The Right Way to A/B Test Landing Pages on Meta Ads for Ecommerce

Summary

Podcast Summary: Marketing Operators

Episode: The Right Way to A/B Test Landing Pages on Meta Ads for Ecommerce
Date: June 30, 2026
Hosts: Cody Plofker (Jones Road Beauty), Mark (Guest host, subbing in for Connor MacDonald/Connor Rolain)

---

Overview

This episode dives deep into the strategies and philosophies behind A/B testing landing pages for Meta (Facebook/Instagram) ads within ecommerce. Sparked by a lively discussion on X (formerly Twitter) and notable threads from industry leaders, hosts Cody and Mark explore real-world tactics, operational challenges, and nuanced insights from managing massive ecommerce ad accounts. They also traverse tangents into AI, product launch strategies, and the shifting landscape of DTC growth.

---

Key Discussion Points & Insights

1. The Current State of CRO & Landing Page Testing

[04:40–13:03]

• Testing Philosophies:

  • There are myriad ways to A/B test landing pages—no consensus “right way.”
  • Businesses often default to what works for their own operational realities, not to a universal best practice.

• Operationalizing Testing:

  • Jones Road uses around six template landing pages, chosen via iterative A/B tests (e.g., listicle vs. performance LP) depending on the product.
  • A key insight: “We overdeveloped pages and under-optimized them.” (Cody, [09:41])
  • Diminishing returns emerge from creating bespoke landing pages for every ad; leverage comes from smart, focused testing.

• Homepage as a Control:

  • Frequently, the main site homepage outperforms bespoke landing pages—a shift from years past.
  • “There’s so much more leverage in that. Like, if we’re gonna run traffic to that and it’s already beating it, let’s get it as good as possible.” (Cody, [11:15])

Notable Quote:

“Everybody seems to do it a little bit different. There’s a thousand ways to skin a cat, and that’s true with landing page testing as well.”
—Mark, [04:44]

---

2. Tradeoffs: Structured A/B Testing vs. Launching New Funnels

[13:03–19:48]

• Structured Tests:

  • True A/B tests provide clarity but are slow and resource-intensive.
  • Relying too heavily on split-testing can lead to operational bottlenecks.

• The "Funnel Congruency" Question:

  • Creative, messaging, and offers must match between ads and landing pages for tests to be valid.
  • Mark shares an anecdote from Gut Culture: when a radically new offer ($88 vs. $39 product) was introduced as a strict A/B test, it tanked performance due to existing ad learnings and price-segmented audiences within Meta.
    • Takeaway: Meta’s ad algorithm serves different audiences based on historical signals, including price and landing page design.

• When New Funnels Beat Tests:

  • Instead of A/B testing, sometimes launching a new ad set with dedicated creative and funnel structure yields better incremental results.
  • “Meta is using the details from the landing page and the PDP that you’re driving to in order to influence who it’s serving ads to.” (Mark, [17:17])

Notable Quote:

“That’s not even really a test. You can almost think of it as just being incremental to what you’re currently doing… launching them as separate funnels.”
—Mark, [15:50]

---

3. Landing Page Diversity & Meta Algorithm

[19:48–22:55]

• Diversity as a Signal:

  • More varied landing pages can help increase reach and potentially escape audience saturation in Meta’s optimization system.
  • Some brands deliberately deploy an array of landing page templates (e.g., bottom-funnel vs. top-funnel), mapped to ad types.

• Risks of Over-Diversity:

  • The operational cost of maintaining many pages is high (broken links, asset management, update complexity).
  • Efficiency and clarity often come from consolidating to fewer, high-performing templates.

Notable Quote:

“It just feels like… a burden over time having more and more landing pages live… it becomes more difficult to make decisions around what’s working, what’s not.”
—Mark, [28:00]

---

4. Practical Testing Frameworks & Real-Life Examples

[22:55–34:38]

• Testing Major Product vs. Minor Products:

  • For top SKUs (“programs”), it may be worth running extensive template and funnel diversity experiments.
  • For smaller products, focus on a single, well-performing page or even just the PDP.

• Optimization Tactics:

  • Use A/B testing for big swings and directionally important decisions (e.g., homepage vs. listicle).
  • Prioritize operational simplicity; avoid maintaining excess landing pages for campaigns with limited upside.

Notable Quote:

“If we do test it, we’ll test a listicle against a homepage, and I almost hope the homepage wins because then I don’t have to go and maintain this other one.”
—Cody, [27:00]

---

5. Nuance: Upper-Funnel Metrics & Velocity vs. Precision

[34:38–38:46]

• Testing Small Segments:

  • If testing on lower-traffic portions of the account, consider optimizing for upper-funnel metrics (e.g., PDP views) to get directional learnings faster.
  • The speed of testing (high-velocity, low-precision) often outweighs the benefits of waiting for statistical significance.

• “Canary in the Coal Mine” Tests:

  • Sometimes tests are simply sanity checks—run a quick split to validate nothing is catastrophically broken, then ship.

Notable Quotes:

“No one’s taking me up on it, but I really think the right way to run an incrementality program is really high velocity of low precision tests.”
—Cody relaying Zach from Haus, [36:42]

“You just want to get this thing live with some level of confidence that it’s not like a terrible net negative… I’m all for tests like that.”
—Mark, [37:26]

---

6. Large-Scale Theme/Product Redesigns: How to Test

[39:06–46:47]

• Testing Full Redesigns:

  • When testing a major site overhaul or new theme, balance change velocity with variable isolation.
  • Best practice: test “like for like” where possible—make only necessary changes to the test variant so performance differences are attributable.
  • Use multivariate tests where traffic allows—add variants with increasing complexity to track which features move the needle.

• Rolling Roadmap:

  • After rollout, layer in new features and enhancements incrementally, rather than launching too many changes at once.

Notable Quote:

“If it works, you don’t care why… if it doesn’t do well, I think that’s the harder one. My opinion: I would change as little as possible.”
—Cody, [41:01]

---

7. Product Launches as a Growth Engine

[49:26–63:54]

• DTC 3.0 and the Evolution to Multi-SKU:

  • Brands that once thrived as single-SKU, high-LTV businesses (AG1, create, etc.) will increasingly shift to multi-SKU to maintain growth in crowded markets.
  • Rapid SKU proliferation can exhaust audience goodwill and diminish per-launch impact over time.

• Category Expansion Challenges:

  • In beauty, it’s difficult to draw new audiences with adjacent products; “new” must mean truly new to net new customers—not just extensions for current fans.
  • Cannibalization risk is high for similar offers; more radical innovation is risky yet potentially more rewarding if it captures a genuinely new market segment.

• Positioning Dilemmas:

  • Adding products can undermine original brand promises (AG1 as the "all-in-one" now launching supports like creatine).
  • Maintaining trust and positioning while expanding the catalog is a delicate balancing act.

Notable Quotes:

“The consumer is tired of it... if your launches are outpacing the rate at which your audience is growing, then you’re just kind of gonna drain it.”
—Mark, [55:57]

“It’s harder than I thought to have a new product or new category that really appeals to a new audience. You kind of have your pool...”
—Cody, [53:34]

---

Memorable Moments & Quotes

• On “Funnel Diversity” and Algorithmic Learning

  “Meta’s using the details from the landing page and the PDP… in order to influence who it’s serving ads to.”
  —Mark, [17:17]

• On the Operational Realities of Testing

  “There is an operational cost of all of that… I almost hope the homepage wins because then I don’t have to go and maintain this other one.”
  —Cody, [27:00]

• On Fast, Directional Tests

  “I don’t want to wait three weeks to test everything. I’d rather just get a new PDP live and obviously get some metrics on it… with the speed that… even if you make one wrong call… you’ll have enough winners just by the speed…”
  —Cody, [36:42]

---

Timestamps of Important Segments

• [04:40] — Kickoff: Approaches to landing page iteration/testing
• [09:41] — “We overdeveloped pages and under-optimized them.” — Cody
• [13:03] — When to use A/B tests vs. launching incremental new funnels
• [15:50] — The case for landing page diversity as not an A/B test, but additive
• [17:17] — How Meta uses landing page signals to optimize ad delivery
• [19:48] — Balancing diversity and operational complexity in landing pages
• [22:55] — Real-life example: BAU vs. product-specific landing page testing
• [27:00] — The burden of excessive landing pages
• [34:38] — Velocity vs. precision in A/B testing, upper-funnel metric use
• [39:06] — How to test a full site redesign or theme overhaul
• [41:01] — Cody on minimizing changes for better testing clarity
• [49:26] — Product launches as a flywheel (or pitfall) for DTC growth
• [53:34] — Hitting the limits of new-product growth: “It’s harder than I thought…”
• [55:57] — Mark on launch fatigue and the law of diminishing returns
• [58:38] — Can categories/brands “cross the chasm”? (AG1 example)
• [61:49] — Dynamics of single-brand trust vs. multi-brand supplement purchases
• [62:29] — Brand-building around product launches (Gruens, etc.)

---

Tone & Language

Conversational, candid, and tactical with anecdotal war stories. The hosts offer nuanced views based on direct experience, providing a no-BS lens for operators on the front lines of ecommerce growth.

---

Takeaways for Ecommerce Operators

• There is no single “right way” to A/B test landing pages on Meta—match your approach to your business’s constraints, traffic, and product mix.
• Consider landing page diversity as a tool for algorithmic “reach”—but be mindful of operational drag.
• Don’t hesitate to ship tests before perfect data; high-velocity, directionally sound experimentation pays more long-term.
• Major site or product changes should be isolated or stacked methodically; after big lifts, roll in enhancements incrementally.
• Expect the DTC terrain to keep shifting—brand and product expansion strategies must adapt as markets saturate and fatigue sets in.

---

For Further Listening

• Dylan Ander on Homepage Performance ([11:15])
• Brad from Scalability School on New Funnels ([13:03])
• **Jackson’s GM on Upper Funnel A/B Testing ([33:15])
• Open Residency & Gruens for Advanced Product Expansion Blueprints ([62:29])

---

Summary prepared by Podcast GPT.
For more tactical breakdowns from the Marketing Operators team, revisit their previous episodes tackling creative testing, measurement, and growth operations.

Transcript

[Mark]

All right, we are back with another episode of Marketing Operators. You've got just myself and Cody today. Cody, how are things on your end? We're getting close to the end of Q2. How are you feeling?

[Cody]

Can't believe we're halfway through the year. Can't believe we're halfway through the year. It's been a. It's been a year. Definitely got a little bit more gray hair than I had at the beginning of the year. Hard, hard year.

[Mark]

Yeah, I hear you. I hear I got something. This showed up last night. I feel like you'll be excited about. Check this out, dude.

[Cody]

Nice. I love it. I love it. What are your plans? What are you doing?

[Mark]

My plan is I'll set up the Mac Mini with Claude and then I'll just be able to, like, sign off forever. Like, that's my understanding is like, I could just go away from keyboard for until the end of time.

[Cody]

Yeah, pretty much. But you gotta do HQ on that. You gotta make sure you got HQ set up.

[Mark]

Um, no, it's true. There's. There's a lot of cool AI stuff going on, and I just felt like I'm a couple months behind on the Mac Mini hype. I'm a couple weeks behind you. I think maybe a month behind you. So I'm getting on the bandwagon.

[Cody]

All right, well, I won't do it on the pod because I think people probably are sick of hearing about AI, but happy to share notes individually if you want on what we're doing because, yeah, I got one. And then I'll say, we just got one for jrb and so we have, like, an agent that we're calling Alfred from the Butler and Dark Knight Rises in Batman. I thought Alfred was, like, a good name for one. He seems like a really helpful guy. So we're setting that up and. And having that be in our slack. But it's been cool. So excited for you.

[Mark]

So you name. You name the agent. Do you name the Mac Mini itself? I almost feel like naming Mac Minis. Naming hardware could go the route of, like, boats, you know, like, we might need. They're like. They're important vessels.

[Cody]

The thing is, you can have multiple agents running on one device. Like, we'll probably. We may even have, you know, like, we're using like. Like. Everyone calls it Hermes, but it's. I call it Hermes because that's how you spell it. So we're using. We're using, you know, Hermes as the agent, but you could have multiple Hermes on one Mini. So we're we're naming our agents, but, yeah, maybe. Maybe we'll name the vessel as well.

[Mark]

Yeah, I'm on board for it.

[Cody]

What would you name yours? What would you name yours? Very important.

[Mark]

You know, I. I'd have to think of some sort of cute wordplay. I don't have anything off the top of my head here. I'll come back with some additional notes on a future episode, but I like the idea of it.

[Cody]

What about a boat? You sell Ridge got all this money, you're now Zach stuck, you get a boat. What do you name your boat? Have you thought about it?

[Mark]

Oh, dude, no. You're putting me on the spot here. I'm, like, terrible at coming up with things off the. Off the cuff here. No, I don't have an answer for you. Again. There's just, you know, you walk around a harbor or something, and there are just some really clever names. I would take it really seriously. That's what I mean. I just feel like dedicated AI Hardware might be going that same route. All right, I got one more thing for you before we jump into it. The Social Network, too, I think it's called the Social Reckoning trailer came out.

[Cody]

Oh, I didn't know that.

[Mark]

Yeah. So this is exciting. Jeremy Strong looks fantastic as Mark Zuckerberg. Like, he's gonna be. He's just gonna be incredible. But I was disappointed, though. From the trailer, there was no mention of removing credit cards or location fees. Can you believe that?

[Cody]

Did they mention switching from DMAs to zip codes or whatever they changed for in the trailer?

[Mark]

I assume it'll be in the movie.

[Cody]

Yeah, I wonder. I actually wonder, like, on a serious note, if they. If they go. And I don't know, like, what time frames it's on, but if they go into, like, iOS at all and fights with Apple, that would be really.

[Mark]

I mean, I'm sure it's out like it's about a whistleblower. I think it's about what was the political thing that had the really scary sounding name?

[Cody]

Cambridge Analytica.

[Mark]

Cambridge Analytica.

[Cody]

Thank you.

[Mark]

Yeah, I think it's about that.

[Cody]

Great name for a Mac mini, dude.

[Mark]

Oh, dude. Yeah, see, that's good. Now we're cooking. All right. Yeah.

[Cody]

Sweet.

[Mark]

So, dude, we got some great movies coming out. I'm excited for the Odyssey, and I'm excited for Social Network, too.

[Cody]

Yeah, I actually really want to see that. One note on that, and then we'll actually talk about marketing. So I met Jeremy Strong, went to the Telluride Film Festival last year. It's awesome. It's on Telluride, Colorado. And it's really cool because, like, it's such like a safe town and it's just, it's like the whole town shuts down for the film festival. So, like, celebs just like walk the street and you see people. And I would just see Jeremy Strong walking back and forth like all day. I don't know why. He's just like the funniest guy. Like, I think he's an incredible actor, but he's just like a funny guy that like, even you just watch him walk, like, and he's just like. He was very nice. I met him. He's really nice. But I think he's a great actor and I've heard he's like, really good in this.

[Mark]

Yeah, yeah. No. Extremely talented individual. All right, sweet. Let's get into it here. We're kicking off the episode. Talking about on site testing CRO landing page testing, the best way to do it. This was sparked by a conversation that happened on X. Andrew Ferriss was sparking conversation around the best way to ab test landing pages on meta, which, like, I want to get into it, but like one of the main takeaways was like, everybody seems to do it a little bit different. So there's no, there's a thousand ways to scan a cat. And that's true with landing pages or landing page testing as well. I know at Jones Road you guys have been ramping up CRO, your CRO efforts, your landing page testing. So I'm just curious, like, how are you guys currently approaching developing iterations of landers, new landers, and then how are you ultimately testing those?

[Cody]

Do me a favor first because I saw the tweet and I have it open now. I haven't actually spent time going through it. Like, can you give me a little bit more context? So like Andrew was wondering and start conversation just about like, is it better to set up in Intelligence and meta like, like one to one at like, is that kind of right?

[Mark]

So I mean, it started out with just a very, very high level directional question of what's the best way to test landing pages on meta. Some people brought up the native A B testing within meta. Some people said intelligence. Some people said in berry hot link to one of his videos, he did like a full deep dive on it. And I went through that a little bit. A lot of people talking about just taking your best creative, building out new ads with these new landing pages, launching those in addition to what you currently do, that that might be the best way to test new Landing pages. So it's like that's the spectrum at which people were discussing it. And yeah, we were just as a community trying to get to the bottom of the best way to do it.

[Cody]

Cool. I'm curious, I'm curious to talk about some of those in more detail than hear how you guys are doing it. I would say it depends. I think a few things. It depends if this is like an a B test. So I think you have, you have, you know, iterative A B tests that are like, hey, I'm going to test this page, but I'm going to change this copy or I'm going to, you know, change the destination from this pdp, right? Are you going from, you know, are you going from landing page to checkout or are you going from landing page to pdp? Like something like that. That to me is a. A B test and intelligence, and you just have to make sure you have enough traffic to do that. So we currently have bunch of different landing pages live in our ad account and at most times we're setting KPIs for this. We have AB tests that are running in intelligence. The pros. Right. Like, because I don't think there is like a right or wrong is what it seems like for this. So maybe we'll just talk through, like, pros and cons of each.

[Mark]

Oh, for sure. This isn't a quiz.

[Cody]

Yeah, no, but I'm just saying, like, I'm going to give what I think the pros and cons are and feel free to jump in because I think there are pros and cons of each. The pro of that is you get clear learnings. You get, hey, this, this did better. Right? First of all, you. The audience pool, you know, you're sending the same traffic to it and it's split and you're getting. You're changing, hopefully one variable at a time. So you're. You are, you know, doing that. Those are the pros. The con. It's the proper way to test. Probably. The cons of that is obviously it's slow. You're testing one thing at a time. Also, depending on how many landing pages you have, you might need to do that across, you know, many of them. And we have now fixed this. But we, but we have a. We used to have like a flicker. I've even seen it like on some of your guys as well. Like, there's like a little redirect. You know, I don't know if that is how. How pro or bad it is, but I even did see a tweet a few months back of somebody looking at their Meta CPMs when they had, you know, software a b testing live and they saw their CPMs go up. So I have no idea if that's true or not, but I just do want to call that out.

[Mark]

Motion just dropped their 2026 creative benchmarks report and it's been getting shared everywhere. Slack channels, LinkedIn, Twitter, sharing it in our private group chats. And it's great because everybody's been asking the same four questions forever. What is normal? How many ads should we actually be shipping? What is a healthy hit rate and which formats really win? The report analyzes over 575,000 creatives from 6,000 advertisers and over a billion dollars in ad spend to answer these exact questions. And the report has some really interesting findings, like the fact that only 4 to 8% of ads actually become winners and over half of ads actually lose. And for Motion customers, this report is especially helpful. You can upload it into your Motion dashboard with their Runneth AI chat and compare it directly against your vertical benchmarks. Hit the link in the show notes, I promise you won't regret it. And as always, go to motion app.com and tell the marketing operator sent you. Now what happens. What you just laid out was you have like a winning, call it a winning landing page. And also let me ask like you've talked a lot about like the AI enabled workflow that you've built out with CRO. Is that pretty iterative? Like do you have like a winning landing page? And then you were, you were building variations of elements or copy or like whatever else you were changing one thing at a time and then launching that as a test via intelligence. Is that like the majority of your landing page test, would you say?

[Cody]

Yeah, that's a good question. So yeah, we probably have like six templates. I will say I think we have over developed pages. Like even prior to Claude when we were on journey, like I think we over developed pages and under optimized them. And so I went in the direction of like creating a ton of pages. Like every ad, not one to one, but like every ad should have its own page. And it's just there's a lot of diminishing returns there. There's not a lot of leverage to that and a lot of diminishing returns. And it didn't really improve performance the way that we wanted and it's not worth the cost, you know, the time. So we have like six templates, right? Like let's say you have a Listicle, you have an advertorial. We call it a performance lp. We have a few different templates. When we have a product launch, what we'll do is we'll take what we think are the two best ones or the two winners. And our first test, right, we've got a new product is we just. We just set up a AB test for the three of those, right? Like, we might run to homepage, we might run to a listicle, we might run to our performance lp. And I wish it was like, listicle always wins. Performance always wins. It depends. Some products do better on a listicle, some products do better on another one. So that's like our first test usually. And then we try to refine it in a way where like, all right, cool. Now this listicle works. What's the best test on that one? Is it the order of things? Is it. You know, we don't really test offers much, but is it the offer? Is it the destination? So that would be it. And so we might have spend on six different products on our ad account. And, you know, six different landing pages are kind of getting a B tested. Um, that would be like the first one. And then we also, again, we still test running the homepage. Cause like, that is working a lot. When we had Dylan Ander on a few months back, he talked about it and I'm loving that. Like, on. On one hand, it's like we're spending all this time building these landing pages and. And they're not performing, right. And. And on the other hand, it's like, that's actually great. Like a. I actually don't love our homepage. So that's why we've been testing our homepage a lot. It's like, there's so much more leverage in that. Like, if we're gonna run traffic to that and it's already beating it, like, let's get it as good as possible. But also like, that's just one page to manage now. That's so much easier. So I. I think that's one component. Two other things we're trying to do is obsessed with like, David Herman's been talking about this a lot. Like landing page diversity being actually like a signal to improve your reach. And I believe in that, especially with gem and sequence learning. So not everything is an either or kind of like creative. It's kind of in addition. And so I don't necessarily know the way to test it, but it wouldn't be an a B test. And you're trying to get out of the Mindset of which template do we go with? But it's actually like, how do I get these different ones going? So we tried to. When we had a new launch recently and I don't know how successful this was, but we were like, hey, we're just going to have, we're going to build five landers and we're. And we're going to have landers for different funnel stages. So we have like a EPDP is like an enhanced PDP that's like a bottom funnel page where you got your buy box at the top of it. You know, tons of social proof. That's a bottom funnel page. Let's have our bottom funnel statics going to that. Let's have our, you know, we, we pulled a page out of the Ridge Playbook and had our branded videos. Like that's going to more of a top of funnel page. Either a homepage or a 10 Reasons why advertorial. That's like long form that goes through like a longer funnel. So like trying to do it that way, I can't tell you how effective it is, but that's definitely the goal. And in that case, like those are not like a strict test. That's more of like a throw everything at the wall.

[Mark]

I'm glad you said that. That's where some of the tweets landed. Brad from scalability School, another great marketing podcast and Andrew was kind of one over on that one too where if you're testing, they would call it testing. I would argue it's not even really a test, but they would say if you're launching like a new template or something like whatever you want to consider like a net new landing page, that you're actually best off doing it as new ads, new ad sets. Because Meta will use that information to potentially serve to different people. And I think that's, I think that's an interesting one. We just saw a really interesting data point that kind of pointed to that. So this was with Gut Culture, the fiber brand that we launched earlier this year. We tested a significantly different offer. One was our like main one was $39. You get these gifts with purchases one month subscription. What we tested was an $88 bundle quarterly subscription. So if you look at like, if you look at like a Minerals or a Primal Queen or an ima, like a lot of these subscription brands are pushing really high AOVs early on for like quarterly subscriptions versus monthly. Just makes a ton of sense if you can get it to work. So we wanted to a B test this. We set it up via intelligence. And it was no joke, the worst test we've ever launched. Like, and you can imagine how this was set up. So we've got our ad account, we've got our ads built out. Everything's been driving to this $30 per month subscription funnel. We've been doing this for months now. We just all of a sudden one day wake up and we begin splitting that traffic From Meta into two ways, going to the $39 and the $88 funnel. The $89 funnel, something like that. Worst test we've ever launched. Conversion rate on the $88 product is essentially zero. So we like double our CAC and it screws the whole thing up. And it was like, it was far and away the worst thing that we launched. Initially, I was like, oh, this is, I mean, at the very least it was a, it was a strong directional learning. That's where I'm at with CRO2, where I'm like, as long as I get a clear read one way or another, I'm like thinking of it as a win. And so what we do though is we say, hey, maybe there's an issue with the A B test here. And this comes down to like, how much does the offer landing page or how much have the learnings that have accrued within Metta? How much does that dictate who Meta is serving those ads to? Because I could hear the argument, and this is kind of the way that they were positioning it on X that Metta is serving to people that buy things that are 30 or $40 like that. There are just, there are price conscious consumers. Meta's obviously familiar what people are buying online and it's just going to serve to different people. And not only might it serve to different people, but it's learning to serve to those people over the last couple of months. So us showing up and splitting the traffic without Meta having any idea that we're doing it is actually just a terrible way for us to actually ab test that offer. So what we instead did was we just launched it as a new funnel and we have seen it be like a very solid performing part of the ad account. And I think that's a great example of like how on site changes can influence who Meta serves ads to. And if you're doing that, you're actually best off not doing it as an A B test and instead of just like launching them as separate funnels. And that's why I said it's not even really a test. You can almost think of it as just being incremental to what you're currently doing. Like, if we are splitting that traffic and seeing terrible results on, on intelligence and then launching it as separate ad sets and seeing pretty good results, then very clearly meta is like driving different people to each of those offers. And therefore I would, I would argue that it's likely more incremental or we're able to serve to a different person that we otherwise wouldn't be. So that kind of got my wheels turning around. Like, what are other opportunities for us to be launching new. Whether it's landing page templates, offers, whatever else, not as tests, not a B testing, but instead thinking them as. Thinking of them as, you know, diversity for the meta algorithm.

[Cody]

Super interesting. How, how much are you talking about? Because I think there's two things there. How much are you mentioning the offer in the ads? Like, because there's, there's what you're saying, which I 100% believe in. But then there's also funnel congruency is like some, some landing pages just don't work with specific ads. And if you were talking about your old offer and your different ads or something like that, like, it's also like, yeah, maybe it's a better landing page, but it's just a bet, not a worse landing page for those ads.

[Mark]

So that's a good point. I mean, I think it is a relatively good example of listicle page that we were driving to. Each of them, like, exact same value props, exact same messaging, exact same copy. It's really just the offer that's different. And the ads are fundamentally not that different either. We're hitting the same sort of value props, we're educating around the product the same way. Um, there are definitely like some differences as we launched it as a separate funnel, but to me, it just was a, was a pretty good directional sign that meta's using the, the, the details from the landing page and the PDP that you're driving to in order to influence who it's serving ads to. And if that's the case, then that the intelligence test was just kind of DOA from the beginning.

[Cody]

Yeah, that makes sense. So, so two things on that. Yeah, there are some things that are not a landing page, a B test, and they're a funnel test. And so if like the ads didn't make sense, like, for example, you know, we'll do like sometimes when we have a macro influencer, we'll do a, we'll do a landing page with them. Like it. Only that landing page only makes sense with the ads. And so there'd be no point to intelligence test it for that. You know, sometimes we'll, we'll do a new funnel for them and then we'll do an intelligence test on their page to get like a variation. But yeah, that's like just a funnel and it only makes sense. So I think that congruency and like you're trying to find incrementality and you're just looking at the total funnel, not just like the site metrics. But the other thing is, I do agree with that and believe that obviously Meta scans the landing page. It actually scans a few different pages and can infer so much about what's on that page and who it's going to do it down to the offer, clearly. Right. Like we have the hypothesis right now that some of our reach issues are price related and actually that's, it is like we don't really do offers and stuff and discounts. But like when we see promos, obviously there's, there's really interesting signal data that like Mandar from Blood out has been showing us. That's like, you know, Meta just doesn't think you can acquire more people at your. That want to buy a full price right now, which is like, is it meta? Is it the economy? Like it's, that's actually probably both because that's always a funny conversation. It's like, is it meta or is it the economy? But so clearly meta, to your point, I think like not only knows who's interested in what product, but even like at what price point, like I'm sure they know. So I think that's a really good example. What, what would you do if you were gonna test a new landing page? Let's say your winner was a listicle on a product and you wanted to test a different landing page, you know, a different template. So it's not like an iteration, but it's like a new page. Would you test that in Intelligems? Would you test that in Meta? Would you do an A B test in Meta? How would you do that?

[Mark]

Well, let me give you, let me give you another example because again, like we have a relatively rigid approach to testing and I don't know if it's perfect. It's definitely like, it's definitely what we've operationalized around. And again I'm of the mindset of like, if we can just get more done and we have committed to a system of thinking and we continue to make decisions with that system of thinking, we're going to make better and better decisions over time. Is It a perfect system? Not at all. But I'll give you another example. We just launched these Atomic Purple Power banks a couple months ago. We're sold out of them now in the US but we launched those originally driving to the Atomic Purple Power bank pdp. So we've got all the ads, all the videos are driving straight to the pdp. There are best performing ads in the account. A lot of our Evergreen or BAU Tech products drive to a listicle that it's just a standard listicle. Six reasons why we've got Marcus Brownlee featured on it. Things like that. That's our go to landing page. My thinking was, and this is like, look, I understand that this is flawed even as I'm about to say it. If our best performing landing page for our BAU is a is a six reasons, it is at the very least worth us testing that six reasons against the Atomic Purple Power bank page. Because even though these totally and even though it is like very product specific, like people are still getting introduced to the brand and they still likely need some of the education of that six reasons to in order to. To to convert on the product despite it being basically an additional step between the the PDP that they were originally being driven directly to. So we tested that and the way that we literally set this up within Intelligence is we set up a URL split test. We say, hey, if you're being driven to the Atomic Purple Power bank page, if this is the landing page, send half this traffic elsewhere. Now then we target via UTMs like traffic from Meta. So all of a sudden we're testing all the traffic. Anything on Meta that's driving to the Atomic Purple Power bank page. We are AB testing against the six Reasons landing page. That's how we set up the test. We found that the six reasons continued to win. So we switched everything over. Like that was basically the the approach or at the very least moving forward. Any new ads that got launched, like sometimes the way that we'll do this also is we'll just effectively end the AB test and make it 100% split. So we'll just. Everybody who matched the criteria I mentioned earlier is driven to the six Reasons page. And then any new ads built moving forward we would use the six Reasons page. What that completely ignores is any sort of diversity on the landing page side and it completely ignores the idea. It's like, well maybe some of these Power bank Pages should be PDPs and some of them, these Power bank ads should be PDPs and some should be Landing pages. We've completely removed that level of nuance, but we at least identify, hey, we observed, you know, hundreds of orders in each cell, tens of thousands of people on the site. We saw like a 15% increase in revenue per session by driving to the six reasons. That's sort of like the global optimization that I really care about and that's typically how we've structured a B test. So lack of nuance. But we're at least we've operationalized around the testing and we're getting clarity around like what is the global best landing page for, you know, this type of ad?

[Cody]

Yeah, I agree with that. I think that makes a lot of sense. And there's no, there's clearly no perfect way to do that. I was just thinking you said on a few podcasts ago that your toxic trait is more volume. I just found that to be funny and I don't know why I just thought of it while you're talking through it where it's like super, just like regimented, like, hey, we're just going to test this way and we're just going to go have this practice and maybe it's not perfect, but I totally agree with that.

[Mark]

Before Q4 hits, the smartest e commerce operators aren't just asking how do we spend more, they're asking how do we make every visitor more profitable? That is where after sell comes in. After sell helps e commerce brands increase revenue per visitor, contribution margin and incremental profit by optimizing the highest intent moments cart, checkout, post purchase and even the thank you page brands like Hexclad Ridge and Pilar increasing their AOV across every cohort using these zero risk upsells we've tested and they have no negative impact on conversion. It's just more free money in your pocket. And with rocked thanks. You can offset your after sale subscription. No redesigns, no risky funnels, just better economics. Want to see how much money you can make with after after sale? Go to aftershell.com Rocket Revenue Forecast, put in your numbers and see how much more profit your funnel will make before BFCM starts. I have no idea how else to get to to understand the value of multiple landing pages in the account. It just ends up being like that's why, that's why I made the argument earlier. Like I don't feel launching multiple landing pages within the account is actually testing them at all. I can hear the argument that you're making the ad account better and that's I guess how you're managing it. You're Saying like, hey, performance is good, business is good. We're hitting our goals. We happen to have all these landing page testing live. Maybe I've seen some interesting stuff around CPMR where lower CPMR stuff are being driven to different landing pages up top. And that's a sequencing thing that you mentioned earlier. So maybe there's like some. Some data points that you can kind of point to, to like, feel good about that decision. But otherwise I find it really difficult. And I find we would just be.

[Cody]

You.

[Mark]

You end up running the risk of like, analysis paralysis where it's like, you don't know exactly what value is being created here. I don't know exactly how to ab test this in my account. I don't know how to, like, continue improving on the results that I'm getting.

[Cody]

I agree with that. I'm a little surprised you're saying, because usually I'm the one who's like, you got to do a structured incrementality test. And you're like, no, like, you can see it in the business, you know, but like, you're not saying you don't find value clearly in like, new funnels and building landing pages for new funnels or like, not testing them. You're just saying you don't call that a test, but you still think it's worthwhile. Or am I wrong?

[Mark]

Well, we end up. We end up like. What we do now basically is we have landing pages for. If you think about our different categories, we typically have a main. We have like an evergreen BAU where like, we are. We are acquiring people like, independent of like a given colorway or collection. And we typically have like one really mainlander for that. And that would be like the six reasons. Wallet page or whatever, Smart wallet page, something like that. Then the, like, bottom of funnel stuff. The only time that we're really driving traffic to anything other than that page would be something like catalog ads. That's like the level of nuance. Now we have a ton of landing pages live in the account because we have things like Father's Day ads will go to a Father's Day page. We'll launch a new capsule. We're launching a collection called Jungle Ops. And it's like, that will get its own dedicated page, so it will go there. So we have a lot of different pages, but we don't have, like, if you think about each of those as like a given, you know, program within the business, we really just have one optimal landing page per program instead of multiple landing pages per program. And we just are doing less of the, like, hey, let's get multiple landing pages live across different ads in the same ad account. Not test them, let them live alongside one another within a program. Just because I think, think that is just. I again, I just, I don't know how to make decisions around it.

[Cody]

How many programs do you have? How many like skus do you have spend on in your ad account? So I think that's partly why. Right, like, so the one I explained where we tried this full funnel thing, again, I don't even know if it's worth it. That was like projected to be like a top three SKU for us, right? Like a top three product. I would absolutely not do that on anything that's not in like top three. Right. Because our top three products are probably getting 60% of our ad account spend together. And then, and then we probably have six to 10 different products that are like, some of them are actually not even worth building a landing page for. Like, some of it we run to PDP and my opinion is like, hey, I don't really know that this is going to be an acquisition product, like a high spend. And like, I want the MVP to be strong enough that it converts going to be like fine enough to go to PDP that then I can take the time to build a lander. Now with Claude, that's all different because you could just build, you know, build it easily. But yeah, everything else besides the top three should have one landing page. And that's where if we do test it, it's. We'll test a listicle against a homepage. And I almost hope the homepage wins because then I don't have to go and maintain this other one. There is like a operational cost of, of all of that.

[Mark]

There's an operational cost of it. And I think there's almost like it just feels like you're. It's a burden over time having more and more landing pages live in the account. And again, I think it like, I think it becomes more difficult to make decisions around what. What's working, what's not. So that is like, I don't know. That's where I land on it.

[Cody]

That's where you land on it. Let's see. I like what you did there. So, so, so unrelated to it, like, I a hundred percent agree. We're consolidating because it is like, it's much easier to have 404s and stuff broken when you have more pages. There's just like a lot more that can go wrong if you're Updating something and you have an image that you can no longer use. You have to look at all the pages. Right. Like it. It. There is a lot to it.

[Mark]

Or here I'll give you one more on that. Like, like we go in, we go in and out of these promo periods and we like reskin all the landing pages so that our evergreen ads get the updated messaging of like hey now we're now it's Father's Day or whatever. And if I had more landing pages live it's just like it is just yeah operationally harder to track down all those things. It's not even breaking but it's just like if I want to continue to iterate and optimize over time it becomes really difficult to do it if I'm running across 8 pages instead of like 2.

[Cody]

Completely agree. Completely agree. The one where so some nuance. And again I don't have data on this except for percent new and CPMR when we launched we're no longer using Journey but we did for a little bit before Claude. The Journey pages aren't a subdomain and there are also different pages. This is when we're doing like the very specific like Persona funnels. And so it was different messaging, different creative on them, different demos. We did see really good performance and it really improved at the time our percent new and CPMR and yes it was whitelisting ads and partnership that should help with that. But I also think the landing page and so my hypothesis was and who knows if it is. Is it the subdomain? Maybe the subdomain had a lack of data, kind of like a new pixel and it, it allowed us to explore a new audience and get out of rut. Maybe it was creative, I don't know. I just do think there is something there and I think that's more of a new funnel, an incremental funnel. But I do think it's there. I'm. I'm interested in testing like third party pages to see if you know you can do that. And like there's this one company, I forget what it is but where like they, it's. It's not like it's like a quality edit but they don't do. It's like much more. They, they kind of facilitate it with like large publishers like New York Post and stuff like that. And they charge a percentage of spend but like interested in trying something like that. Right. Just to get like to use actually the domain as like a signal source.

[Mark]

Yeah, I think it's an interesting idea and I guess just what it comes down to. And all businesses are shaped differently because I think the right way to put it would be like, if I think about programs, again, we have like our BAU Wallet program, but then we also have three new collections at any given point. It's just. How much diversity, how much landing page diversity do you want to commit to and maintain within each of those programs? And I agree. If I was. If I was more of a. And then also for Ridge, right, We have like four different categories across five different markets. So it's just like we have so much redundancy that we have to be accounting for. But if I was a brand closer to like, you know, low sku, high LTV business, where I'm like, really the path to growth there is that horizontal expansion. We've talked about this all the time. The AG ones, the Creates the rises. I can just like landing pages, Persona landing pages, like you just mentioned, are really a key part of my growth. And that's actually. I'm not going to get the new capsule launches like Ridge does where we're launching three every month. I instead need to get that diversity in other ways. And then maybe you're just a little bit more incentivized not to be a B testing more pages, but just be launching more. More in the account and trying to build it up that way.

[Cody]

Yeah, no, even for. We've had that conversation even for how we split our account because, like, you know, we've really tried to split it by like Persona or Demo, but it's like. But then if we also are doing it by product and by, you know, sku, it's like, now we have so many. It's much easier to do it that way if you're a, you know, single skew or few SKUs. But if you have a lot more, it's. It's harder. And you kind of have to pick how you. How you're going to slice your account, which is. I don't know what else you want to say about Landers, but it's almost a good segue into, like, into the other conversation we were going to have of product expansion. We talk about incrementality a lot, but how do you actually operationalize it to make your business better? That is one thing that I've been really leaning in with my team recently. And HAUS has played a tremendous role. We use it for all of our experiments, all of our geolift testing, but we now use it for our MMM as well. I've been a design partner. I've been one of the early design partners on House's cmm. C stands for causal. So it's one of the only MMMs, if not the only MMM that I've seen that's actually using your causal experiments to build the model. And so that allows me to just trust the data so much more. So it's not a black box. So it actually informs our roadmap and has been so crucial for allowing us to operationalize around incrementality. The House team is world class. I can't speak highly enough about them. They've also built a really amazing community with some of the best DTC growth operators out there. They have a few exciting events coming up soon that they call the Houzz Growth Lab. One is in LA on May 19th and the other is in New York on May 21st. I highly recommend checking it out if you're in the area. If you want to check it out, learn a little bit more about cmm. Go to House IO operators to start making better data driven decisions today.

[Mark]

How much traffic are you able to drive to a given AB test? And like, if you want to end up being more specific, like I was mentioning at Ridge, like we test relatively broad. We're testing like all of our BAU ads within Meta between. We're splitting all that traffic between these two landing pages. So we're like driving a ton of volume really quickly. If you want to be more nuanced after that. If you want to run a new variant or an iteration on a landing page that's just kind of a subset of your account, you're like driving down the amount of volume that you can drive and it's going to dramatically extend the time in which it'll take to figure out if you have a winner or not. We had the GM of Jackson on a couple weeks ago. Connor and I spoke to him and one of the things he advocated for, which I thought was interesting, was just more AB testing with upper funnel metrics as the key metric. So like if I am testing, I've just got like a subset of my account running to a very specific Persona landing page. I want to test a new element on that page. I set up this A B test rather than waiting weeks to get statistical significance around like driving better conversions, maybe my goal should just be PDP views so that I can reach that conclusion much quicker. And even if I'm wrong again, it's just like the velocity at which I can make that decision and being confident that it's at least slightly better is just going to that's what's actually going to pay off over time. What do you think about that, like general strategy of adjusting your key KPIs when you're testing on smaller amounts of traffic.

[Cody]

That's super interesting. I'm excited to listen to that one. I don't think I do a good enough job looking at those funnel metrics and getting people to the next step and stuff. And I think I more just look at revenue per visitor and stuff like that. So I'm excited to hear more related to that. This is one I always toy with again, especially we hired a new CRO person we're talking about. All right, what can be. Dylan Ander calls it a hot fix. What can you just change on your site without testing? And I think there's ways you could look at and monitor data just to see if it was a good idea or bad idea. Because again, with how quickly we can develop things now, I don't want to wait three weeks to test everything. Right. I just, I'd rather just get a new PDP live and obviously get some metrics on it. But there's that what needs like a proper AB test where you get full stat sig and then what can you do? And this is very controversial to a lot of people. What can you do a test on and end it early? Just get directional learnings. Because I remember there's two things that are behind it for me. I remember talking to Zach from House, the founder of House, like two, three years ago. And at the time he's like, no one's taking me up on it, but I really think the right way to run an incrementality program is really high velocity of low precision tests. And I was like, you're probably right. Yeah, I'm like, you're probably right about that. And I think again, I don't know anything about statistical power, but it was like, if you're 80% power, it's worth it. And so I remember talking to Brian from when he was at True Classic and he was telling me that the way that they tested, I don't think they still do it, but for a long period of time it's like they'd see an idea, they develop it, they'd get a test live, they'd run it for a day or two and then ship it. And like, you know, you could probably get a lot of false positives doing that way. But also like, if it's not something that's gonna have a huge downside, like what you gain in the velocity of Testing is probably going to outweigh any. Like, you'll have enough winners that you can bag by just, like, the speed that, like, even if you make one wrong call that maybe wouldn't have looked good if you went to full Scott Sig. Like, to me, it's probably worthwhile. It's riskier. How do you feel about those two things?

[Mark]

I totally agree. And that's kind of what I was getting at. I mean, we ran. We ran the most house tests of any house client last year. And it's not because they were all extremely high precision tests. Right? Like, we are, we are more than comfortable sacrificing precision for volume. Yeah, I totally agree with that. I think. I think it was Drew Marconi from Intelligence. I think he was also on the Andrew Ferris podcast and he called. I believe he called it a canary in the coal mine test, where, like, you just test it to see if it's, like, terribly broken. Like, like, that's just. Hey, if it's not like, we're. We, we. Whether it's strategic or subjective or whatever, like, we, we have this difference, this new design, we like it more. We've decided just test it to make sure that, like, nothing is catastrophic about it. You can easily run that for a day or two or three and then just call it and be like, okay, we don't really care if it's better or not, and we definitely don't care if it's statistically significant. We just want to get this thing live with some level of confidence that it's not like a terrible net negative. And I'm all for, I'm all for tests like that. It's just better identifying, like, what you truly want to get done with a given test and then just making sure that, like, the way that you're measuring it aligns with it.

[Cody]

I agree. So what are your thoughts on the. The guy's upper funnel thing? Is that something you're. You.

[Mark]

I like it. Especially on lower volume stuff. I mean, when it comes to. To gut culture, like, the fiber supplement, like, is such low volume that our initial test, like, we were looking at all sorts of soft metrics, and that's where it's like, let's just see what people are clicking around on the heat maps. Let's look at the rate at which we're getting people from the listicle to the pdp. If we're doing that at a significantly higher rate, like, we feel relatively comfortable that that listicle is better because otherwise the alternative would be like, literally just sitting around forever. Running a listicle test. So I'm like, I'm very comfortable doing that. And I just think if I'm a brand again, this goes back to like, if I'm a brand who wants to be running more landing pages in my account and I would assume that I want to continue optimizing those landing pages. I want to be iterative with those as well. Then I also probably need to get comfortable with like looking at softer metrics in order to make decisions around those tests.

[Cody]

Yeah, I like that. Are you setting up like custom events in IntelliGems to track anything like on those pages? Do you ever do that?

[Mark]

What we do mostly is we'll use Intelligems and then we'll build the segments in GA4 and then we could look at all the, all the events within GA4 between the two intelligem segments.

[Cody]

Got it, got it. I'm allergic to GA4, so it's.

[Mark]

Dude, it is terrible. I would like, I don't wish it upon anybody, but you end up having like GA4 has a bunch of events. It has all the stuff set up. We'd be better off like moving it over to Intelligence, but we haven't yet. All right, I've got, I've got one more testing thing. This is a very personal one. We are launching a new site. By the time this comes out, the new site will be at the very least in the middle of an A B test. So full, full theme redesign. Big like 9 month project. The whole, the whole soup, soup to nuts approach. We're going to test it via intelligence. We're going to do a theme test. You can imagine if we're redesigning the website and rebuilding the theme. We made a lot of like, we made a lot of different, we made a lot of different decisions. Like in terms of we have some goals around. Like we want to be driving upt, driving up average order value, making kits more navigable. We changed the hamburger menu. We've got all these new like modals across the site that we could use for education and navigation and things like that. One of my concerns with this test is that if we were to. Now we're in the process of merchandising out. So like actually making the decisions around the new theme and web design. One of my concerns is we will change every variable we possibly could. We run this big A B test and like, I mean none of them are good. Like if there's a really good chance that the test is a slight negative or even a big negative. And if we make too Many variable changes, then I have no idea like what to attribute that like drop in conversion to. So I'm curious if you have any thoughts or advice on like the best way to. How would you think about this? How would you merchandise the new web design in order to run an A B test? And then I guess the second question is like, what do you prioritize in testing after the A B test where you actually begin rolling out some of those new features?

[Cody]

This is very relevant because we talk about the same thing all the time, more with for example, redesigning a pdp, right. And actually Fable came out yesterday so I tested it. But we had a meeting and onboarding our new CRO person and it's like, well, how do we want to set this up? Do we want to iteratively a B test it and have six, seven months to get this new pdp or do we want to do it and we do this thing where we actually usually. And it's also, I think controversial. We don't usually a B test, we usually multivariate test or like maybe it's not technically multivariate but like we'll have four variations. We'll have a control plus four if we have enough traffic on it. And we were working with an agency and that's how they did it and just, you know, adopted it. But so try to go in between where it's like we are changing more than one at a time. We just did this with like a collections page test as well. We in an ideal world. So you can get multiple variations. In an ideal world you have one variation between them and you stack them. So let's say you're testing a PDP and you change the button on variation one. Well, on variation two, you keep that button that you change but then you add another thing and then on variation three, you go those two things. So at least there's some consistency amongst them. I will say we don't always follow that. Like sometimes there are two variations changing like the amount that we want to change on our pdp. Like I don't know that it makes sense. Like I want to make our review section better. It's so low on the page. Like I don't know that I want to spend two weeks on an A B test for that. So I think we're going to go more of like, we might go like themes instead of like variations. It's more of like what's like a hypothesis and a theme and then like go wider. At least have something. But yeah, so we had that happen Last year where we redesigned a pdp, we tested it, it didn't do well. And we were like, oh shit, we really like this PDP better. We changed so much what it is. It was a really good learning and I think we're much better at zero now because of that. But it was definitely, that was a challenge. If it did well, you don't really care why. And then you could test it from there. But if it doesn't do well, I think that's the harder one. And then obviously that sucks. So my opinion, if I were to do it, I would, I would change as little as possible. If it works with your new site, build a hypothesis. If part of your site design was like, no, we actually want to improve our. The way we're merchandising and like, then that is part of the new theme. And I would just do that.

[Mark]

Rich Panel CEO just predicted his own SaaS product will lose 90% of its revenue and he's the one building what will kill it. They are betting that AI agents will replace traditional support tools entirely. No dashboards, no seats, no workflows. So what did they ship? An autonomous AI agent that handles customer support from end to end, reading order data, issuing refunds, updating shipments without a human ever touching it. It means you can run a faster, leaner support team at a fraction of the cost while resolving tickets instantly. 24. Seven rich panel is cannibalizing their own business to build the future of CX and support for brands. Because if they don't, someone else will. Go check out what they're [e-mail] and learn how you can automate your entire support operation. Today the initial test is going to be like, let's get the two versions of the theme as, like, for, like, as possible so that we remove really some of the like, random variables. And we're even like, you know, as we've been building out the new theme, we've continued to test new strategies on our existing theme. So we found wins. Like, we have this cool, like, bundler that's like working pretty well now. Now the bundle is not built out in the new theme. So we're even like adjusting the current theme, removing the bundler again so that, that, so that it can be a proper A B test against a new one. Because again, my fear is like. And it. But it makes sense though, right? Like, actually an easier example is like, we have this really, like, we have this progress bar built out that conditionally triggers based on like, what products you're adding to the cart so that we can do like, different break points by all of our different categories. And it's been like, it's kind of a hairy thing to have to deal with. We don't have it ready yet in the new themes. We're actually just going to turn off the progress bar in the old theme. So I'm actually down to like, not only merchandise a new theme as similar to the. As similar to the old theme as possible, but even change the old theme to match what is capable in the new theme just so we can like, remove as many variables as possible. So that's step one. And then what I like what you just said is we could very easily launch with the multivariate test, have a C variant that makes a few more unique decisions on top of the new theme, and then we're effectively running like a test between those, like, concurrently as we're testing against old theme as well. So. All right, I'll keep you posted. That's. That's good feedback.

[Cody]

Yeah, I'm curious. That's likely how I do it. So. All right, and last thing is, are you. Did you not build in the bundles in the cart just because of, like, timing and development and you'll plan to get them later, or was that like, active decision?

[Mark]

Okay, no, no, no. So they're coming and like, and that's the thing where I think what this will look like for us is like, let's. Let's a B test the new theme in as, like, for like, of a way as possible. We are, we are like basically consciously understanding that we are not testing all of the wins that we think we potentially have. We're sacrificing some of like the newness of. In order to just get a clean read on. Like, hey, is this not. And as honestly, like, honestly, as long as it's not like, way worse, then it's like, okay, great, we can like, roll out with it and we can be confident in that decision. And then we have like a very quick testing roadmap after that where we could be rolling out the bundler, we'll roll out the progress bar again, we'll roll out all these other features that we've had and the new ones that we've built, but we'll just do that more like incrementally over time.

[Cody]

All right, I said last thing on this, but I'm going to. I'm asking another question. So what was the decision to do a new theme? Is it. Is it mostly design? Is it actually like the architecture? Is it the function? I Know you guys used to have multi site. I don't know if you still do. Like are you doing one? Is it AI? Like is it all of it? Like what? Walk me through the thought process there. Yeah.

[Mark]

Okay, so I have a big doc, so we could take this as a talking point for a future date as well. It is kind of all of the above. Like one of them is like the founder is just tired of the current design. Okay, perfect. That's like, that's super, super valid. So that's one. It's not far and away, not the only one to our current theme we've had for like eight years or something. So there's just like crud built up. It's like a really old theme. It hasn't been maintained all that well by different developers over time. So like there is a lot of value in just like let's just go blank slate. I'm not technical, so I don't know all the technical terms. But I also think going blank slate and building it up more correctly will allow us to test faster over time, launch new features. Just that cleanline will sort of compound over time better international support. So we're still multi site for all of our different international stores. And then within some place like the eu, we want to be leveraging Shopify markets more for personalization. We just needed a theme more purpose built to like handle some of those nuances. So like that was a big one. And then there's the other ones that I mentioned earlier where like I think we can improve a lot of. I think we can improve a lot around navigation, bundling, you know, driving upt order value, things like that, education, all those just like nice to have things we kind of addressed in this theme redesign. So those are probably the four big buckets. But again, I'll come back in a couple weeks with, with the doc that I built out because it was really like cross department. Like everybody kind of had like different objectives that we could roll into this very large theme and this very large project of like the, the site redesign.

[Cody]

Cool. So we'll hear about it in a few weeks if it does well. Otherwise we won't speak about it again.

[Mark]

We won't even publish this episode.

[Cody]

Yeah. That's awesome. How long did this take was this? Like a year? Six months?

[Mark]

Yeah, we kicked off like we were looking at designs in November, like early, early design explorations in November. We got a really good like homepage design just before Christmas. I was feeling really good about that going into the new year and then spent, you know, January and February, like dialing in some of the smaller pieces there. And then the last couple months. Couple, I say a couple months. It's been like the last six weeks developing, like literally coding the theme. And then it's taken us two weeks just to like merchandise it. We've gotta like, dude, I mean you can imagine we have thousands of variant images and we have to update all of them so that they're standard. Like we're changing. Like we have like a very certain gray that we use on the current theme and it's going to be different now. So like we have to change every variant image that exists on the site. So there's like a lot of small creative stuff that we've gotta hammer out.

[Cody]

Cool, cool. That's awesome. Excited to see it.

[Mark]

All right, sweet. The last thing I had for you, product launch. Product launches for growth. And the reason this came up was AG1 launched creatine, which I think is. Is interesting and we can take this whichever way you want. One thought that I've had recently, we spent a lot of this was 2024 talking about DTC 3.0 high LTV low SKU count brands. The creates the rises, the, the grooms, these like very simple businesses that compound really nicely over time. And we talk about like sort of just expanding the acquisition around those via landing pages and channels and things like that. Like you just have to operationally you're like as simple as you can get in order to really just focus on being a performance marketing engine. And I feel like we're at this point where, and we talked about that as DTC 3.0 because those are really, really exciting businesses to have run. I feel like it's just going to become a rock fight really quickly in that market as AG1 launches creatine and Create launches Electrolytes and you know, Marsman launches, you know, I don't know some other thing. Right. Like, I just think over time all of these high LTV subscription brands that are fantastic businesses in and of themselves, if they want to continue growing and capturing the value that they think that they can capture, they end up becoming multi SKU brands and they end up like seeking growth through product development and product launches. So I think we're going from a part of the industry being extremely single SKU focused to like very quickly becoming multi SKU focused. And that just comes back to like Ridge has been very multiscu for a long time. All of our growth has come from like category expansion and building redundancies across extremely different products. Do I just Think maybe we see more of that in, in types of brands that we haven't in the past. But where do you sit on product launch for growth right now? Like, what. What would you be interested in getting into?

[Cody]

Yeah, you're right. It's kind of like an end of an era where like, it's, it's, it's remarkable ag1 that they got to their size. I think they got to 600 million publicly, you know, without. With essentially one skew, which is like, wild. It's crazy. Yeah. I guess Mars Men has one. I have no idea if they have plans to do, you know, others, but they seem to be crushing it. But again, it's also, they're. They're new and on their way up and will likely add other ones, I would, I would guess. And then the majority of brands are, you know, they're like a momentous and they have a line and they have a brand and they do, you know, a line around it or like, again, this is like not a D2C brand and they're at like probably hype hyperscale at this point, but like Bloom nutrition. If you look at them, they're like, you know, they've got like everything and it's just like this, like, it just is one. It seems like it's almost like a shark ninja thing where, like, you have to be on this innovation pipeline. It's like, all right, creatine's hot. We got to go do creatine. You know, gummies are hot. We got to go do that. Like, it's just like you're almost leveraging your brand IP to then like be in these markets, you know, like, I don't know that they're competing on, they're competing on distribution and brand. And if you're in, if you're in Target, you know, which is again, a very different thing. And you need, you know, you need some creatine. You're going to get bloom instead of. Right. Whatever else is in there. So I guess it depends. I can only speak from our side. I actually had this thought the other day when listening to Zach stuck on Operators, which was really good. Do you listen to that one yet? It was good. It was really, really good. 1. We over developed. We have too many launches right now, partly because the goal was like, hey, we need to grow. We're too relying on a few products and we need to grow to do that. So we've had launches too much, like, far too much. Like probably two launches a month. Right. And they're not all net New. Some of them are shade extensions. Some of them are like promos, limit edition things. It's just too much. And it worked two years ago to have something every few weeks. It's just the consumer is tired of it. I thought it was that we weren't reaching enough new people and that they were only tired of it because we were reaching the same people. Part of it is, I think it's not about new products. It's about, you know, new audiences. And the products have to be bringing new audiences. And I think for at least us in beauty, in our category, like, you know, a lot of your audience is really your brand, and it's not.

[Mark]

It's.

[Cody]

It's harder than I thought to have a new product or new category that, like, really appeals to a new audience. Like, you kind of have your pool and you have to do, like, other things to reach them. It's not just product where maybe that's different for, like, a ridge or something like that. So for us, we've. We've gone too fast and we're going to slow it down and really try to do other things.

[Mark]

And what do those. Like, what are other things in this case?

[Cody]

Dude, I wish I knew. No, I'm kidding. I mean, I think a lot of it is going to be brand. A lot of it is distribution. Yeah, totally. And there's still some. I think we also develop too and launch too quickly. Like, we haven't really had time to really market them. We do it for two weeks and then we're on to the next thing. And so I think that's part of it. Like, we have seen some success with, like, some. I also think it's like, how different is it? This is at least in beauty, where it's like a very brand thing. Like, if you launch something that is too similar to what you have. Let's say we have a foundation and we launch another foundation. And yes, maybe it's better for different skin types and it can open up your market. Like, there's still a lot of overlap with our existing one and there's some cannibalization. Let's say you launch something so that's safer but less upside. Let's say you launch something that's very different and it's a new category, like a. We launched body or whatever, and that wasn't expected to be big or something like that. But it's like, okay, you think this opens you up to a new market. It maybe has more upside because there's potentially expand. Sam. But it's also riskier it's so competitive and beauty and you're kind of known as this. Right. You know, Salt and Stone is a deodorant brand and you know, right. Summer Fridays, this came around like the great, the great, great brands that have a really clear position and can expand in these categories and have the trust and credibility that carries over it. But most people aren't like, hey, I'm a. Very few people are like, I'm a Jones Road through and through. Like Jones Road makes amazing makeup products and I'm going to get my skincare from somebody else. And you can expand that, but that's just, it's. That's very challenging. So I don't know, I think that's where I have. And so we're going to, we are going to slow it down, consolidate. Really focus on quality over quantity in this area. My. Probably more than you. My, My toxic trait is definitely speed and urgency. And I think I have to do a better job of like thoughtfulness and strategy. So we're going to slow it down. Really, really be much more disciplined. And then we have to find other ways, use it as a constraint to kind of find other ways to grow.

[Mark]

Yeah, I think that's good. We went through that same thing. Like this current year is we had a very deliberately, you know, truncated merchandising calendar last year. We were launching stuff like weekly all the time and it was like high volume. And we just like, we're not getting the juice out of it. One thing that you mentioned that I think is like really worth, I don't know, double clicking on is you said the consumer's tired of it. One thing that we saw on the Ridge side was like our first few launches were like huge Father's Day 2023. We came out with like leather ridge wallets, like a midnight, midnight black and a tobacco brown leather. And they just crushed. And you can imagine, like, I think your brand has a certain amount of like pent up demand for newness and you're just like draining that over time. So it's not even just that the consumers, I guess it's at the consumer's tired, but it's just like your audience is like relatively finite. And if your launches are outpacing the rate at which your audience is growing, then you're just kind of gonna drain it. And we basically saw our first like eight Colorway launches, like all just get smaller over time until we figured out like, what are the Colorway launches or what are the capsules and what are the silhouettes. We need to include in that in order to be reaching new audiences. That's how we actually got them to begin growing again. Because otherwise you're just kind of like, you're just trying to get water out of the same stone.

[Cody]

Totally. And I don't, I don't know your guys lead times, but that's a challenge. I mean early times aren't a year, but same thing. Like we would launch a, you know, especially promotional things. Like we launched a mini Miracle bomb or Miracle bomb palette right. One summer, like two years ago. Crush, best performance we've really ever had. Right. Like sold through, you know, 30,000 of them in a few weeks. We next year ordered double of that. I don't want to say flop, but like less demand than the first year. And, and it's, but some of the challenges, like obviously now we know that like things don't anniversary well and we had that pent up demand and it's coming down. But you don't, you don't initially know that unless you've gone through that before. And so you, you plan for that and you order into it and you're like, oh no, it's not there. So that definitely has been a really challenging one. But it's, it's challenging with any inventory thing because it's not like a quick thing. Hey, I'm just going to change my ad spend. It's like, no, I, I either bought way too little of this or I bought way too much.

[Mark]

And to your point in beauty that like maybe you guys are known as a makeup brand and therefore it is like harder for you to get into like non makeup cosmetic categories. Do you think AG1 will have a similar sort of issue where it's like, yeah, look, I know you for being my greens and I'm not necessarily looking to you for my creatine. Or do you think maybe there's more like portability in like the supplement space?

[Cody]

I thought years ago that like they could do it because they had like the trust and they got, you know, Huberman and stuff like that. And like they do have the trust. So people kind of have a positive association with them and get it once up. I think the one challenge for them is it kind of is counter to their positioning. It's kind of like they have a little bit of like the innovator's dilemma where if now they're coming out with the, you know, I think they came out with an Omega and if they come out with a, you know, not a multivitamin but a creature or stuff and actually by the way, we have the same thing. We get the same feedback. It's like, well, you've told me for years that, like, this is my all in one.

[Mark]

It's for sure. Yeah.

[Cody]

So I think that's. That's the thing. And same thing, right, where like, we now get feedback because we've. One of our lines, like Bobby used to say, you don't need more beauty products, you need better ones. And so now people are like, what the is this? And we're like, well, we're not saying you need it all. Like, it's kind of more for different strokes for different folks, but it is. You do lose that credibility and trust of the positioning. So you have to be really, really mindful of that.

[Mark]

I wonder if in supplements. Well, actually, I can imagine it being more beneficial as a supplements brand to expand because you can have, like, cohesiveness in your product suite or you can think about them like, hey, I'm thinking about a consumer that's going to use all these things together. And maybe this is the point that I'm going to make is like, right now, people spending money on supplements are spending it across, like, a couple different brands, right? Like someone like me, I have. I have some create at my house, and they just came out with creatine and electrolytes. Before, I had elements and I had momentous, I had electrolytes separate from. From creatine. So those are getting combined. Then we have something like gut culture, which is electrolytes and fiber. So now all of a sudden, like, they're just like, not all that compatible because. Because they can kind of bleed into one another. So if I'm a brand, you know, we had. We had Matt on from Cadence a couple months ago, and I love the Cadence story because they're building out such a cohesive vision for, like, they know who their person is and. And they know that the products that they're making are going to integrate into that person's life, like, in a very compatible way. And that just feels like a more natural sort of extension versus beauty. You're, like, forced to be compatible. I don't think. I don't think there's people coming out with. I mean, I have no idea, but, like, the equivalent of creatine and electrolytes is probably less common in beauty products than it is in supplements where you're just doing like a. A little powder mix.

[Cody]

Yeah, exactly. No, so it's funny because the reason I got up is like, have like, these three here on my desk, right? So it's Just funny I have because I'm friends of a few mind body green things. So creatine plus electrolytes. So it's like, all right, what am I going to take today? Creatine plus electrolytes. Element only creatine or creatine with taurine. And then, you know, it's so, it's, it's really funny. Like there's so much overlap between these, I think. Yeah, the only one. Like, I don't know if you do this, like I do take some supplements outside of like, you know those or like a groons or an 81. And that's the one where I do like to have it be single brand just because I'm like, all right, I just want to find one brand that I trust. I know it's good value and it could be thorn or it could be momentous and I just trust them. And if I'm going to get something from Amazon, like I'm just going to get that brand every time. So that's, I think, the only one. But that's their positioning. I think the challenge with AG1 is that's not their positioning.

[Mark]

Yeah. And I just think it'll be interesting to look at. Like we've talked about product development as forms of growth for a long time because I think our businesses have demanded it earlier. Whereas all the supplement brands that we've talked about, the like quote unquote D2C 3.0 style brands are now, if they want to continue maintaining high growth rates as the space becomes more competitive, there will be more overlap and we're just going to see all sorts of brands that weren't grappling with product development now have to sort of think about that. So I'm interested to see how the space plays out now.

[Cody]

It wouldn't be an episode of marketing operators if we didn't talk about David. But, but I mean, again, I'm. I am chomping at the bit to try their ice cream.

[Mark]

Dude, I. Dude, you know what's embarrassing? I. Maybe I said this before. I DM'd Peter. We've never spoke. I was like, dude, how do I get a pint of ice cream? He didn't, he didn't message me back.

[Cody]

That's so thirsty. I did DM him after he was on open residency. But yeah, it didn't mess with me to be back. But that's, that's funny. No, I really wanted to go to their like scoop shop thing, but I think they're doing it incredibly well and even when I listened to him. I was like, this is going to be a multi billion dollar business because they're also standing up separate brands for each of them. They're kind of doing like the, the Gruens playbook. So I think that's a really great way to do it. That, yeah, I think they'll be super successful. But even, even in general, like, I don't know if you notice, like, I feel like they've pivoted their branding. Like they're marketing beauty. Like if you look, they had the Isabella or Gigi Hadid thing and then like they had this girl on like a, like a social hero video to, to promote the ice cream and the scoop shop and like she was a model. Like, it's no longer this like Jack David type thing. So they're essentially saying, hey, if you eat our ice cream, you'll look like this. Which is like, which is, which to me is fascinating. But it seems to be working very well for them.

[Mark]

Yeah, yeah, yeah. No, I'm excited. I'm extremely bullish on that business.

[Cody]

Yeah, the reviews on the, on it are pretty good. Dude.

[Mark]

I, you know, I would just love. I've just, I've thought about it so much this week. It'd be just be so sick to just have like an ice cream treat, 200 calories or whatever it is, 30 grams of protein. I'm like, dude, life's going to be great.

[Cody]

I can't wait. I can't wait. I feel like for all the earned media we've given them, this is another

[Mark]

like just shameless attempt to try to score us some ice cream.