up here in the front.

Thank you. That's, that's pretty good. piggybacking on the answer to the question now, so by that gentleman, where you said that your website has got all the product information, the reviews, the PDPs, everything about the product, all of that user comments, et cetera, does it make sense to build the websites

on kind of like an RIG or a kind of like it's own LLM because it's got more information than the PDPs. Sorry. I'm missing the question.

“So does it make sense to build an LLM inside your own website?”

Oh, like a chatbot? Like a chatbot. On your site. Yes. Yeah.

I mean, look, it's going to happen one way or another, right? If you're on Shopify, I bet by January, there's going to be a shop, shop, little purple widget that you'll be able to enable, and the whole promise is going to be, we know, you're, you know, we know everything about your store and what you sell and all that. So if you want a chatbot, you can use ours.

You know, meta. I'm sure you saw Mews just come out recently. Their entire business model is affiliate commission on the back. So they're entirely incentivized to try to get on everybody's site as well. It's going to happen.

When I launched festivals with Mr. Beast, we had a custom app that we built for all the chat on the front end, and people use it like crazy. It's an amazing tool. So I've no doubt it's going to happen, the, like me personally as somebody who has gotten super deep in the world of AI, my question would be, okay, who owns that data, right?

When you're having a conversation with somebody random, who owns that conversational data, are you paying a commission, I'm an Indian again, so I'm going to try to figure out how to not pay a commission and go direct or build my own tool. But I would highly recommend it, and, yeah, again, the thing about AI is the more data you can collect and siphon for yourself, right, not give the other platforms.

The more it's going to benefit you, like on a compounding basis. So the more you could do stuff like that, but then keep all the data, right, the better. Thank you. So Mews is relatively new. So what do you say is, how brands can use that to grow their business?

I mean, you use this more for personal use, what I understand right now. Yeah, I mean, I think from a marketing standpoint, Mews is trying to go after the personal angle, because instinct is probably their biggest or a competitor right now, which is a big

“personal assistant angle, and that's how they're getting, you know, the normies on AI, basically.”

You can still, it's still a very powerful tool, so, and it has connector abilities, right?

So you can still connect it to everything, like everybody in here should have a system that's plugged into their Clavio API, their Postcript API, their Shopify API, you know, basically every ads platform, every reporting platform, like these platforms all have APIs, whether you know it or not. So you can plug all of it in, and then, I mean, if you whether use Mews or use Codex or

Cloud Code or something else, you know, it really comes down to your preference of the model, but also how these systems organize files, that's why I recommended HQ earlier, because HQ allows you to have a universal memory system across any model or any of these harnesses you use, right? Mews, Codex, Cloud Code, they're all considered harnesses.

It's kind of like, it's just the window you type in, and then there's a model that comes in, whether it's, you know, Fabel or Claude or Astra, and then there's the memory it pulls from. So in my setup, I have the memory, as HQ, no matter what window I use to chat, and, and because of that, I can go between each of them without feeling like there's any loss

of information, or that, you know, if I use Mews today and Codex tonight and Cloud Code tomorrow, they're all going to be fully in sync, because I have my own memory system built out, which I highly recommend.

“That's probably the best thing you can do for your brand, as you kind of build this AI infrastructure,”

is make sure that you've got all your stuff in one place organized so that everybody's kind of working off the same set of updated information. It's not that one person's cloud has been advanced for the last six months, they've been

doing a bunch of stuff, and now the new guy has to basically start from scratch, right?

Then the outputs of what you do are completely different. The other thing is, if you actually have this brand kit, right, every partner you work with over the next few years is going to need that brand kit, because all your partners are going to be using AI in some way, shape, or form. And if you don't give them a skill file or a, you know, a gentic friendly brand kit to use

and plug into their process, well, when they're running your creators, or they're building you static ads, or they're building you an advertorial, or, you know, whatever they may be doing, their stuff is going to look ridiculous compared to yours, because they don't have those same kind of memory and skill files. Does that make sense?

Yeah. All right. We got time for one more question. Any last question?

Over here I see one.

Sorry.

This is a bit of a, how long is a piece of rope question, but what is a good benchmark for

“moving over to a landing page from a collection page in terms of conversion rate, jump up?”

You're saying to go, uh, so, so, if you were to add a landing page, as like, a click through page, what's a good, like a typical collection page, what do you see is it like, you go from like a 2% conversion rate to a 12% or what's, you know, what's a industry benchmark for your, yeah, I mean, I think it depends.

So like, if you're running a straight, like, I could tell you, I remember for a hair care

brand, we were running site directly to the collections page, and then we tested the same bundle that we were trying to get from the collection page on a landing page. We saw a 10% conversion on the landing page, just because everything was focused there. There's no link clicks out, you know, it's a very focused experience on the collections page.

“I think we were like, probably two to three and a half percent on average.”

So, I don't know if that's helpful, but, you know, the other thing to basically look

at is just like, all right, is your cost pad to cart going down, is your cost per initiate check out going down, are you seeing that there's less frustrated customers going through or people feel more educated as they go through, you know, another one that I like to do is these, like, in between, like, the listicle pages, for example, right, you're basically just placing something in between the ad and where you're trying to get

them to. And, you know, for me, I'm like, if I can hit a 35 to 50% click through rate on that page, I'm in the money. And it also shows that, you know, the audience, the ad, the messaging, and the landing page all have some good consistency there.

But yeah, I would say, you know, it also depends, because like it depends on what you're selling. Super expensive, you might have a 5% increase in your click through rate or in, you know, add the cards, but if you saw something expensive, that might, might make you an extra few million bucks.

So yeah, hope that helps. All right, well, I hope you guys use the skill. I was one shot in some pages last night that works well. If you have any questions, you can shoot me an email, [email protected], or find me somewhere here.

But thank you for the time. Thanks for listening. We'll be back next time to cut through the noise on CPG, retail, and e-commerce.

“If you enjoyed this episode, why not share it with a friend?”

And be sure to subscribe wherever you listen, so you don't miss the next one.

Keep reading