Latest / AI DTC WTF / Why AI UGC Is a Trap (And What Actually Works) | Cody Plofker

Transcript

Why AI UGC Is a Trap (And What Actually Works) | Cody Plofker · AI DTC WTF

• 0:00
  If you work in D2C and you use AI and
• 0:05
  you're wondering what the
• 0:06
  F is going on every week,
• 0:08
  this is your podcast, the AI D2C WTF
• 0:11
  podcast, your home for
• 0:13
  tactical tips, strategies, and
• 0:14
  ideas that you can implement right now in
• 0:17
  your AI workflows to make
• 0:18
  your brand or agency more money.
• 0:22
  Welcome to another episode of AI D2C WTF.
• 0:25
  So good to be with you
• 0:27
  again, Will, and today's
• 0:28
  guest, Cody Ploffer, legend, obviously,
• 0:31
  everybody loves this guy, doing a ton
• 0:32
  with AI. The episode
• 0:34
  was absolutely mental, in my opinion, in
• 0:36
  terms of things that he
• 0:37
  talked about. What's your take on
• 0:38
  this? Yeah, I think mental is an
• 0:40
  understatement. It seems like he has
• 0:42
  become his own CRO agency
• 0:44
  in and of itself. And it's like very
• 0:46
  clearly driving results,
• 0:47
  but also like how to properly
• 0:49
  use CloudCode and how to incentivize your
• 0:51
  employees to use CloudCode. I think he
• 0:53
  just did a really nice
• 0:54
  job touching on all these things that
• 0:55
  we're all sort of stressing out about.
• 0:57
  But if you're interested
• 0:58
  in landing pages, this is the one to
• 0:59
  listen to. Yeah, AI
• 1:00
  landing pages, creative process,
• 1:04
  what is, you know, the way that the
• 1:05
  systems work in his organization and
• 1:07
  organizational AI usage. So
• 1:09
  a lot of good topics touched on today.
• 1:11
  And so let's go ahead
• 1:12
  and get into the episode.
• 1:17
  As I was just saying to you, everybody
• 1:18
  loves hearing from you and
• 1:19
  loves hearing what you're up to.
• 1:21
  So the first question I want to talk
• 1:23
  about is what you mentioned
• 1:24
  to me at the Meta Performance
• 1:26
  Marketing Summit and showing me which was
• 1:28
  how many of these landing
• 1:29
  pages that you've built,
• 1:31
  which you admitted was not a great use of
• 1:33
  your time as the CEO.
• 1:34
  But you said it was legit,
• 1:36
  like had been huge in terms of it. One
• 1:38
  question I had is, you're building all
• 1:40
  these and you're going
• 1:41
  through them. How are you keeping track
• 1:42
  of the tests of all of
• 1:44
  this, like in a way that makes
• 1:45
  sense? Because you're doing so many
• 1:46
  iterations. I mean, everybody does this a
• 1:48
  little bit differently.
• 1:49
  But since you're building them manually,
• 1:50
  it's not like through a
• 1:51
  tool or something else,
• 1:52
  or maybe you built a testing tool
• 1:53
  yourself. I'm just curious.
• 1:55
  Yeah, I built a roadmap tool. That was
• 1:56
  like actually one of
• 1:57
  the first things I did,
• 1:58
  you know, and I don't follow it. I don't
• 2:01
  organize it. I just hired somebody new,
• 2:04
  which we'll talk about to like, run this
• 2:06
  whole process, you know,
• 2:07
  for me, because I'm just gonna
• 2:08
  like give her everything that I built and
• 2:10
  do. But I'm just like
• 2:11
  terrible at taking the time to do it.
• 2:13
  But theoretically, it would be in there.
• 2:14
  Otherwise, it's just like
• 2:15
  an intelligent stuff like
• 2:17
  that, you know, but yeah, built a lot of
• 2:19
  pages, a little combo,
• 2:20
  built new landers, and then just
• 2:21
  doing a ton of, you know,
• 2:23
  website CRO with Cloud as well.
• 2:25
  So I'll just one more follow up question
• 2:26
  on this. You've gone through
• 2:28
  and how have you decided which
• 2:31
  elements because a lot of the pages that
• 2:33
  I saw that you had built,
• 2:34
  and maybe you can send us a
• 2:35
  couple of these, we can include them in
• 2:37
  the show notes that we could
• 2:37
  talk about these are things
• 2:38
  that you built. But like, you know, they
• 2:40
  have obviously, it's the
• 2:41
  PDP, and then it has a whole
• 2:42
  bunch of features under it, like it has
• 2:44
  different, you know, skin tone variation
• 2:47
  looks at it. It has,
• 2:48
  you know, all these different features
• 2:49
  that I'm sure you've
• 2:50
  seen have converted before.
• 2:52
  How do you decide which to put where? Or
• 2:54
  is it, you know, based
• 2:55
  on what you know of like,
• 2:56
  what's been the best performing combos?
• 2:58
  Yeah, that's a good question. So those
• 2:59
  were actually sections, I
• 3:01
  can just walk you through
• 3:02
  what the process was. Those were our PDP
• 3:04
  sections that I built for our
• 3:05
  PDP for a new launch that we
• 3:06
  had. And so for me, the most foundational
• 3:09
  thing like, is not even
• 3:11
  like I've said this in a lot of
• 3:12
  tweets, like, and I'm like, right now,
• 3:14
  and I hate like boasting or
• 3:16
  bragging, but I'm just excited
• 3:16
  about this. Like, I'm five for five on
• 3:18
  intelligence tests that I've ran end to
• 3:20
  end in Claude. And by
• 3:21
  end to end, I mean, like starting with
• 3:23
  like the insights and the
• 3:24
  research and the strategy.
• 3:25
  And so what's important to me, and what's
• 3:27
  what I think is cool is not
• 3:28
  just that I can design and
• 3:29
  dev stuff, but it's it's because you can
• 3:32
  do that so fast, I actually
• 3:33
  think the strategy is more
• 3:35
  important. You only have a certain
• 3:36
  limited number of testing slots that you
• 3:38
  can take up. And so being
• 3:40
  able to do those to a much higher degree,
• 3:43
  and I think I'm really
• 3:44
  on to something because
• 3:44
  prior to this, like, we have never had
• 3:47
  five winning tests in a
• 3:48
  row. Usually it's like one
• 3:49
  out of 10, two out of 10, you know, and
• 3:51
  these were like some pretty big wins,
• 3:52
  like five 10% lifts.
• 3:55
  And why I think that that has happened
• 3:57
  10% lift on CVR. Yeah,
• 4:00
  CVR. That's insane. I mean,
• 4:02
  relatively speaking, like that's crazy.
• 4:04
  Yeah. And and the fact that
• 4:05
  it's multiple in a row. And the
• 4:07
  reason why is this zero skill that I
• 4:10
  trained and I just like, I didn't know
• 4:13
  where this was going to
• 4:13
  go is just one Sunday, I was just like,
• 4:15
  hey, I feel like Claude
• 4:16
  could do a pretty good job with
• 4:17
  CRO and obviously tweeted about it. And
• 4:19
  so I just started like feeding Claude
• 4:21
  everything I had access
• 4:22
  to. And so gave it intelligence, MTP, I
• 4:25
  gave it a Slack channel of all of, you
• 4:29
  know, just how to go
• 4:30
  through that. We gave it spreadsheets, I
• 4:33
  gave it transcripts of
• 4:36
  looms from a past CRO agency,
• 4:38
  I gave it like, you know, I always have
• 4:40
  granola notes on when I'm
• 4:41
  recording my podcast. So I've
• 4:42
  had Dylan Ander on twice that, you know,
• 4:44
  CEO of heat map gave it
• 4:45
  those notes. I gave it, you know,
• 4:46
  he later sent me his book as a PDF, like,
• 4:48
  just kind of like built
• 4:50
  these knowledge bases and trained
• 4:51
  it. And then and what I'll do is I'll
• 4:53
  just go to Claude. I'm like,
• 4:54
  hey, this is what I want to do.
• 4:55
  Here's what I have. What else do you
• 4:56
  need? You know, and I'm like,
• 4:58
  well, what do you need for UX
• 4:59
  best practices? And Claude tells me, hey,
• 5:02
  Baymard. I'm like, all
• 5:03
  right, cool. Go scan the web,
• 5:04
  find that stuff. So kind of has like all
• 5:06
  of the CRO frameworks
• 5:07
  and right. For example,
• 5:08
  Claude is trained on when coming up with
• 5:10
  a test. It's trained on the ice
• 5:12
  framework, right? Everyone's
• 5:13
  familiar with the ice framework. So it's
• 5:14
  trained on that. It's trained
• 5:16
  on, we look at highest traffic
• 5:19
  pages above the fold, like that's going
• 5:21
  to be the higher leverage areas. You
• 5:23
  know, there's just like
• 5:24
  a certain testing framework that we've
• 5:26
  picked up over the years from
• 5:27
  working at different agencies
• 5:28
  and stuff. It's trained on that. So it
• 5:30
  has the CRO basics. Then it has our data.
• 5:32
  So it has at the time,
• 5:33
  it was IntelliGems, MCP. Now it has GA4.
• 5:36
  We have the heat map API.
• 5:38
  So it's pulling in data from
• 5:39
  multiple sources and able to, you know,
• 5:42
  analyze that data and have
• 5:43
  whatever it is, you know,
• 5:45
  dashboards. And then where I think is,
• 5:47
  you know, the coolest one
• 5:47
  is the qualitative stuff.
• 5:49
  I have this one really big customer
• 5:50
  intelligence markdown file. And it has,
• 5:53
  it analyzes a lot of
• 5:56
  subjective insights. So it has our Junip
• 5:58
  MCP reviews. It has kind of what else?
• 6:02
  RichPanel CX Tickets API. It has
• 6:05
  OuterSignal MCP. It has ListenLabs, which
• 6:09
  is like a customer survey
• 6:11
  focus group thing we do for prospects. We
• 6:14
  can kind of build panels of people and
• 6:16
  send surveys to them
• 6:18
  and do customer interviews. And then we
• 6:19
  use Typeform as well
• 6:21
  for a lot of like internal
• 6:22
  customer advisory stuff. And so Claude
• 6:24
  analyzes all of that. And I have skills.
• 6:28
  So any of these skills,
• 6:29
  like CRO skill, landing page building
• 6:31
  skills, anytime they run,
• 6:32
  they actually refresh those
• 6:34
  APIs, hold new fresh info into that
• 6:36
  customer intelligence markdown file.
• 6:38
  And so the reason that's important is to
• 6:40
  your, is a long way of saying to your
• 6:42
  question of where do
• 6:43
  you even start? How do you even know what
• 6:44
  to test? Is like, I just asked Claude,
• 6:47
  like it'll analyze all
• 6:49
  of this. And so I think that's been to me
• 6:52
  the biggest factor. But
• 6:53
  then I can go and design four
• 6:56
  variations of each test, stacking
• 6:58
  variables and be able to
• 7:00
  design and develop features.
• 7:02
  So the one you're asking me about, Claude
• 7:04
  knows this markdown file
• 7:06
  analysis thing, knows that
• 7:08
  shade matching friction is the number
• 7:10
  one, you know, thing. I
• 7:11
  then have a skill, I just,
• 7:13
  you do something once and you build a
• 7:14
  skill that does competitive
• 7:16
  intelligence and looks at eight
• 7:17
  different websites and sees how they
• 7:19
  handle this friction, what
• 7:20
  they do on their PDB sections,
• 7:22
  builds a mockup of it. Once we get it in
• 7:24
  a good place, you know,
• 7:25
  gets that over into Shopify.
• 7:26
  So that's kind of like a little bit of
• 7:28
  the, of the workflow
• 7:29
  that has been really cool.
• 7:31
  Yeah, that's awesome, man. I guess I have
• 7:34
  sort of a two follow ups on that one.
• 7:37
  The first question is, you start running
• 7:40
  this process, at what
• 7:41
  point are you as the user,
• 7:44
  if you will, getting involved? Is it not
• 7:46
  until like you, the
• 7:46
  page is fully developed,
• 7:48
  and then you start providing feedback on
• 7:50
  the page development and say
• 7:52
  like, hey, let's change this
• 7:53
  block, let's change that block? Or do you
• 7:54
  have sort of checkpoints
• 7:55
  throughout your markdown files
• 7:57
  to say like, hey, Cody, like, this is
• 7:58
  what I'm thinking? What
• 8:00
  are your thoughts on that?
• 8:01
  Yeah, that's a good question. Every,
• 8:02
  every phase I'll get involved.
• 8:04
  Okay.
• 8:04
  Usually there's like, you know, there's a
• 8:06
  plan. So usually I'll,
• 8:07
  I'll, I'll do a brief first.
• 8:08
  I'll be like, like, let's say we're
• 8:10
  trying to come up with a
• 8:10
  zero test or I try to build
• 8:11
  these sections. There's some component of
• 8:14
  a brief where I'm like,
• 8:15
  hey, build me an HTML file,
• 8:16
  like pulling from these sources. And
• 8:18
  obviously I'll build a
• 8:18
  skill for building this brief.
• 8:19
  And part of it, I like to see, because
• 8:22
  everything has got to be human in the
• 8:23
  loop. Like you can't
• 8:24
  just do like AI swap for it. I like to
• 8:26
  see the brief, which has
• 8:27
  the insights, the data,
• 8:28
  the qual and quant, the test plan. Like I
• 8:31
  just try to think of it, of
• 8:32
  like, what would a world-class
• 8:34
  zero or digital product function be
• 8:37
  without AI, you know? And like, what
• 8:39
  would I want? I would want
• 8:40
  these briefs I'd want based on this user
• 8:42
  research, based on this
• 8:43
  data, based on competitors.
• 8:45
  Like, so I just had Claude doing that. So
• 8:48
  I'll review that, give any
• 8:49
  feedback, give any notes,
• 8:51
  kick it off. Then we go to the next
• 8:52
  phase, which is, you know,
• 8:53
  wireframes, mockups, you know,
• 8:55
  and it'll do some mockups. I'll give
• 8:57
  feedback on that. And then once we get
• 8:59
  those into a good spot,
• 9:01
  there's probably multiple rounds of
• 9:02
  visions. We'll then take
• 9:04
  that into Shopify, you know,
• 9:06
  have a whole, whole workflow for building
• 9:09
  out those Shopify pages from Claude.
• 9:11
  That's where there's a ton of back and
• 9:12
  forth. There's no like,
• 9:13
  I want to be super clear.
• 9:14
  Like, this is, this is a really poor use
• 9:17
  of my time. I do it at
• 9:18
  night because it's fun. I do it
• 9:20
  on weekends for fun. Like there's
• 9:21
  multiple, multiple,
• 9:23
  multiple rounds of revisions. This
• 9:24
  is not like a one shot, you know? It's
• 9:26
  like, I can one shot a
• 9:27
  page 80% good. That final 20%,
• 9:30
  that's taking a long,
• 9:32
  long period of time.
• 9:33
  Yeah, I certainly hear that. And like,
• 9:35
  sometimes getting even
• 9:36
  the page into the Shopify
• 9:37
  pages, like, and like ensuring the API is
• 9:39
  like firing correctly can
• 9:40
  be such a pain in the ass.
• 9:41
  The other question I have is, it sounds
• 9:43
  like you're using like a
• 9:44
  lot of social listening tools,
• 9:45
  you know, to help build these, build
• 9:48
  these pages. And it's
• 9:49
  almost like a chicken or the egg,
• 9:51
  you know, situation in the sense that
• 9:53
  like, are you thinking like,
• 9:54
  I'm going to build this page
• 9:55
  first, and then I'm going to build ads
• 9:56
  around it? Or like, I have these ads now
• 9:59
  that I are performing
• 10:00
  well. So I'm going to build the page for
• 10:02
  those ads, and then subsequently build,
• 10:04
  you know, additional ads
• 10:04
  on that sort of angle or persona. That's
• 10:06
  a good question. Almost
• 10:08
  like thinking of them together.
• 10:10
  So some of them are, you know, for these,
• 10:12
  these, for these pages is,
• 10:14
  so I would I have for the
• 10:16
  landing pages, I have probably five or
• 10:17
  six templates. So those
• 10:19
  are like different pages,
• 10:20
  and then like in Shopify, their
• 10:21
  templates. So call it a listicle, we call
• 10:23
  it a performance LP,
• 10:24
  but it's more just like, you know, your
• 10:25
  standard LP with a buy box,
• 10:27
  we have like an enhanced PDP,
• 10:29
  that's like looks like a PDP, but
• 10:31
  conversion optimized. We have like a 10
• 10:32
  reasons why advertorial
• 10:34
  that I wrote from Zach Stuck have, you
• 10:37
  know, like, so there's
• 10:39
  like four or five of those.
• 10:40
  Nice. And, and then those are templates,
• 10:42
  Shopify can duplicate
• 10:43
  them, you can put copy of them.
• 10:45
  So those are just like the building
• 10:46
  blocks. And then like my goal, and want
• 10:49
  to chat with you about
• 10:49
  this, because like, we're not I'm not
• 10:50
  there on the outside yet.
• 10:51
  Like, I want to just like, boom,
• 10:53
  have an idea and be able to build a
• 10:54
  funnel. And to me, a funnel
• 10:55
  is like, a toolkit of ads,
• 10:57
  let's call it like 12 plus ads,
• 11:00
  diversified creative, you
• 11:02
  know, statics, like, do like
• 11:03
  a minimum viable test and have that go to
• 11:05
  like, an adjacent
• 11:06
  landing page. That makes sense.
• 11:07
  Cool. No, I like that a lot. Honestly,
• 11:10
  like using social listening
• 11:11
  insights to do CRO to then to
• 11:13
  do ads, because I've been doing the
• 11:14
  opposite, and just like using insights
• 11:16
  from ads to generate
• 11:17
  the landing pages. But then at that
• 11:18
  point, that angle may
• 11:19
  have fatigued out. So like,
• 11:20
  I feel like for even discovering net new
• 11:22
  angles, like that
• 11:23
  approach makes a ton of sense.
• 11:24
  The other question I have is, and then we
• 11:27
  can, I'm sorry, move on, Andrew.
• 11:30
  The good shit. This is the real shit
• 11:33
  right here. We're good.
• 11:34
  People love sharing like, you know, CRO
• 11:36
  agents and shit. And like, I
• 11:37
  have a CRO agent that's like,
• 11:39
  sorry, CRO QC agent. And like, I think
• 11:42
  it's pretty good. And folks
• 11:43
  ask me all the time, like,
• 11:45
  you know, can I just like have a bunch of
• 11:46
  QC agents to go through
• 11:47
  this? Like, why do I have
• 11:48
  to be involved at all? And maybe you can
• 11:51
  sort of like talk to that?
• 11:52
  And like, how just like having
• 11:54
  QC agents isn't going to solve your
• 11:57
  problem? And like, why
• 11:58
  having human in the loops? And
• 12:00
  ultimately, like
• 12:01
  having taste matters here?
• 12:03
  Yeah, I mean, I don't even know what
• 12:04
  like, so how are people
• 12:05
  using QC agents? Like,
• 12:06
  what components of it?
• 12:07
  Yeah. So for example, like your page
• 12:09
  should have like social proof,
• 12:11
  and a block that has all of, you know,
• 12:14
  the press articles
• 12:15
  that mentioned you like,
• 12:16
  and if you don't have these like 20
• 12:17
  things, like your page won't perform or
• 12:19
  whatever it is, right?
• 12:22
  Yeah, I don't know. I think everything is
• 12:24
  the way I think about
• 12:25
  it. Like, it's just,
• 12:26
  it's the same way that you would manage
• 12:27
  people. Like I was actually doing a
• 12:30
  unrelated thing earlier. And yeah, I was
• 12:32
  like whisper flowing and
• 12:33
  giving feedback. And like,
• 12:34
  if somebody walked into the room I was
• 12:36
  in, like, they would think I was like on
• 12:37
  a call with a person.
• 12:38
  Like, I was very much just like talking
• 12:40
  and be like, no, this is like the
• 12:41
  framework we need for
• 12:42
  this was like redoing our KPIs. And like,
• 12:44
  I just think about it as
• 12:45
  managing people. And it's the
• 12:47
  same way. Like, you can also choose to
• 12:49
  take junior mid-level
• 12:50
  employees and not give them a lot of
• 12:53
  clarity and direction and not be in the
• 12:55
  workflow, you know, and see
• 12:57
  what you get. Or you can do a
• 12:59
  good job of giving them clear direction
• 13:02
  up front, giving them proper
• 13:03
  context, and helping them tweak
• 13:05
  along the way. And so I just think it's
• 13:07
  the same thing where, like,
• 13:08
  I'm not trying to completely remove
• 13:10
  myself. Like, that's where you get slow.
• 13:11
  And at the very least,
• 13:14
  you have to spend a ton of time. Like, I
• 13:16
  put an ungodly amount of
• 13:18
  time into building our design
• 13:19
  system. Like a painstakingly amount of
• 13:22
  time. And now I have this giant, you
• 13:25
  know, GitHub markdown
• 13:26
  file of our design system, which just
• 13:27
  took so long. And so now I can one shot a
• 13:30
  page, have it be 80%
• 13:31
  there. But there's, it's still, it's not
• 13:33
  perfect. AI is deterministic, and it
• 13:35
  still messes up some
• 13:36
  stuff. So yeah, you, you have to be in
• 13:38
  the loop. I think building skills and
• 13:40
  context layers are the
• 13:41
  highest leverage things that you can do
• 13:43
  to allow you to be more hands
• 13:45
  off later. But you, you, it's
• 13:47
  not just like a, let me set up this
• 13:48
  thing. I just think that's,
• 13:50
  that's just all that like, AI
• 13:52
  engagement bait on Twitter. Yeah. Oh, I
• 13:54
  built this thing that's
• 13:55
  launching 1000 ads. And I've made,
• 13:57
  that's where I've made the mistake on,
• 14:00
  like, on Ciro stuff, landing
• 14:01
  page stuff. I've been much more
• 14:03
  detailed. When I've tried to do ads, I'm
• 14:04
  like, Oh, man, I'd love to
• 14:05
  launch 1000 ads. And so like,
• 14:06
  I'll get like a good reference, I'll get
• 14:08
  like a few ads. And then
• 14:09
  I'll try to like scale it. And
• 14:10
  it's just like, it's, it's just too much.
• 14:12
  And it like becomes slop, I feel like.
• 14:14
  Yeah. I mean, I completely agree. Like
• 14:16
  similar to you, I like just
• 14:17
  hired someone to sit in cloud
• 14:18
  code and generate ads. Like, because I
• 14:19
  tried to do it initially, like just
• 14:21
  generate me 100 ads,
• 14:22
  like here, it's social listening, you
• 14:23
  know, here's ad
• 14:24
  performance. And like, it's shit,
• 14:25
  you know, like the outputs. But anyway,
• 14:27
  sorry, Andrew, you know, take it on.
• 14:29
  So first of all, well, you know, again,
• 14:30
  you don't have to ever
• 14:31
  apologize to me. It's okay. If we go
• 14:32
  on a riff, it's it can be a 30 minute
• 14:34
  riff. That's fine. I think,
• 14:35
  you know, let's go back to
• 14:37
  this creative strategy thing just
• 14:38
  briefly, or creative overall.
• 14:40
  And Cody, you just alluded to
• 14:41
  it again. You know, I think that a lot
• 14:43
  this is what the biggest one that a lot
• 14:45
  of folks are trying to
• 14:46
  solve as it relates to AI that's like,
• 14:47
  okay, we're going to take over creative
• 14:49
  and we're gonna figure it
• 14:49
  out. You know, Will's putting out a ton
• 14:51
  of proposals on this, right? From self
• 14:53
  made of doing AI for
• 14:54
  creative for people. Two questions, and
• 14:58
  maybe you both can answer
• 14:59
  this question, which is,
• 15:02
  first, what is a system now or what
• 15:05
  systems need to be in place now to make
• 15:09
  higher converting creative?
• 15:11
  And if they're not already in place, and
• 15:14
  second question is, what are in six
• 15:17
  months, or at the end
• 15:18
  of the year, at the end of 2026? Like,
• 15:21
  what are where is creative
• 15:24
  workflow going to be? Because
• 15:25
  right now, it's so I feel like it's
• 15:27
  piecemeal. I feel like it's all over the
• 15:29
  place. And I feel like
• 15:30
  nobody, there's very few that have
• 15:32
  mastered the workflow aspect of this of
• 15:35
  the briefing to the ad,
• 15:37
  you know, to all of those pieces to bring
• 15:39
  the teams together. And
• 15:40
  obviously, like different tools
• 15:42
  are solving this to some degree, like HQ
• 15:43
  solving this, like to a
• 15:45
  degree, you know, but I'm curious of
• 15:47
  what you guys think and of like, what
• 15:49
  it's what it is now to be successful. And
• 15:51
  like, what's it going
• 15:52
  to be in six months? Because Cody, you've
• 15:55
  been really spot on with, with
• 15:56
  predictions, frankly,
• 15:57
  I did some research and the things you've
• 16:00
  said have largely become
• 16:01
  true. So I'm curious of what you
• 16:03
  think.
• 16:04
  Also, I have not like mastered, like, I'm
• 16:07
  like a laser focus person.
• 16:09
  And so all my like effort has
• 16:11
  really gone on like these CRO stuff. So
• 16:12
  like, Will is way ahead of me on like,
• 16:15
  image gen and stuff like
• 16:16
  that. Like, I want to get a lot better
• 16:17
  there. I think it is, I
• 16:20
  think it's interesting. Like,
• 16:23
  I wouldn't want to be a software like an
• 16:24
  icon or like, I think
• 16:26
  something launched last week,
• 16:27
  like Tempo. Like, I think a lot of that
• 16:29
  is very commoditized by
• 16:30
  people who can build stuff. And
• 16:31
  I've seen a lot of what Will has built is
• 16:33
  like, I think whether
• 16:34
  you're doing it, you know,
• 16:35
  directly in a Nano Banana or ChatBee GPT,
• 16:37
  or like you're doing it from Cloud Code,
• 16:39
  pinging API. Like, I like that. I
• 16:42
  consider Cloud like the white collar
• 16:45
  operating system that like,
• 16:46
  you never have to leave Cloud and you can
• 16:48
  just like do everything
• 16:48
  from there. Because think about
• 16:49
  we have, we have Motion MCP pulling into
• 16:51
  there. We have our Polar MCP.
• 16:53
  We have, you know, Meta MCP,
• 16:55
  like, you just have all of your sources.
• 16:58
  We have shop, like, so I
• 16:59
  just think getting all of that
• 17:00
  stuff in one place is incredibly
• 17:02
  valuable. And one of the things we're
• 17:04
  working on a ton internally is
• 17:05
  like, even if you don't know where it's
• 17:07
  good, like, and this is so
• 17:08
  hard, like getting people
• 17:10
  to change their workflows and behaviors,
• 17:11
  if they're used to using
• 17:13
  docs or sheets or stuff
• 17:14
  like that, and now asking them to do it
• 17:16
  and, and Cloud is like very different.
• 17:18
  But I just know that
• 17:19
  it's going to be valuable having
• 17:21
  everything connected in a
• 17:22
  context layer and talking to each
• 17:23
  other. And I don't even know that I know
• 17:26
  what the value is going to be
• 17:26
  yet. But I just think that's a
• 17:28
  huge part of it is the context is like,
• 17:30
  you know, like, the models are not
• 17:31
  everything like the models
• 17:32
  are great, but like, they're all good.
• 17:34
  The models are good. Actually
• 17:35
  having your context is the most
• 17:36
  important thing. So I think that's
• 17:38
  extremely important. And
• 17:39
  the more you can document and
• 17:40
  orchestrate that stuff. But like, Meta is
• 17:45
  probably going to do it.
• 17:46
  Meta is going to do a lot of it.
• 17:49
  That's what I think is going to happen.
• 17:51
  Yeah, yeah. I mean, to what you said, I
• 17:53
  think is true. And you
• 17:54
  know, we have now I just looked,
• 17:55
  we have like 70 members that are
• 17:57
  connected to the Foxwell MCP, like our
• 17:59
  knowledge base. And,
• 18:01
  you know, that's every call transcript
• 18:02
  we've ever had. That's
• 18:04
  everything in Slack. That's every
• 18:05
  SOP. And like bringing that all together,
• 18:08
  what you said, even about
• 18:09
  the landing pages is absolutely
• 18:10
  absurd. Like how, I mean, it's how
• 18:13
  valuable that is in terms of being able
• 18:15
  to bring an answer forth
• 18:16
  of what's really going to work. So I
• 18:17
  think that's, I think that's a good
• 18:18
  shout. But yes, I do agree
• 18:20
  that Meta is going to do a certain
• 18:21
  percentage of this. And yes, I think
• 18:24
  tools are tough. I think that
• 18:26
  the challenges is really sitting down and
• 18:28
  looking at like, where are the
• 18:29
  inefficiencies in your own
• 18:30
  workflow? And how are you trying to solve
• 18:32
  it? I don't know, Will,
• 18:32
  what do you think about this?
• 18:33
  Like, what are people doing? You know,
• 18:35
  what can people do better
• 18:36
  now? And then what does it look
• 18:37
  like in the future? Yeah, for sure. I
• 18:39
  think there's sort of like
• 18:40
  three pieces of the equation,
• 18:42
  as it at least pertains to ads and, you
• 18:45
  know, by association
• 18:47
  landing pages. And I definitely am
• 18:49
  a broken record with this. But it's like,
• 18:50
  the first thing is like a
• 18:51
  gap analysis and saying like,
• 18:53
  you compare and juxtapose yourself to
• 18:55
  other folks in your
• 18:57
  vertical, like what sort of formats,
• 18:59
  personas, angles, emotions, are they
• 19:01
  testing that you're not? And sort of
• 19:03
  index all of your ads have,
• 19:05
  you know, Claude's on it 4.6, just like,
• 19:07
  go through all of your ads, you know,
• 19:10
  look at your personas,
• 19:11
  angles, emotions, and say like, this is
• 19:14
  how I compare to my competitors. It's
• 19:16
  sort of like piece one.
• 19:17
  Piece two, of course, it's like a time
• 19:18
  series analysis, where you
• 19:20
  just like use a meta API,
• 19:22
  pull in all of the ads you've ever run
• 19:23
  for the past, like, you
• 19:24
  know, 365 days or five years or
• 19:26
  whatever, tag them all with the same
• 19:28
  taxonomy, where you have like persona,
• 19:31
  angle, emotion format.
• 19:33
  And so you can sort of say like, well,
• 19:34
  you know what, I turned
• 19:35
  this ad off a year ago,
• 19:36
  but like, I haven't run that, you know,
• 19:39
  persona angle format since
• 19:40
  then, right? So you're able to
• 19:42
  sort of resurface these old winners,
• 19:44
  rather than like lose all these
• 19:45
  compounded learnings. And I think
• 19:47
  that's where a lot of folks get wrong is
• 19:48
  that they have had
• 19:50
  probably so many successful,
• 19:52
  successful equations in the past that
• 19:54
  have just like been lost to
• 19:56
  time, right? It's such a pain
• 19:57
  to go like, go look through your meta ad
• 19:59
  library and see like what
• 20:00
  has worked in the past. And the
• 20:01
  third piece is the social listening,
• 20:03
  right? Like perpetually
• 20:04
  scraping your reviews, subreddit
• 20:06
  comments, Facebook comments. And then if
• 20:09
  you add those all together,
• 20:10
  you know, give them proper
• 20:11
  weights. You have a really good sense of
• 20:15
  like, what I need to be
• 20:16
  creating, right? Like here are my
• 20:18
  holes from the gap analysis. Here's
• 20:19
  what's worked in the past. And then like,
• 20:21
  here's what consumers
• 20:22
  are saying about myself now. The cherry
• 20:25
  on top is using, you
• 20:27
  know, Reddit and TikTok and,
• 20:29
  and Google Trends to see is there
• 20:31
  anything trending right now that I can
• 20:34
  like quickly spin up ads for
• 20:35
  and like take advantage of that. So like
• 20:38
  if you can spin up ads within
• 20:40
  like 24, 48 hours, I know it's,
• 20:41
  it's hard, but it's actually not that
• 20:44
  hard. If you're able to do
• 20:46
  that, and, you know, sort of add
• 20:48
  that as like the, you know, as the gravy,
• 20:51
  then like you're going to be
• 20:52
  in a really, really good spot.
• 20:53
  So that is like sort of my equation and
• 20:55
  how I think about things.
• 20:56
  And it's like definitely been
• 20:57
  pretty successful.
• 20:58
  Quality is important, but like obviously
• 21:00
  the, the, the strategy is
• 21:02
  important and getting the right,
• 21:04
  you know, the writing sites, the right
• 21:06
  data, right? But also,
• 21:07
  yeah, like just, just speed to
• 21:08
  launching fun. Like I'm making that a
• 21:10
  core KPI of my growth team
• 21:11
  is just like, we just need to
• 21:13
  have a, we're going to say 24 hour funnel
• 21:15
  launch time. Cause we, we,
• 21:17
  we review data Monday. We have
• 21:19
  a deck, we're trying to automate that
• 21:20
  deck, but it's just like, all right,
• 21:21
  cool. Here's an idea.
• 21:22
  We have this angle is taken off. Here's
• 21:24
  an iteration on it. Like
• 21:25
  that used to be, and, you know,
• 21:28
  pre AI, I'm sure some teams were better
• 21:30
  than others, but like that
• 21:31
  just takes, it just had taken
• 21:32
  too long where you get a idea and it's
• 21:34
  like, Oh, we'll brief that. We'll put it
• 21:36
  in Asana. We'll go to design
• 21:38
  revisions. We'll build a landing page.
• 21:40
  It's like a month later,
• 21:41
  you're like, you missed the
• 21:42
  opportunity. So it's like, no, let's get
• 21:43
  that live tomorrow. Like
• 21:44
  that's, that's it. And even if
• 21:46
  like, that's just a V1 and then you can
• 21:48
  go and then brief UGC
• 21:49
  creators on that and like scale that
• 21:51
  up. I just think being able to get some
• 21:53
  quick insights on
• 21:53
  that is, is so valuable.
• 21:55
  Yeah, completely. I mean, that's
• 21:56
  something that you're doing scalability
• 21:57
  school podcast with Zach
• 21:58
  Suck. It's something I've been within his
• 22:00
  teams, right. It's just
• 22:01
  absolutely so rapid. Like it's,
• 22:04
  it's, it's, it, the speed is, is a huge
• 22:06
  part of the component of
• 22:07
  what makes it successful.
• 22:08
  Can I tell you what some of this comes
• 22:09
  from? I had a landing
• 22:11
  page that somebody sent me
• 22:12
  and it was a new template I had seen
• 22:14
  before. And I sent it to a
• 22:15
  few people and I sent it to Zach
• 22:16
  and Zach said, I'll tell you how it's
• 22:18
  doing in 24 hours. And I, it
• 22:19
  haunted me. I was so upset.
• 22:22
  I would know, but I thought about it for
• 22:23
  like, like literally
• 22:24
  until now. And I was like,
• 22:26
  that's what it takes to be successful.
• 22:28
  And I, if I asked my team to launch this,
• 22:31
  it would be an embarrassing length of
• 22:33
  time to get it along.
• 22:35
  Yeah. I mean, I think so much of it is,
• 22:37
  is going through and yes,
• 22:39
  assessing the systems you have
• 22:40
  now and, and, and completely adjusting
• 22:43
  all of this. I mean,
• 22:44
  it's, it, it's a different,
• 22:46
  totally different ball game as it relates
• 22:47
  to digital marketing within, I think so
• 22:50
  much of the AI stuff
• 22:51
  is around ads and it's focused on ads,
• 22:53
  but it's not, you know, we
• 22:54
  all know that launching ads that
• 22:56
  match to landers in the persona driven
• 22:58
  funnel is like ultimately what helps
• 23:00
  scale a lot of these
• 23:01
  big brands or, you know, like yours and,
• 23:04
  and others. And it helps
• 23:05
  even small, you know, brands
• 23:07
  scale up. Like it's a bigger part of it.
• 23:10
  And, uh, and there was way
• 23:12
  too many tools in the middle and
• 23:13
  it doesn't need to be that way anymore.
• 23:15
  And I think if you, tools and
• 23:16
  people, too many people in the
• 23:18
  middle too, that's the problem. Yeah.
• 23:20
  Yeah. That as well. Like
• 23:21
  you have like, you have like,
• 23:23
  you have to brief a copywriter and you
• 23:25
  got to brief a designer and
• 23:26
  you have no idea what their
• 23:27
  priorities are and what's on their
• 23:28
  workload. And then you have too many
• 23:30
  approvals versus like,
• 23:32
  again, and I'm not doing this outside of
• 23:34
  at night, but let's say it's
• 23:35
  one person. Let's say it's me.
• 23:36
  It's like, all right, cool. I have an
• 23:37
  idea for an ad. I'm going to
• 23:38
  go make some ads. I have an idea
• 23:40
  for landing page. I'm going to make a
• 23:41
  landing page. Like that's so much of
• 23:42
  actually what the value is,
• 23:43
  is like, you don't have to rely on other
• 23:45
  people to, to actually produce stuff.
• 23:47
  Yeah. I agree with you. I also think that
• 23:49
  there's not enough in terms of employee,
• 23:52
  employee alignment on this. There's not
• 23:54
  enough incentive for them to be faster.
• 23:56
  Right. So like, this is something that
• 24:00
  I've been working with a couple of
• 24:01
  agencies on that are in
• 24:02
  this kind of scaling phase. And that's
• 24:04
  what we're working on
• 24:05
  with them is basically like,
• 24:07
  how do we design systems that when the
• 24:08
  employee saves 40% of time,
• 24:11
  taking Cody, what you said
• 24:12
  before, I think you said, if you were
• 24:13
  starting an agency today, one of the
• 24:15
  three things you would do
• 24:16
  is an AI automation, the service for
• 24:18
  brands, and you audit,
• 24:19
  automate, and take a percentage of the
• 24:21
  savings. And that's essentially what I
• 24:23
  have gone and said to these folks.
• 24:25
  That was an old tweet.
• 24:25
  I forgot about that one.
• 24:27
  We're going, we went deep cuts on our
• 24:29
  research on this, but like...
• 24:30
  I think I said that. I think I said like
• 24:32
  whitelisting as a service.
• 24:33
  Yeah, of course.
• 24:34
  Whitelisting is incredible.
• 24:35
  And that's one other. There was one other
• 24:37
  one on the tweet. I forgot one.
• 24:38
  There are definitely some Foxhole founder
• 24:39
  members that are
• 24:40
  functionally whitelisting
• 24:41
  agencies now. That's like pretty much...
• 24:43
  No, no. I meant like, I meant like a
• 24:45
  white, like ads whitelisting.
• 24:46
  Oh, ads whitelisting. I got you.
• 24:48
  Yeah, that was...
• 24:48
  Also good. Sorry. Sorry.
• 24:50
  Also good business. But no, like, so my
• 24:52
  point is, I think a lot of
• 24:53
  that is employee incentive
• 24:55
  when you're sort of, okay, look, if
• 24:57
  you're going to do this, we're going to
• 24:57
  save X amount of time.
• 24:58
  Let's put that into dollars and that's
• 24:59
  going to be a bonus for you, right?
• 25:01
  That's a big part of it. So when you're
• 25:04
  utilizing, you know, Claude
• 25:08
  and you're building all this,
• 25:11
  like, don't you ever run out of storage
• 25:13
  and all this? Like,
• 25:14
  tokens? Like, how do you
• 25:16
  functionally do this? I'm
• 25:17
  just curious. Because from...
• 25:19
  Yeah.
• 25:19
  Yeah. Because from my end, like, I'm
• 25:21
  hitting the wall. I feel like
• 25:22
  a lot of people are. I don't
• 25:23
  know. I'm just curious of your take.
• 25:24
  I'm on the... So I have my whole team on,
• 25:26
  you know, Claude
• 25:27
  Enterprise plan. Most people are on
• 25:29
  pro power users either request extra
• 25:32
  usage and, you know, we
• 25:33
  approve it or are, you know,
• 25:35
  upgraded on like a max plan, which is I
• 25:37
  think like 100 bucks a month.
• 25:38
  If they're proving that it's,
• 25:40
  you know, worthwhile. I'm on a personal
• 25:41
  plan, which is a $200 a
• 25:43
  month max plan. I would say...
• 25:44
  And I'm like, what I would say as a power
• 25:46
  user, outside of that like
• 25:48
  one week period where they
• 25:49
  had all of those like usage limits that
• 25:51
  was like clearly bugs, I
• 25:53
  never hit. Like, I push it really
• 25:55
  hard. I use, you know, 4.7 extra high,
• 25:58
  you know, use it. I mean,
• 26:00
  again, I don't use it that much
• 26:02
  9 to 5 because I'm busy in meetings and
• 26:04
  shit. But mornings, nights, weekends,
• 26:06
  like, I almost never
• 26:07
  hit it. So I don't know. I don't know if
• 26:09
  I have problems with it.
• 26:10
  But I've thought about it.
• 26:11
  Like, I would... What would I pay
• 26:12
  somebody to run these CRO tests? Like,
• 26:15
  I'd pay thousands a month.
• 26:16
  So like, I'm paying 200 a month. But
• 26:17
  like, if I had to pay 1000
• 26:18
  a month, like, I think it's
• 26:20
  legitimately building that much value for
• 26:22
  us. And I've also... I mean, we've
• 26:23
  canceled. I just canceled
• 26:24
  three tools, you know, that'll save us a
• 26:27
  few thousand a month that
• 26:28
  not only are saving money,
• 26:31
  but they would actually
• 26:32
  be inhibiting our workflow.
• 26:33
  Yeah, yeah. One thing... So you talked
• 26:37
  about your creator in
• 26:39
  residence program and shipping
• 26:41
  daily content. And so I'm wondering,
• 26:43
  like, how's AI playing
• 26:44
  into this with that group?
• 26:46
  We're not there yet. We're not there yet.
• 26:48
  It's been a hard one
• 26:48
  to build and just find
• 26:50
  like the perfect people and team to lead
• 26:51
  it. We're doing stuff,
• 26:52
  but it's been a slow start.
• 26:54
  I think I'm... So as big of a AI bull as
• 26:57
  I am, I'm actually
• 26:58
  pretty bearish on AI content.
• 26:59
  Like, I think static ads are one thing
• 27:01
  and it's great. But
• 27:02
  like, when I see people talk
• 27:03
  about like AI UGC, I'm
• 27:04
  just like, you're so far off.
• 27:06
  No, no, no. It's not...
• 27:07
  Totally.
• 27:07
  And it's...
• 27:08
  I can't express it.
• 27:09
  Yeah.
• 27:10
  So it's like... And I've read some stuff
• 27:13
  that's pretty good on this
• 27:14
  lately. Like the ClickUp,
• 27:15
  you know, founder or CEO just did a, you
• 27:18
  know, layoffs and he did
• 27:18
  a really good tweet on it.
• 27:19
  And like, he had some really good points
• 27:21
  that he said in it. But I
• 27:22
  think part of it is like,
• 27:23
  you know, he was talking about
• 27:25
  essentially automating as much of the
• 27:26
  administrative stuff.
• 27:27
  So you can spend more time talking to
• 27:29
  customers, you know, and it's like,
• 27:31
  how can you automate as much of the
• 27:34
  operation side or the
• 27:36
  backend stuff or the analysis and
• 27:39
  gestion so you can spend more time being
• 27:40
  creative? Like, that's
• 27:41
  what I would want my creative
• 27:42
  strategist to do is like, almost just
• 27:45
  think, you know, where
• 27:46
  creative strategy used to be,
• 27:47
  let me poke into motion, let me build
• 27:49
  these reports. Like there
• 27:50
  was a lot of time spent in
• 27:51
  an ad account. Hopefully that can just be
• 27:53
  handed to somebody on a
• 27:55
  silver platter of here's exactly
• 27:57
  what the data is to last seven days. And
• 27:59
  here's the ad you should make.
• 28:01
  And then maybe that changes the role.
• 28:03
  Like everyone's talking
• 28:03
  about the role of the media
• 28:04
  buyer changing. Maybe that changes the
• 28:05
  role of the creative strategist.
• 28:07
  And because if that can be done by
• 28:09
  anybody, it's like, I
• 28:11
  actually just want to hand that
• 28:12
  to a creator who really gets social
• 28:14
  content. Like that's, I
• 28:15
  think, the change there.
• 28:16
  So that's my goal is we have some
• 28:18
  dashboards that just show,
• 28:20
  you know, our top ad account,
• 28:21
  but also like organic social trends and
• 28:24
  data, you know? And so hopefully people,
• 28:26
  this is my vision for it. We have the
• 28:28
  studio built out. If you're listening,
• 28:29
  I just need the person to run it. I
• 28:31
  literally, I have the,
• 28:32
  we're just trying to find people
• 28:33
  is like, have like airplayed, like on a
• 28:36
  big TV in our studio, like these
• 28:39
  different dashboards.
• 28:40
  It's like, hey, here are four videos and
• 28:42
  here's, here's top
• 28:43
  trending thing. Acne prone skin is
• 28:46
  trending on TikTok right now. Here are
• 28:47
  the top four viral videos.
• 28:49
  Like make a version of that.
• 28:50
  Like that would be my goal. So it's just
• 28:52
  like long answer to say, like not
• 28:54
  automating the creative,
• 28:55
  but automating all of the, the prior
• 28:57
  stuff, the project
• 28:58
  management, the insights,
• 28:59
  the stuff like that. And then the, you
• 29:01
  know, the, the post-production as well.
• 29:03
  Yeah. That makes sense.
• 29:04
  So you got more time to spend, spend on
• 29:05
  actually making stuff.
• 29:06
  Totally. I want to sort of go back to
• 29:08
  something we were talking about earlier.
• 29:11
  Cause, you know, similar to you, like I
• 29:13
  am like addicted to cloud code, right?
• 29:15
  Like it's, it, I think there are certain
• 29:17
  people that are just
• 29:18
  like find it, you know,
• 29:19
  impossible to pull yourself away from.
• 29:20
  How do you incentivize your
• 29:23
  team to sort of have that sort
• 29:25
  of same level of commitment? Right. And
• 29:29
  like, I, I've, you know, I've done
• 29:32
  certain things internally,
• 29:33
  but like, I'm, I'm curious, like for you,
• 29:35
  Cody, it probably
• 29:36
  takes maybe what, like two,
• 29:38
  three hours to do a landing page now at
• 29:40
  this point, if you were to hand it off to
• 29:41
  someone to your team,
• 29:42
  it'd probably take them with the same
• 29:44
  process for five days. So like, how do
• 29:46
  you sort of reconcile
• 29:47
  that and like, what are you sort of going
• 29:49
  and like, what are you
• 29:50
  going about doing with your
• 29:52
  team and sort of like to teach them or to
• 29:54
  like sort of hand things off?
• 29:56
  Yeah. It's a good question. First of all,
• 29:57
  it kills me cause it's so hard.
• 30:00
  It is, it is. And like, it's, it's
• 30:02
  actually, it's like
• 30:03
  unbearably hard and I'm, we're working
• 30:06
  through it right now. So I have a
• 30:07
  director of data and AI who
• 30:09
  kind of is leading all of our
• 30:11
  initiatives internally. So he's a, you
• 30:12
  know, data analyst by trade and he just
• 30:14
  was super passionate
• 30:15
  about AI. And so he's leading everybody.
• 30:17
  So one thing we did is we
• 30:18
  got everybody on a Claude plan
• 30:20
  and, you know, for a while he did weekly
• 30:22
  trainings. And so he'd teach
• 30:23
  them Claude co-work skills.
• 30:26
  He did a mandatory one on what, you know,
• 30:28
  a GitHub repository is,
• 30:29
  stuff like that. He did one,
• 30:30
  you know, this morning that has a small
• 30:32
  group of people on just
• 30:32
  like, Hey, here's, here's,
• 30:34
  I'm going to build something live, like
• 30:35
  who has something they want
• 30:36
  to build. So just like trying
• 30:37
  to get people excited, showing them how
• 30:38
  it can help. And I would
• 30:39
  say it definitely helped.
• 30:40
  Like everybody is using Claude daily,
• 30:42
  some people to different
• 30:43
  levels. I would say the
• 30:44
  highest level obviously is him. Like one
• 30:46
  of his goals, one of my
• 30:47
  goals for him was to get
• 30:48
  our data warehouse in Claude. We were
• 30:49
  working with Polar to do
• 30:50
  that. So anyone can, you know,
• 30:52
  get insights to that. And, and we're,
• 30:55
  he's building a lot.
• 30:56
  We're building a lot. So like
• 30:57
  he's done a ton. Obviously I have our
• 30:59
  demand planning team built
• 31:01
  a demand planning software
• 31:02
  that we're doing that in there. So like
• 31:04
  that is replacing an
• 31:05
  $80,000 a year software and doing
• 31:07
  it better actually, because it's just,
• 31:09
  it's better. So like
• 31:10
  that's on one level. And then
• 31:12
  I would say like, you know, it's like a
• 31:13
  bell curve. Most of the
• 31:14
  org is like, Oh, I'll, I'll,
• 31:16
  I'll use some skills. I'll, I'll try this
• 31:17
  stuff. And then there, you
• 31:19
  know, there's people that
• 31:19
  are kind of laggers. It's really hard.
• 31:21
  Like he's even noticing
• 31:22
  he's meeting with people kind
• 31:24
  of doing some like internal FDE work. And
• 31:26
  it's like, people are
• 31:27
  really resistant to change
• 31:29
  or really resistant to change their
• 31:31
  workflows. And like, it's
• 31:32
  probably multiple reasons.
• 31:33
  Like, first of all, they just don't know
• 31:35
  what it's capable of.
• 31:36
  Sometimes they're, you know,
• 31:37
  too busy. They worry that they're
• 31:39
  automating themselves out of
• 31:40
  a job. Like that's definitely
• 31:41
  a concern of people. And so it's really,
• 31:45
  really, really hard. But the, what I
• 31:47
  think we're noticing
• 31:48
  is, and the ClickUp guy talked about, it
• 31:49
  was like, you have to create
• 31:50
  disruption. Like I have one
• 31:52
  person who is doing a really good job of
• 31:55
  using it and she hasn't built stuff, but
• 31:56
  she's using it to do
• 31:57
  her workflows that like we've built stuff
• 31:59
  for her. And part of the
• 32:00
  reason is she's got an additional
• 32:01
  scope while losing people on her team.
• 32:04
  One is out of maternity
• 32:04
  leave. One person, you know,
• 32:06
  just left and it's taken time to
• 32:07
  backfill. So it's like, she's had no
• 32:08
  choice but to use it.
• 32:10
  But I think the people that are
• 32:11
  comfortable, it's like,
• 32:12
  well, why do I need to do that?
• 32:14
  And so like, you know, I, I built, I like
• 32:17
  pushing it as far as possible. I also
• 32:19
  don't have the time.
• 32:20
  Like, like maybe that's why it's like,
• 32:21
  like I don't have the time. So I'm like,
• 32:23
  oh, what can this do?
• 32:23
  Like, what can I, I need this info. What
• 32:25
  can I do? But like, I built
• 32:27
  five landing pages and by built,
• 32:29
  I mean, like we had the templates done,
• 32:30
  but I like populated with them with
• 32:32
  content without ever going
• 32:33
  in Shopify. Like I use Shopify CLI to do
• 32:35
  stuff you would normally do. Cause like,
• 32:37
  I'm not going to sit
• 32:37
  there and upload images. So like, I'm
• 32:39
  just like, don't have time too lazy to do
• 32:41
  that. But I just like
• 32:42
  pushing the boundaries and needles. And
• 32:44
  what I find are most people
• 32:46
  aren't doing that. They're just
• 32:47
  like comfortable doing their old
• 32:49
  workflow. So I don't know what the answer
• 32:51
  is. If anyone has, let me
• 32:52
  know. I'm curious what your guys
• 32:53
  thoughts. I just know that we have to
• 32:55
  disrupt it. And I think I tried to be
• 32:57
  a little bit more like, Hey guys, we're
• 32:59
  doing this. This is
• 32:59
  exciting. And now it's like, no,
• 33:00
  you fucking have to do this. If you want
• 33:02
  to work here anymore,
• 33:03
  you're just, you have to do this
• 33:05
  process. You have to put this information
• 33:07
  here. It's just like a new
• 33:08
  way of doing it. Cause that's
• 33:09
  like, I just have been, I have been
• 33:11
  unimpressed with, yeah, I don't know.
• 33:14
  People's like, like people
• 33:15
  think they're using Claude and, and
• 33:17
  they're using it to like 10% of its
• 33:19
  ability. I feel like.
• 33:20
  Yeah. I think a lot of it is, is one is,
• 33:23
  is one of the things you mentioned,
• 33:24
  right? Which is always,
• 33:25
  I think whenever you have employee stuff
• 33:26
  and you're trying to get
• 33:27
  them motivated, the number
• 33:28
  one thing to look at is what are they
• 33:29
  scared of? So they're scared of, you
• 33:31
  know, the things that
• 33:31
  they're automating themselves out of a
• 33:34
  job. So discussing and
• 33:35
  going and saying, look, like,
• 33:36
  this is not something that I tend, I am
• 33:38
  going to replace you for.
• 33:40
  I need you to know that,
• 33:41
  that I am doing this because I'm trying
• 33:43
  to make your life easier and
• 33:44
  to make the work that you're
• 33:45
  doing more efficient. And so like
• 33:46
  doubling down on that, I think also
• 33:48
  giving them a roadmap of what you
• 33:50
  hope it looks like, even if it's not
• 33:52
  going to be what it is. And
• 33:53
  that can be refined every like
• 33:55
  month and going through and saying like,
• 33:56
  this is what we're trying to
• 33:57
  do. These are the things that
• 33:58
  are in my head so that you have more
• 34:00
  transparency into it. So much
• 34:02
  of it as an employee, right? Is
• 34:03
  like guessing what are people, what's the
• 34:06
  leader trying to do? Why are
• 34:07
  they getting into this? And
• 34:08
  so the more that you can go and
• 34:09
  transparently talk about that, it's
• 34:10
  great. The third one we talked
• 34:12
  about is incentives, absolutely
• 34:13
  incentivizing people to say like, look,
• 34:15
  if, if, and when you're adopting
• 34:17
  this, you know, if you can prove to me X,
• 34:19
  Y, Z, that we're saving
• 34:20
  money, like there are financial
• 34:21
  incentives aligned with that, right? This
• 34:23
  is just like when they did
• 34:24
  when in, in, you know, lean,
• 34:26
  lean process, uh, manufacturing and
• 34:29
  everything else. Like when
• 34:30
  Toyota came up with Kaizen process
• 34:32
  and like, what was it? 80, 1980 or
• 34:34
  whatever. It's like, they, they gave
• 34:36
  people a proportional
• 34:37
  amount of money based on what percentage
• 34:39
  of time it saved each employee. And
• 34:42
  that's how everybody
• 34:42
  became obsessed with like, this
• 34:45
  screwdriver is too high for me. So let's
• 34:47
  put this screwdriver
• 34:47
  down here. And you know what I'm saying?
• 34:49
  Like it was incremental. It
• 34:51
  wasn't huge. And I think a lot
• 34:52
  of times that like resonates with people.
• 34:54
  Cause a lot of people sit
• 34:55
  down, it's just like sitting
• 34:56
  down and being like, I'm going to write a
• 34:58
  book and then sitting down
• 34:59
  to write the book and you're
• 34:59
  like, well, fuck, I don't know where I
• 35:01
  even start with all this.
• 35:02
  And you're like, well, look,
• 35:03
  you start small. Like it's all brick by
• 35:05
  brick stuff. It's little
• 35:06
  things. It's what, you know,
• 35:07
  it's like auditing your time and then
• 35:08
  trying to understand that. So I think
• 35:10
  it's understanding that it can be
• 35:11
  smaller and like tracking and modifying
• 35:13
  that and going through and
• 35:14
  sort of showing people the
• 35:16
  possibility of like, look, you're going
• 35:18
  to keep working here. You're a great
• 35:19
  employee. And if you
• 35:19
  want more time with your family, like
• 35:20
  this is one way to do it, you know, like
• 35:22
  this is a way to make
• 35:23
  that more efficient for you. Um, and like
• 35:26
  incentivizing them that
• 35:27
  way too, if it's not the
• 35:28
  financial stuff. So not like threatening,
• 35:30
  but I'm saying like, you know, so like
• 35:31
  you want to spend time
• 35:32
  with your family, better get on it. It's
• 35:34
  like, no, like this is an
• 35:35
  opportunity. Okay. Like this is
• 35:36
  something you can do. It's, I think those
• 35:37
  are some, those are some possibilities
• 35:39
  you go, go through that.
• 35:40
  The final question I, Oh, go ahead. So
• 35:43
  first of all, I'll say
• 35:43
  you're just so much better at that
• 35:45
  than me. Like I just try to brute force
• 35:47
  it. And I think you're so much more
• 35:48
  thoughtful about and
• 35:49
  like empathetic about like, Oh, here's
• 35:50
  what people are actually
• 35:51
  thinking. I'm like, like,
• 35:52
  I just want people to use it. But how do
• 35:54
  you feel about like
• 35:55
  everything you're saying is right.
• 35:57
  Totally great. Give them the incentive,
• 35:59
  understand what they care
• 36:00
  about, like show them a shave.
• 36:01
  How do you feel about just like, also
• 36:03
  just like kind of
• 36:04
  making it mandatory? Like,
• 36:05
  do you think that's a good idea? Do you
• 36:06
  think that's a bad idea?
• 36:07
  Yeah. As a business owner and as the
• 36:09
  leader of any organization,
• 36:11
  you, you have the right to make
• 36:12
  anything you want to be mandatory. Okay.
• 36:14
  This is your organization. You can do
• 36:16
  that. The issue with
• 36:17
  mandatory statements or mandatory things
• 36:20
  in any organization is
• 36:21
  if you come in and say,
• 36:22
  this is mandatory period, that's a
• 36:23
  problem. Okay. And typically
• 36:25
  what people do in corporate
• 36:26
  settings, big corporate settings is
• 36:28
  they'll come in and say, this is
• 36:29
  mandatory. And then there's like
• 36:30
  a paragraph of text from the leader. And
• 36:32
  the people are like, what
• 36:33
  the fuck is this? And there
• 36:34
  isn't actually some sort of like run
• 36:37
  through of the why from the person
• 36:39
  leading the organization
• 36:40
  or the thesis behind it. You can disagree
• 36:43
  with that person, the
• 36:45
  leader, and some people may do
• 36:46
  that. But if you go through and say,
• 36:48
  look, here's the situation.
• 36:50
  Okay. We're making AI mandatory
• 36:52
  in 40% of what you're doing. And, and I
• 36:56
  want to make sure you are
• 36:57
  all there. And it's important
• 36:58
  because of X, Y, and Z. It's going to
• 37:00
  allow us to scale. It's going to allow
• 37:01
  you to spend more time
• 37:02
  with your family. It's going to allow you
• 37:04
  to have more time to free
• 37:05
  think, which is going to
• 37:05
  strategically be a better alignment for
• 37:07
  us. It's going to, well, you
• 37:08
  know, et cetera, et cetera,
• 37:10
  et cetera. That is a massive, massive
• 37:13
  leverage of being the
• 37:14
  emotional leader and saying the fear
• 37:16
  that I have is X, Y, and Z. The hope that
• 37:19
  I have is X, Y, and Z and
• 37:20
  going through all of these and
• 37:22
  speaking about it honestly and saying,
• 37:24
  I'm going to check back into this
• 37:25
  mandatory thing every month.
• 37:27
  And I'm going to talk to you about it.
• 37:28
  And like, I also may not make it
• 37:30
  mandatory in the future.
• 37:31
  If I feel like this is a bad idea, but
• 37:33
  right now it's absolutely
• 37:35
  something that I do need to make
• 37:36
  mandatory. And I think that it's
• 37:38
  something that we will have mandatory for
• 37:40
  the rest of the existence
• 37:41
  of this company. Right. So like there's
• 37:43
  so much of it is very like, it's this,
• 37:46
  and then it's dogmatic
• 37:47
  statement and that's where it ends. And I
• 37:48
  think a lot of it and getting people be
• 37:50
  motivated and wrapped
• 37:51
  into it is to show them the why and what
• 37:53
  you're hoping is the future. And also
• 37:55
  allowing them to say,
• 37:56
  look, if you vastly disagree, if you
• 37:57
  totally disagree with me on this, number
• 37:59
  one, like you can't come to
• 38:01
  me and be a jackass, but like you have,
• 38:02
  you have to some degree like
• 38:04
  protection. I'm not going to
• 38:05
  fire you for telling me if you think this
• 38:06
  is stupid or you think this
• 38:08
  can be better because if you do,
• 38:10
  that's you doing your job. I want you to
• 38:11
  know you're empowered to
• 38:12
  come to me and say, this is a
• 38:14
  problem. Like I had an employee that came
• 38:15
  to me literally this week
• 38:16
  and said like, Hey, X, Y, Z is
• 38:18
  bad in the community. Like a whole
• 38:20
  different. And I was like, awesome. Like,
• 38:22
  thank you. You know,
• 38:23
  like that's incredible. Like the fact
• 38:25
  that you care enough to
• 38:26
  tell me. Right. And so,
• 38:28
  uh, anyway, that's hopefully that helps
• 38:30
  answer, answer it, but I,
• 38:31
  you can, you can do that. You
• 38:32
  just have to sort of like describe it and
• 38:34
  you have to describe it
• 38:35
  from like the true state,
• 38:36
  true point of view that you have, that
• 38:38
  you would tell a friend
• 38:39
  that's an, and once it gets into
• 38:41
  corporate settings, the true leaders that
• 38:43
  do this, say it and say it,
• 38:45
  say it with an understanding
• 38:46
  and an empathetic understanding. And then
• 38:47
  they man the leaders under
• 38:49
  them to be able to describe it
• 38:51
  and not make it personality list,
• 38:53
  personality list, corporate speak is what
• 38:56
  like Gen Z millennials,
• 38:58
  even like, we hate this stuff because you
• 39:00
  can tell right away when it
• 39:01
  sucks. You know what I mean?
• 39:02
  It's just like the stuff people go on and
• 39:04
  they're like the first AI marketing
• 39:06
  growth, whatever the
• 39:07
  fuck you're like, this isn't real. Like
• 39:08
  you can tell it sucks
• 39:10
  immediately. You're like, this is bad.
• 39:12
  It's like, you know, and so like you just
• 39:14
  think about it through that
• 39:14
  lens. And I think that actually
• 39:16
  has like a lot of implications in, in, in
• 39:18
  being moving things
• 39:19
  forward in organization.
• 39:20
  Makes sense. Very thoughtful,
• 39:21
  much more thoughtful than me.
• 39:23
  Yeah. Ditto.
• 39:25
  Well, this is like so much of what I do
• 39:28
  now is helping
• 39:29
  organizations and agencies that are in
• 39:31
  the founders community, like bring them
• 39:32
  up in terms of talking about
• 39:34
  employee stuff and employee.
• 39:35
  And this is like, I live in that world a
• 39:37
  little bit now, probably 30%
• 39:38
  of my time is coaching people
• 39:40
  within that. So it's something I talk
• 39:41
  about a lot. And AI is
• 39:42
  obviously a big topic because
• 39:43
  everyone's trying to figure it out. So,
• 39:45
  um, good episode. I feel like we got a
• 39:49
  lot of good shit in here.
• 39:50
  I have another question. Do you have time
• 39:51
  to, if I ask one more question, Cody?
• 39:53
  Yeah. Yeah.
• 39:54
  Okay. So the one question is, I talked
• 39:55
  about KPIs and you said that you're
• 39:57
  redoing your KPIs now.
• 39:59
  So how has, cause I feel like, you know,
• 40:02
  there's, there's a lot
• 40:04
  of not focusing on the
• 40:05
  in-platform metrics, as you've said, but
• 40:07
  focusing on the looking at, uh,
• 40:10
  incrementality more often.
• 40:12
  How are you reshaping KPIs in the age of
• 40:15
  AI to be making better
• 40:17
  decisions in the short
• 40:18
  and long term?
• 40:19
  Expectations are just higher. Expect that
• 40:21
  people can, can do
• 40:22
  more, can get more done.
• 40:23
  You know, I just think that they can move
• 40:24
  faster. For us, KPIs are tied to bonuses,
• 40:27
  you know, as their overall rating, but,
• 40:29
  but also bonuses. You
• 40:30
  know, if I'm going to be paying
• 40:31
  out based on it, you know, it's gotta be,
• 40:32
  it's gotta be some pretty high
• 40:34
  expectations of what I think
• 40:36
  people are capable of and what I think
• 40:38
  best in class is capable
• 40:40
  of. And I don't think we're
• 40:42
  there yet, but I'm definitely going to
• 40:43
  set the expectations there.
• 40:44
  And so, you know, when it,
• 40:45
  when it comes down to the volume of
• 40:48
  output that we're expecting and number of
• 40:50
  ads launch or funnels
• 40:51
  launch or things like that, you know,
• 40:53
  definitely trying to pair
• 40:54
  them hand, hand in hand with it.
• 40:56
  You can just get, you can just get more
• 40:57
  done. You can just get more done.
• 40:58
  Yeah. It's interesting. I just curious
• 41:00
  your philosophy on it.
• 41:01
  It's not KPIs, but one of the components
• 41:04
  of our performance
• 41:05
  evaluation is AI usage.
• 41:06
  and utilization. I did a, last year we
• 41:09
  did a bounty and, you know, whoever did
• 41:12
  the most with it got,
• 41:13
  you know, won a bounty. And this year was
• 41:15
  like, all right, cool. Like
• 41:16
  that was a fun thing. Like
• 41:17
  now we're making it mandatory. Like this
• 41:19
  is just a general, and
• 41:20
  obviously we explained it,
• 41:22
  tried to explain it, but this is just our
• 41:23
  general expectation. You
• 41:25
  know, the same way you're using
• 41:26
  the computer and you're using Slack and
• 41:27
  email, this is a tool like
• 41:29
  that and we will train you and
• 41:31
  invest in it. And here's why we're doing
• 41:33
  it. But so it is part of, you know,
• 41:34
  people's evaluation.
• 41:36
  And, you know, it gets reviewed monthly.
• 41:38
  Is that just like token usage?
• 41:39
  No. Cause I don't
• 41:41
  want to incentivize that.
• 41:42
  It's just, they're kind of having to
• 41:44
  prove how they're using
• 41:45
  it, how they're improving
• 41:47
  their role and their, their function, you
• 41:51
  know, with, with it.
• 41:52
  Yeah. Yeah. That makes sense.
• 41:54
  That's interesting. Cody, thank you for
• 41:56
  being on. I appreciate it. And have a
• 41:58
  great rest of your day.
• 41:59
  Awesome. Thanks guys.
• 42:00
  Thanks for having me.
• 42:01
  Thanks for having me there, man.
• 42:01
  Thanks for having me sobre it, man.
• 42:01
  Thanks for having me here. Thanks, Ben.
• 42:02
  Thanks for having me on with this?
• 42:04
  Especially when you have a cry for you guys in the direction of part of your life or
• 42:06
  We'll be right back.
• 42:09
  We'll be right back.