YT YouTube2Text

YouTube2Text → Transcripts → Claude Will Change Agencies Forever (with Mike Rhodes)

Claude Will Change Agencies Forever (with Mike Rhodes) — Transcript

by Grow My Ads · 18,637 words · 2,806 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00You know, there's a lot of noise with
0:02AI. Just like you said, there's people
0:04on LinkedIn or Twitter or YouTube saying
0:07it's crap or you've got doomsday, you
0:10know, everyone's going to get wiped out.
0:12What's the point? And so there's a lot
0:14of this like there's fear, there's FOMO,
0:18there's I don't know what to do. And
0:20then there's also, wow, this looks very
0:23complicated to really do the nerdy the
0:25nerdy stuff.
0:26>> How good you are with AI at this point,
0:28and you kind of alluded to this before,
0:30is really a function of how curious you
0:33are and a little bit your willingness to
0:36or your tolerance for pain at this point
0:39because [music]
0:39it is a very iterative process. It is
0:42not going to work beautifully first time
0:44out of the box.
0:45>> All right, I've got the Godfather of
0:49Google Ads himself with me today.
0:52Um, Mike Rhodes. I, you know, I think
0:55it's probably been 10 years, maybe even
0:57longer. I was learning Google Ads and I
1:01found your community. Um,
1:04it's terrible. I forget the name now. It
1:06was like Web Savvy or that was your
1:07agency.
1:08>> It was PPC Savvy way back then and then
1:11it became Agency Savvy.
1:13Um, yeah, it's a different one now, but
1:15we'll get to that.
1:16>> And Mike, I think
1:17I think I was
1:20uh I think I paid $30 a month. I was
1:22like one of the
1:24first people cuz I by the time you, you
1:26know, had a lot of people in there. I
1:28think it was a few hundred dollars per
1:30month.
1:31>> Ended up at 300 a month, yeah.
1:32>> Yeah, so I was paying this grandfathered
1:34rate of like $30 a month out of my own
1:36money when I had no money back then. I
1:38was just trying to learn. And so I think
1:40I was one of the first, you know, at
1:41least probably 100 people in there. And
1:45uh it's cool. I met um I met Alex
1:48Fedheim in that community. I've never
1:50met him in person, which is wild. And
1:52we've worked together now for for 10
1:54years. And he introduced me to my
1:59business partner in Grow My Ads. And and
2:02so Luke, my business partner, Alex is
2:05the one who introduced us through Skype
2:08back then. You know, Skype's not even
2:10around anymore.
2:11And I started just kind of doing work on
2:15the side for Luke. So I'd manage the
2:18accounts that he would sell.
2:20And this is I was working at another
2:23agency that fast forward today I I now
2:26own half of and uh
2:28uh we ran that and built it to a nice
2:30seven-figure thing. And then COVID
2:33really rocked that agency. Our sales
2:35engine completely changed and that's all
2:37we had. And so that's when I made a
2:40pivot and started the brand Grow My Ads
2:42with Luke where we were just kind of
2:44freelancing at the time and and now Grow
2:46My Ads is, you know,
2:48really exploded over the last few years.
2:50So anyways, it all comes together now on
2:53this call and it's all because of of
2:56you. So there's probably no introduction
2:58needed here, but I guess a little
3:00background Mike and then we can kind of
3:01get into what you're doing with AI.
3:04You've got insane amounts of experience
3:06and history with Google Ads, but not
3:08only that, you were way ahead of the
3:09curve on AI even back in the the day in
3:12the community you would talk about AI,
3:14you know, before it was cool. So
3:16>> I I remember doing a talk at Traffic and
3:18Conversion. I remember Justin Brooke
3:20sitting in the front row. That was 2017,
3:232018. And the first time I talked about
3:26AI sort of
3:27proper.
3:28Um machine learning and AlphaGo had just
3:32happened. And I kind of concluded then
3:34we're we're too early. We'd I think I
3:36said this line of like one day
3:39AI will be like LEGO bricks that we get
3:41off the shelf and put together and we'll
3:42be able to and it'll be that easy. But
3:44right now you've got to learn Python and
3:46build a neural network and
3:48no one's going to do that.
3:50Focus on automation. Come back later for
3:52the AI. I did not predict
3:54Chat GPT
3:56Um but I do remember being on stage in
3:58'22 talking about images and you know,
4:01there was this website back then um this
4:04person does not exist. It probably still
4:06is out there. This person does not
4:07exist.com and it kind of showed the
4:09progression of how AI could create
4:13images of people and it was getting more
4:14and more realistic from '20
4:1713, 14. By 2018 it was pretty realistic.
4:20By 2020 it was like you can barely tell
4:22the difference.
4:24And here we are now where AI is
4:28running huge chunks of
4:30lots of agencies. It's not replacing us,
4:32but it's helping with
4:34huge amounts of work. Um yeah, I'm I'm
4:39fascinated by I get to do what I've
4:41always loved doing, which is be the
4:44scout and and run four or five hills out
4:46into the future.
4:48I kind of live 6 to 12 months in the
4:49future I think and every now and again I
4:51come back to the present. But I don't
4:53like to come back and report on that.
4:55Say here's what I think I saw and I like
4:57to come back and make it really useful
4:59and practical for people and and give
5:03tools, build tools. Somebody on a
5:05podcast yesterday said you you just
5:07you're a builder, right? Um I don't I
5:08caught myself smiling
5:10cuz that's what I did when I was a kid.
5:12I would build stuff out of LEGO and
5:15bricks and build stuff in the backyard
5:17and um
5:18and that's what I'm doing now all day
5:20every day. I just get to build stuff,
5:21but I I have this need to be useful. So
5:24I want to build stuff
5:25and then
5:26give it away or maybe sell it
5:28occasionally. Like the display grid way
5:30back in the day when we first were
5:31talking. Um
5:33um
5:35the profit curve tool and now it's AI
5:38stuff. So yeah, I'm sure we'll talk
5:40about the brain at some point. Um but
5:42it's
5:43tools tools to to help agencies, to help
5:46businesses, to help freelancers, and and
5:50to help people with their careers, too.
5:52I'm really, really
5:54keen on this at at at the moment,
5:56because there's far too much crap out
5:58there. There's two ends of the spectrum
5:59out there. There's either like, "Oh,
6:01it's going to kill us all and and what's
6:04the point?" Or it's
6:06"AI is crap. I tried it once. I did a
6:08Dr. Seuss poem, and it it can't do what
6:11I do, and it's crap." And there's these
6:12two schools of thought, and it's like,
6:13"No,
6:14there is a middle ground here. Let's be
6:15sensible about this. It can't do
6:17everything. It's spiky. It's jagged."
6:21The only way to figure out what it's
6:22great at
6:24and what it's terrible at, and sometimes
6:25those two things are really close
6:26together, which doesn't make any sense,
6:28right? You think, "Oh, it's great at
6:29this. It'll be great at the next thing
6:30and the next thing." But it's terrible
6:32at that. But then it's great at the next
6:34thing. And it's
6:35it doesn't make any sense, but the only
6:36way to find that out is to play and
6:38experiment. And and yeah,
6:41be in a community of people sharing
6:42ideas. Sorry.
6:43>> Yeah, you've been doing it for a very
6:44long time, too. You had So, you had PPC
6:48Savvy for How many years did you run
6:50that agency?
6:52>> Oh, so Web Savvy was the agency. I ran
6:54the agency I ran the agency for about 17
6:58years. So, yeah, sold that about 2 and
7:001/2 years ago.
7:01Uh I had two kids that were still just
7:04in primary school at that point. Still
7:05have the two kids. Uh but one of them
7:07isn't in primary school anymore.
7:09Um
7:09and I was just very, very aware. I had
7:12this thing with my elder daughter,
7:13Kiara, where we were counting the Monday
7:15mornings, cuz we live about 3K's, what's
7:18that? 2 mi
7:19from the school. And so, I would manage
7:21to persuade her to walk once a week. And
7:23the rest of the time, Mom would take her
7:24in the car.
7:26And we started this at about 100 weeks
7:29out. So, halfway through grade four, we
7:31had 100 Mondays left together. And we
7:34we're counting these down. I think we
7:35had a little chart on her bedroom wall.
7:37Um
7:39and then yeah, fast forward to the point
7:41where I was sort of
7:44starting to think about either either
7:45growing and buying another agency or
7:48selling.
7:49And with my younger daughter, we worked
7:52out I had 134 Mondays left.
7:56Uh we're down to about 37 at the moment.
7:59Um so that's how long ago it was that I
8:01sold, but it was just this realization
8:03of once I'd started the thing with Kiara
8:06and when I sold, I think we were we
8:08literally had
8:09What was it? Like 16, 17?
8:12>> Yeah.
8:13>> Left. And it was just I am going to milk
8:15this moment because this ain't going to
8:18happen again. Like once these 16 are
8:20done, she is never going to primary
8:22school ever again. I don't get this walk
8:24with her because she doesn't want me
8:25anywhere near her when she's going to
8:26high school, right? You don't walk your
8:27kids to high school. They're like, "Get
8:29Get away from me."
8:30>> No, yeah.
8:31>> And once you see that, you can't unsee
8:32it. And I was just
8:34very very aware that
8:36the the the this little period of time
8:39isn't going to happen again. So,
8:41uh we'll just
8:44get out of all of that and focus on this
8:47for a bit.
8:48And
8:50then wouldn't you know, AI comes along
8:52in a completely different form
8:54a little bit later and just
8:56and [snorts]
8:57this is the stuff that I would do
9:00anyway, right? This is the stuff that I
9:01would do if I wasn't being paid to do
9:03it. I love this stuff. I love building.
9:06I love running 6 to 12 months into the
9:07future and and figuring this stuff out.
9:11I loved back in the day climbing inside
9:13the Google engine and working out how it
9:16worked
9:17and teaching that. Along the way,
9:18discovered I was a bit of a teacher at
9:20heart.
9:21So yeah, it's it's it's I'm having so
9:23much bloody fun.
9:25>> That's amazing. So you sold and then you
9:27started I mean what is now I I don't
9:29think it had the name that it has today,
9:31but Ads to AI, correct?
9:33>> Correct. Yeah, sorry, just to finish the
9:34timeline. So yeah, I I had released this
9:37free PMax script that something like
9:4015,000 agencies have since
9:42>> blew up, exploded. Yeah, by the way, the
9:44greatest script my typical LinkedIn post
9:47>> My typical LinkedIn post would have
9:48about seven impressions. Not not not
9:51comments or anything, impressions.
9:53>> Wow.
9:53>> And then that PMax post, I think between
9:57the three of them I ended up having had
9:59a million impressions. It was
10:01ridiculous.
10:02>> Amazing.
10:02>> And so, yeah, right after I sold it, it
10:04was my IP.
10:05Um I talked about it with the new
10:07owners, "No, no, that's your stuff, you
10:08take it." Okay, then.
10:10And I was speaking at Gianpaolo's event
10:12in Bologna, so I figured, "Well, that's
10:14a good time to launch it."
10:16Um started selling the free version of
10:19the script and then I'm sorry, so
10:20started selling the paid version.
10:22Where are we now? October 2023.
10:24And it just went bananas. And it just
10:26kept going bananas because Google didn't
10:28come up with their version of that until
10:30what? Almost 2 years later.
10:33And then off the back of that, lots of
10:34people were like,
10:36"Scripts, I need to learn scripts. Can
10:37you teach me about scripts?" So, then I
10:39had workshops and stuff teaching scripts
10:41and Google Sheets.
10:43That led to me starting to build more
10:45and more stuff with AI in 24.
10:48So, then I built it so I would learn how
10:50to do it cuz I've never taught
10:53textbook stuff. I want to know how to do
10:55the thing if I'm going to teach it.
10:57So, I started building
10:58agents. Um
11:00not what we would call agents today.
11:01It's changed a bit since then, but
11:03started building a whole bunch of stuff
11:05with AI.
11:06And then started teaching that and
11:08bringing these build the agent courses.
11:11This is a little over a year ago. And it
11:13was off the back of that that everybody
11:14went, "That was awesome. That was a
11:16great 8 weeks, but now what? What have
11:18you got?" So, that's when I started Ads
11:21to AI, beginning of last year,
11:23as the community of scouts and builders
11:26and people that just want to get their
11:28hands dirty and play with this stuff,
11:30but don't want to do it by themselves
11:32and want a head start. I don't want to
11:34figure everything out from scratch.
11:36Give me a starting point.
11:38Filter out the
11:40the noise. Give me the signal. So, I do
11:42like a weekly video each week, and we
11:43have a monthly Q&A, and then I'm
11:45building stuff all the time, skills and
11:47tools and all these different things,
11:49and it's just so much bloody fun. It's
11:51great.
11:53>> We We were chatting before I, you know,
11:56>> [clears throat]
11:56>> we started recording this, but uh I was
11:59saying that, you know,
12:02uh you know, like I here at Grow My Ads
12:03have a team of 20 employees. I'm very
12:06much in the weeds of the agency still.
12:09Um
12:09>> Right.
12:10>> So busy, and [clears throat]
12:12>> [laughter]
12:13>> you know, you you've been through it,
12:15right? Uh
12:16and
12:18you know, there's a lot of noise with
12:20AI. Just like you said, there's people
12:22on LinkedIn or Twitter or YouTube saying
12:25it's crap, or you've got doomsday, you
12:28know, everyone's going to get wiped out,
12:30what's the point? And so, there's a lot
12:32of this like there's fear,
12:35there's FOMO, there's I don't know what
12:37to do, and then there's also, wow, this
12:40looks very complicated to really do the
12:42nerdy the nerdy stuff, right? Like,
12:44sure, I have a
12:47I had a ChatGPT account, Claude account,
12:49obviously Jim and I, and I
12:51I would kind of use them to just search
12:55like most people, and then and then
12:57Claude I started gravitating naturally
12:59more to Claude. I just enjoyed it
13:01better.
13:01>> Mhm.
13:02>> And then yeah, sure, I started adding in
13:04spreadsheets and doing things through
13:06through chat. But, when I see the
13:09terminal, which Luke, my business
13:11partner here at Grow My Ads, he's for
13:13the last couple years been really
13:14nerding out on AI.
13:17>> Of Luke.
13:17>> And
13:18he's in you know, he was showing me the
13:20terminal. He's got multiple, you know,
13:23screens running, you know, he's got
13:24codex running here and Claude code here,
13:27and he's going back and forth, and it
13:29looks
13:31chaotic and complicated, and I'm like, I
13:33don't have the time, right? It It's It's
13:35just like I'm not going to spend the
13:37time to learn this. Then,
13:40Alex Fedheim, who is a part of Ads to
13:42AI, started talking to me about what
13:45he's doing, and he He's like, "Hey,
13:46listen, Mike has actually made this very
13:49easy to learn. Go spend like a few hours
13:53and just go through his tutorial videos
13:55in the group, and just play around with
13:58it." So, I decided, "You know what? I
14:01I'm going to spend a weekend. This is my
14:03weekend. I'm just going to learn what
14:04Mike has. I'm going to go step-by-step
14:06through your videos, and I'm just going
14:08to see what it's about." And you made
14:11that
14:12unbelievably easy. I think I told you I
14:14still don't really know what I did,
14:16because, you know, you you you get the
14:18terminal up, and you've got to download
14:19these things, and it's doing all this,
14:21and I'm like, "It's working. Whatever's
14:23happening is working."
14:25And
14:27uh lo and behold, I I got hooked, right?
14:29And then I started playing around, but
14:31the brain made it very simple. It was
14:33like it Like you said, it's like an
14:34agency OS in the box, and
14:38um it's because then of how easy you
14:40made it that I you know, Claude code to
14:42me is like I I'm in that every single
14:45day. It's just a part of my workflow
14:48now, and I'm barely barely even
14:50scratching the surface now. And we're
14:53now at you know, I'm at a point now
14:55within the organization of like
14:57we can get into this a bit later is
14:59like, how do I deploy this with a team
15:02of 20? You know, it gets
15:03>> [laughter]
15:04>> It gets a bit uh
15:05complicated in that, but anyone who's
15:08watching this, and you're just using
15:10chat, right? You're just using going to
15:12Claude on your web browser and chatting
15:15or or chat GPT,
15:17um it's it's really not that difficult,
15:21and it's it's once you get it set up,
15:23especially your video training on how to
15:25do it, it it's very simple. In fact, I
15:28love it now. I I like it to me it's it I
15:31would rather spend my time in Claude
15:33code than a Google Ads interface today.
15:35I would rather just work within Claude
15:37code itself.
15:38>> Yeah. Um and so coming from a guy
15:40>> started.
15:41>> Yeah.
15:41>> That's how it started. I I would be
15:43doing work and every time I would go to
15:46a different window outside of this
15:48horrible thing called terminal. And
15:49don't worry, um dear listener, if you're
15:51listening to this and going, "Oh, I
15:52don't want to do this thing in this
15:54thing that looks like it's from 1987,
15:57this black screen." I have a fix for
15:58that. But whenever I would go outside of
16:01that, and I got used to it, and you can
16:03turn it from black to white or purple if
16:05you want.
16:06I would be asking myself like, "Why am I
16:08going here? What am I doing here? Why am
16:11I logging into Gmail? Hey brain, can you
16:13do that?" Yeah. "Okay, we can just pull
16:15in all of the emails and you can alert
16:17me when there's one that looks
16:19important, not that Amazon just
16:21delivered your thing, or this thing just
16:23billed you more, but you need to reply
16:25to this in the next 2 hours. Can you let
16:26me know so I don't have to keep going
16:28over to Gmail just to check just in
16:30case? And all those emails that come in
16:32overnight, can you write draft replies
16:34to those? Don't send anything, you're
16:35not allowed to send anything, but could
16:36you write drafts
16:38so that I wake up to 13 drafts ready to
16:41go rather than
16:43an inbox full of noise that I've got to
16:44spend the first 2 hours of every day on
16:46and get knocked out of flow? Yeah, I can
16:48do that.
16:49Great. And it's just like you have this
16:51conversation with this
16:53occasionally incredibly stupid, but most
16:55of the time incredibly brilliant
16:58Yeah, there's there's lots of ways to
17:00think about it. I think AI should stand
17:02for alien intelligence because it's it's
17:04it's human-like,
17:06but it's also really weird, and it will
17:08do the strangest, dumbest things. We
17:11tend to What's the word I can never
17:13pronounce? Anthropomorphize
17:15a lot of stuff, right? We see faces in
17:17the clouds and
17:19that's just a million years of evolution
17:21so that we can spot the tiger or that is
17:23a snake or a stick. I need to know
17:25really quickly. Yeah, that's just the
17:27way our brain works and so we tend to
17:29think these things
17:31are human, but they're not. They're just
17:33they're mimicking us. They're just
17:35holding up a a mirror. So when it does
17:38something wrong, it goes, "Oh, I'm yes,
17:39I'm sorry. I shouldn't have done that.
17:41You're absolutely right. I messed up."
17:43Because it thinks that's what we want to
17:45hear and that's what it's seen in all of
17:47the training when people mess up, they
17:48say stuff like this.
17:50It's not sorry. It doesn't have
17:51feelings. It's just mimicking and it's
17:54really bloody annoying when it deletes a
17:56bunch of your code and it says, "Oh,
17:58you're absolutely right. I shouldn't
18:00have done that."
18:02Okay, what are you going to do now? It
18:04just like cost me 500 bucks. I'm like,
18:06"Yeah, that's terrible, isn't it?" Yeah,
18:08you know, it's it's frustrating at
18:10times, but
18:11>> Yeah.
18:11>> it is absolutely incredible technology.
18:15I I promised to an interim step. Can we
18:18go there?
18:19>> Yeah, let's do it.
18:21>> Um
18:23What's the best framing for this?
18:24Skills, I think.
18:26I'm jumping ahead, I think, by doing
18:28this, but I think it's worth mentioning.
18:29I'll just mention I'll throw it out
18:30there now. 8020skill.com
18:33is a site I built
18:35to show people to to make you play with
18:39it in a very easy and
18:42fun way because there's some games
18:44hidden in there if you go through the
18:46steps
18:47to
18:49viscerally show you
18:51how good AI can be. I've made it as drop
18:53dead simple as I can. There's three
18:56little demos. You don't have to do
18:57anything. You just watch. Then you get
18:59to level two
19:00and for each of the things on level two,
19:03you download a skill.
19:05You download some sample data. It's like
19:07literally click and it downloads. Click,
19:09it downloads. And there's a prompt. and
19:12the prompt is on purpose really, really
19:15short cuz there's no like all magic
19:17stuff. A lot of people when you see
19:18these like 14-page prompts, they're
19:20like, "Well, I could never write that."
19:21Exactly, but you can write a one-line
19:23that says, "I want to do this with that.
19:24Can you do it?" Great.
19:27And then you use a tool
19:29which you mentioned before. Claude has
19:31this desktop app. So, you go download
19:34that. Takes you 10 seconds to set up.
19:37And we've got chat, which like you said,
19:40chat GPT. Claude has their own version
19:42of that, Claude.ai. That's at this end
19:44of the scale.
19:45At the other end of the scale, you've
19:47got Claude Code running in a terminal
19:49window like Luke was doing, like you're
19:51now doing, like I'm doing every day.
19:53That's the uber geeky version that looks
19:55like it's from 1987.
19:57Code work
19:58is about here. It's slightly less
20:01powerful, but it's got the same
20:03interface as chat. So, everybody knows
20:05how to use it. It's just a text window,
20:07but there's a ton of power hiding under
20:09the hood.
20:11And all I want listeners to do, if
20:13they're still scared of AI, is make make
20:16it to level two of 80/20 skill.
20:19Click, click, click, download these
20:21strings, and just drag them into Claude
20:24Code, and watch it go to work using a
20:27skill.
20:29So, you pick who you are. Are you an
20:31agency owner? Are you a freelancer? Are
20:33you a business owner? So, that the use
20:35cases are more tailored to you because
20:37I've discovered that the closer
20:40a use case you can show someone that's
20:42that's as close as possible to their
20:43world,
20:45it's this moment of like
20:47"Oh, AI can do that." But if you show
20:49them a use case that's got nothing to do
20:51with them, it's like, "AI is AI is
20:53shit." This is This is my world here.
20:55This is my Venn diagram with hands. This
20:57is my world or your world, and this is
20:59AI me showing you AI. And it's like, if
21:01I show you AI that's not your world, AI
21:04is rubbish. AI is rubbish. Even if it
21:06overlaps a little bit on that Venn
21:07diagram, AI is still rubbish. Unless
21:09it's almost exactly overlapping
21:12and then there's this moment of
21:14Oh my god.
21:16And once you see it, you can't unsee it.
21:19And then, like you did, you get really
21:21curious.
21:23You spend a few hours, yes, mucking
21:25around. You don't know what you're
21:26doing, but you hit this moment of like
21:29Wow. Okay, and now I want to lean in.
21:32Now I want to spend a couple of hours a
21:35week learning how this stuff because I
21:36can see how much this is going to help.
21:39And once you reach that point, you
21:40really almost can't imagine going
21:42backwards. You can't go back to a world
21:45where I have to do everything by hand.
21:47That's what this website is designed to
21:49do. There's There's a few levels.
21:51There's a few different
21:53demonstrations, examples in there
21:55with sample
21:57um meeting notes or a sample
21:59spreadsheet. So, you don't have to
22:00wander around and oh god, now I've got
22:02to download some data in order to test
22:03this thing out.
22:05But please just play with it. Please
22:07make it to level two. Have a look for
22:09the Super Mario option and um and have a
22:11play with that. I've made it fun. That's
22:14kind of the the free stepping stone so
22:16that then you
22:17experience it and then if you gain then
22:20yeah, go play with Cloud Code, which is
22:22just another button in that app. You can
22:25start using Cloud Code in there on a
22:27wide screen instead of the scary-looking
22:28terminal. But that I think is the It's
22:31that moment. You got to go from zero to
22:32one. You got to see what's possible and
22:35and feel it rather than just be told it.
22:38It's not a course. It's not It's like
22:40riding a bike, right? You can't learn to
22:41ride a bike by reading a book. You got
22:43to get on the bloody bike, fall off a
22:44few times, and then
22:47there's this moment where you're like,
22:48"Oh, I'm riding, Dad. Oh, this is
22:50awesome."
22:51It's the same thing with this.
22:53>> It is that um
22:55it's I had that moment when I actually
22:57went through
22:58uh your course and
23:01uh you get hooked. and now and then
23:04you're energetic. It's probably it's
23:06probably the same feeling of like your
23:08first time setting up a Google ads
23:10campaign that works and you get like
23:12conversions and you're like
23:13really straight and just like
23:15ball straight. You yeah, like it is that
23:18moment and then you can't sleep because
23:20your your your brain's racing. You've
23:22got a million different ideas and and it
23:26just takes I think like you said
23:30in your world seen what it can actually
23:33do. Now anyone listening to this video
23:34is obviously going to be a part of like
23:36running Google ads accounts. So your
23:39what's the website again?
23:408020skills.com?
23:42>> 8020skill.com.
23:43I should probably buy the one with the S
23:45as well. I've got a million domains. I
23:46should probably buy the other one.
23:47>> I'll have skill 8020skill.com. I'll have
23:50the link in the description below on
23:52this video so people go check out all of
23:54them.
23:55Yeah, I'll link to ads to AI your group
23:57as well. I think everyone who is trying
24:00to self learn this right now just go
24:02take Mike's The course is free too.
24:04>> That course that you went through I've
24:05since put it up on YouTube so I'll send
24:07you the link to that too. It's like 50
24:09videos. That whole thing is free to
24:11learn.
24:12>> Yeah, well that's what what that's
24:13amazing.
24:14Why would you do that? That that was
24:16that right there that that was such a
24:17like that course itself was worth I
24:19forget what I paid but it's like worth
24:2010x that now so
24:22>> Oh so yeah so ads to AI is is really
24:24it's the community and it's all the code
24:26for the brain to get you started. I'm
24:27not giving the brain
24:29>> Okay. Okay. Yeah, yeah. Yeah, yeah.
24:30>> the 50 videos that explain what it is
24:32and how to use it and how to build on it
24:35and how to start doing Google sheets and
24:36Google ads and
24:39anything you [clears throat] want my
24:40Gmail, calendar, meeting prep, email
24:44triage. Like these are all things that
24:45you you mentioned the good morning
24:47command just before we started
24:48recording.
24:50There's heaps of us in the community
24:51that start our day by walking up to an
24:53AI and going /gm and saying good morning
24:56to a bot. Not on Twitter or somewhere
24:59else, but saying good morning to a bot
25:01and it goes
25:03Yeah, first thing I do, too. And it
25:04tells me
25:05>> edit it. I I'm editing every day. I'm
25:07like, "Well, actually, let's take this
25:08cuz I you know, I have it scrape YouTube
25:10channels I like and give me summaries of
25:12the new videos and like emails." It's
25:14So, I'm like every day like it's been
25:16now well over a month and so it's
25:19evolved into this thing.
25:21>> Great. That's exactly what it should be.
25:23>> to me. Yeah, it's it's incredible.
25:26>> You imagine walking into an office and
25:28you've got the world's perfect assistant
25:30who just knows you and they know exactly
25:33what you want. Do you want to know about
25:34the day ahead? Do you want to know what
25:36your competitors are doing? Do you want
25:37to be given ideas
25:40um or better yet, know what your
25:42competitors are doing, which one of
25:43those performed the best and here are
25:45some ideas for your next video that will
25:47actually blow up on LinkedIn or Twitter
25:50or YouTube. And like, what is it that
25:53you want? Do you want meeting prep notes
25:56for that meeting that you have at 10:00
25:57a.m.? When do you want those delivered?
25:59Do you want that now or do you want it
26:01at 10:00 to 10:00? Do you want me to
26:02text it to you if you're in the car? Do
26:04you want me to read it to you as a
26:05podcast on your way to the meeting?
26:08Like,
26:09that's that is the
26:10that's the hardest bit.
26:12Is just knowing what you want.
26:13Imagination is the only constraint now.
26:17It's
26:17>> Yeah.
26:18>> I describe it, we mentioned Lego before.
26:20The way I describe it is we now have the
26:21world's biggest Lego kit. 4,000 pieces
26:23in this thing, but you only need to
26:25learn about 20 of them to get tons and
26:28tons of value.
26:30>> I think everyone who's watching this
26:33start like they shouldn't even go try to
26:36self learn. I'm telling you, your brain
26:39that you've already developed because
26:40you can do what I'm doing. I'm I'm
26:43I'm starting with the brain, but then
26:45I'm building on top of it, right? For my
26:48own needs. Like, I know what you have
26:50preloaded skills in there for Google Ads
26:52stuff.
26:52>> Just to get people going, yeah.
26:53>> Right. So, I think like one of them was
26:55a negative keyword thing, but I have I
26:57already was had a negative keyword sort
27:00of skill I was trying to create. And so,
27:03I've I actually cross-referenced what's
27:06Mike's do that mine doesn't. And then I,
27:09you know, it took 2 weeks of reiteration
27:11cuz I And that is a part of the problem
27:13with all of this is you end up becoming
27:15a like it's never going to be perfect.
27:17[laughter]
27:18>> And the hardest thing is knowing when to
27:19stop, right?
27:21You can go from idea to implementation
27:23in literally 5-10 minutes and build
27:25version one. But once you see version
27:27one, you look at it and you go
27:29Oh, wow. Um that was cool. That was
27:30easy. Um oh, you know what would be
27:31really cool if we had a drop-down there.
27:33I I want to be able to choose between
27:35all campaigns and choose an individual
27:37campaign.
27:38Oh, done. Um I actually want to be able
27:40to color code those. Oh, I need filters.
27:42I want a slider so I can say like only
27:44show me the search terms with more than
27:46100, 500,000 impressions. Can you Oh,
27:49it's already built it. Right. Um and
27:51it's just like
27:52because it's so fast to get the idea out
27:54of your head and build the thing,
27:56the ideas keep coming. You get in flow.
27:58And the hardest thing is walking away
28:01from the screen at the end of the day.
28:03It's like Yeah, I'm I'm coming I know
28:05dinner's ready, but I
28:06I just like quickly want to give the
28:08machine more stuff to do while I'm not
28:11here. Can you build this? Can you build
28:12that? Like
28:13>> It is incredible like I'm going to share
28:14my screen real quick. This is coming in
28:16from a guy and that you know, you won't
28:18even probably be uh too impressed by
28:20this, but um this is coming from a guy
28:23>> I keep top of it. That's exactly what I
28:25built it for. Is I don't want to be
28:27prescriptive, right? It's like here's
28:29the Lego kit. Let me explain the most
28:31important pieces. Let me build the first
28:32couple of models for you to show you and
28:34then go build.
28:36>> I want to show people like I'm a guy who
28:40uh was afraid to use cloud code, right?
28:43And get in the terminal. Now, I this is
28:47This is like our audit process. I'll do
28:49our large audits that come through. So,
28:51these will be like big spenders, 50K
28:53plus per month usually is where like
28:55it'll get sent to me.
28:56And um you know, I honestly used to just
28:59do only a Loom video and then if I had
29:03if we had the resources and time, I
29:05would have like a slide deck created
29:07from that. All manual work right there.
29:09>> Yep.
29:09>> I now have by giving it my brand
29:13guidelines, so my my brain now knows
29:16Grow My Ads brand very well. Anything I
29:18create, spreadsheets, whatever, it it
29:20knows what to go get for the brand
29:22guidelines.
29:23>> Yep.
29:24>> And I now was I've been able to take
29:27This was built from a Loom video and
29:28then working with some previous um you
29:31know, reiterations of this. This is not
29:34the first one. This audit was created
29:37essentially in 5 to 10 minutes. There's
29:40some reiterations here. This
29:42>> is a This was done I mean, for me to
29:44have my team manually build something
29:46like this in the past would have taken
29:48so much time and resources. This was
29:50done in 10 Now, the audit itself still
29:53took time um
29:55and I'm not I With audits, I I actually
29:58don't use Claude code cuz like I do I I
30:00don't even download data from their
30:02account cuz I I just I feel like they
30:04gave me access. I'm not even going to
30:05download a CSV of their data here. So, I
30:07do manually review their accounts cuz
30:09it's not hooked up to our MCC. So, I
30:12don't have the API integration there.
30:14>> But, I can take a Loom video of my audit
30:18and the sales All of our sales calls are
30:20recorded. It take I take that, I take
30:23transcripts, feed it in, and then I go,
30:25"Go build me an audit deck." It now
30:27knows what that means, right? Uh based
30:30off of this audit. And then sure,
30:32there's a few things I got to go
30:33reiterate. It nails it almost every
30:35time. And they I'm talking I've closed
30:37We've closed like 10K plus per month
30:39deals with this exact process I'm
30:41talking about. And this is coming from a
30:43guy 1 month ago who wouldn't even
30:46think of using Claude code. And now I'm
30:48closing 10K per month deals with the
30:50power of it. Very simple.
30:52>> use that quote.
30:53>> There [laughter] we go.
30:55You can use whatever quote you want.
30:56>> You but you had this really cool line in
30:57there. If I had the time and resources
31:00to do it, that is a great way to attack
31:02AI, right? It's like, what are the
31:05things you do? Not just let's automate
31:07the stuff we're already doing. That's
31:09That's an okay place to start. Start
31:11with that. Free up an hour or two a
31:13week, but then use that hour to start
31:17thinking a little bit broader. What are
31:19those things that you occasionally do
31:21that you would love to do more of, that
31:22you would like to deliver to a client if
31:25only you could spend 20 hours on it. But
31:28what do you know? We don't have a spare
31:2920 hours lying around, and I don't want
31:32to
31:33pay for 20 30 hours of the team time to
31:37put this thing together. But my god,
31:39that client would love it if we did, but
31:41they're just not paying us enough. But
31:43now that is possible. We did this
31:45challenge back in February in the
31:48community.
31:49Um
31:50We built a couple of different things, a
31:52landing page audit tool, and a
31:53competitor intelligence report that went
31:56off and you had told it, "This is me and
31:58here are five of my competitors." And it
32:00went off and looked at all of their
32:01sites, read all of their copy, took
32:03screenshots of the site, and bundled
32:05that all up into a beautiful 18-page
32:08PDF, as you say, fully on brand.
32:12You don't send it without checking. You
32:14check everything.
32:16But it solves the blank page problem. It
32:17gets you 80 90 maybe 98% of the way
32:21there,
32:22and you need to change it a bit, tweak
32:24the voice, but the more you do it, the
32:26more it learns who you are, how you
32:27speak. You give it rules and say, "Never
32:29use the word genuinely again, please.
32:31Just Just get rid of that.
32:33Get rid of the M dashes, all those silly
32:35things."
32:37But people were then able to go give
32:40that to a favorite client, build
32:42reciprocity, build the relationship,
32:45and position it as we can do more of
32:47these. Next one's not going to be free.
32:49First one's free, but do you want us to
32:52do one of these once a month, once a
32:53quarter? Do you want to do the same five
32:55competitors every time? Do you want to
32:57mix it up? What would you like? Because
32:59we now have the resources to do that.
33:01And there is this gap where that is
33:03going to work, right? There comes a
33:05point, I don't know when,
33:07couple of years from now,
33:09where
33:10most small businesses are going to say,
33:12"Oh, that's just AI."
33:14Um yeah, that was just a thing that you
33:16did, only took 5 minutes. Why would I
33:18pay 500 bucks for that?
33:20But right now, the fact that you know
33:22how to use these tools and you can wield
33:24them in that way and AI is just one tool
33:27that you're using as part of the mix.
33:29It's not like anybody can just walk up
33:31and go, "Oh, competitor intelligence
33:32report, please." and get this 18-page
33:35gorgeous report that looks and sounds
33:37like you and is what you would create
33:39and put your name to.
33:40Most people
33:42can't do that.
33:43Yeah, how
33:45how good you are with AI at this point,
33:47and you kind of alluded to this before,
33:49is really a function of how curious you
33:52are and a little bit your willingness or
33:55or your tolerance for pain at this
33:57point. Because it is a very iterative
34:00process. It is not going to work
34:02beautifully first time out of the box.
34:05But you can build a really quick version
34:06one or version 0.1 really quickly. And
34:10if you're willing to just spend a little
34:11bit of time, as you said, to learn,
34:13spend a weekend, to understand the LEGO
34:15bricks, to learn how the moving parts
34:18work. And yet,
34:20the thing I've built is designed to get
34:22you from zero to one very quickly, but
34:24not from one to 100. That's That's on
34:27you, right? Then you go iterate. You do
34:29what you've done with this thing. You
34:30add your own branding. You add your own
34:32sections to the report. I don't like
34:34that thing that Mike built, throw that
34:36out, but I want this new thing.
34:37>> Yep.
34:38>> And it just But it But then
34:40it gets it perfectly right. It's like
34:42that dream employee you wish you could
34:44have hired. Of once you say, "Oh,
34:46remember we always do this as section
34:48five." it never forgets that. And people
34:51will be yelling at the screen going,
34:53"Yeah, but it hallucinates and it gets
34:54stuff wrong all the time." It depends
34:56how you build it.
34:57>> Right. Right. You know, um
35:00I I have a comment on cuz you said
35:02businesses or clients, you know, will
35:05say, "Oh, you just did that with AI,
35:07right?"
35:07>> Mhm.
35:07>> And we've already had that um with some
35:10clients. Like, "Oh, that was just
35:11generated with AI." But uh to me, I like
35:16to attack that. So, first, like I if I
35:19send an email and I recap it, like cuz I
35:22do a lot of Loom videos, right? And so,
35:23I will I'm just up front in a lot of
35:26times. AI summary.
35:27>> Yeah.
35:28>> So, I'll just I I'm not going to fake
35:29it. This is a summary from AI based off
35:31of this video. If you want the nice
35:33>> useful.
35:35>> Right. Now, that audit, if they said,
35:38"Well, AI did that." Yeah, you're
35:41absolutely right. We used a tool to help
35:44build our knowledge that we just gave
35:47you, right?
35:49>> Yeah.
35:49>> Who cares if a human wrote that PDF doc
35:52>> 10 years of wisdom.
35:54>> AI? You're not paying You're paying for
35:56an outcome that we're able to achieve. I
35:58mean, an agency, I don't care what
35:59anyone says, an agency is nothing more
36:00than a brain trust of human beings,
36:02right? So, you're you're going to this
36:04group collective group of humans that
36:07can do something, and in this case ads,
36:09right? Better than what you can
36:11internally. And so, it's more valuable
36:13for you to just go pay them to go do
36:15that to get the outcome you're looking
36:16for. And so, I'm I've you know, I've
36:19told my team, like, just Yeah, attack
36:22that. Tell them you cannot replicate
36:24what I just did. So, please go go to
36:26chat GPT, Claude, whatever, and see if
36:29you can replicate that. You can't
36:30because we have our own brain, right?
36:34So, the brain trust of Grow My Ads, the
36:36Thank God we've recorded
36:39uh for the last 5 years almost
36:41everything with Loom. So, we have
36:44thousands of thousands of videos and we
36:47have already transcribed all of them and
36:50fed it into now what is turning into the
36:53Grow My Ads brain. Right? Yeah. So,
36:56yeah, you can't You cannot go just do
36:59what we're doing. Yes, we're using AI to
37:01help produce a final asset, but the
37:05data, the knowledge, the expertise, all
37:08of that is us and this is the outcome of
37:11that. So, I I just kind of attacked
37:13that. I'm not afraid of someone saying,
37:15"Well, just AI did that." That's not
37:16that's not the case.
37:18>> Yeah, Kirk Williams had a lovely post
37:19the other day. I think he's He put a
37:21little registered trademark after it. I
37:23think
37:23"a Claude it." Cuz people are doing
37:25>> Oh, yeah. I did see that. Yeah.
37:27>> dumping data into Claude and going,
37:29"Audit this." And
37:31the the way I like to explain this is
37:34AI the output from AI is absolutely just
37:38a function of what you put in, right?
37:40And if you don't put in much effort or
37:42much it
37:44I like to picture it as this universe
37:45full of millions and millions of
37:46planets. There's this huge big planet in
37:49the middle that I call planet average.
37:51That's what you're going to get back.
37:52You're going to get an answer from
37:53planet average every time unless you
37:56take the time to push the AI out to the
38:00edges of the universe. There's a Think
38:02of it this way, there's an answer on
38:03every single one of those planets.
38:06But if you don't try very hard, you get
38:07the answer from planet average. And
38:08guess what? AI has been trained on the
38:10entire internet, all of Reddit, all of
38:12Twitter. That wasn't a great idea. And
38:14it's going to give you a meh, very
38:16average answer. But if you add all of
38:20the context about that client, if you
38:22add all of your 10 years of wisdom, all
38:25of these transcripts, all of the ah, but
38:27we do it this way because of this and
38:28the you add all you start to push it out
38:31to the edge and you get a very, very
38:33different, unique answer that no one
38:36else is going to be able to get or
38:38replicate because that's your wisdom.
38:40It's your way of doing things. You've
38:42just codified that. We've been codifying
38:45stuff for years. Nobody looks at the
38:48financial reporting, oh, you used Excel
38:50to do that.
38:52>> No.
38:52yes, I did.
38:53>> Because
38:54>> [laughter]
38:55>> Because it would take me about 14 weeks
38:56longer if I didn't. So, yes, of course,
38:59as you say, of course I used AI to help
39:02with that, but AI didn't do
39:05all of that. It's just one of the tools
39:07that we used.
39:08>> I like the way you frame it.
39:09What have you seen? So, all right, the
39:12someone who's like
39:14totally like, I've only used like chat
39:17or cloud AI and just the the chat
39:19interface.
39:20So, you said, hey, go to 8020skill.com,
39:23get to level two, right? And then start
39:25start learning at least and and and
39:27seeing what the power is there. I I'm
39:30saying everyone should just go buy the
39:32brain and your course. Now,
39:35so, if you haven't played around with
39:37any of this, just go do those things.
39:38But, those people that are dabbling, um
39:42are already learning, right? What are
39:44you seeing in the group right now
39:47uh in regards like case studies or
39:49outcomes where they are now automating a
39:53large part of what they might do with
39:55managing Google Ads accounts or they're
39:58an agency and they're automating a large
40:00portion of, you know, how they run their
40:02agency.
40:04>> Um oh, there's so many.
40:06Um
40:07Benny comes to mind immediately. Benny
40:09started playing with the brain at
40:11Christmas
40:12and within 3 days replaced Motion.
40:15So,
40:17most of us use a software tool, be it
40:19ClickUp, Asana, Notion, HubSpot,
40:21something.
40:22But we have this nagging feeling that
40:24I'm only scratching the surface. I'm
40:25only using 10% of it. I remember
40:27thinking that way about Infusionsoft. I
40:29remember going to a conference thinking,
40:30I only use 10% of this Infusionsoft
40:32thing. And then the guy on stage, who's
40:36a genius at it, says, "I'm only using
40:3810% of this thing." I'm like, "Oh,
40:40I'm only using 1% of it if he's using
40:4110." Like, but I'm paying 100%.
40:44Right. All using tools like that. And
40:46now, this isn't where you start,
40:48absolutely, but if you're really curious
40:50and you have a play, you can do that.
40:52You can replace Supermetrics. You can
40:54replace Asana. You can build it to be
40:57just the exact way that you want it to
41:00be, that just does the thing that you
41:01want.
41:03And it works for you.
41:06Um
41:07audits and creating, you know, audits
41:09and reports and all that stuff.
41:10Once you wire up the Google Ads API, and
41:13yes, you've got to put the guardrails in
41:15place, and
41:16it it comes down to trust. So, I'm I'm
41:18making a lot of assumptions here that
41:20we've built trust over time, right?
41:22Which you do by observing the AI, by
41:25looking very carefully at what comes
41:26out. But it's very similar here to
41:29working with a junior employee. We used
41:31to have this thing when we would uh
41:33hire maybe a VA for someone in the
41:35Philippines. Of me, we, you. Right? This
41:38is me doing it. Loom comes in here a
41:40lot. This is me doing it. I've recorded
41:42it. You've got the transcripts. You can
41:43go back and watch those. And now we're
41:45going to do it we Well, we'll do it
41:47together, and we'll see if you hit
41:49roadblocks. And once you do it perfectly
41:51five times in a row, now we trust you.
41:53Now you can do it. You need to follow
41:55the checklist every time, please, which
41:56of course nobody does. But now you're
41:58doing it because we've built the trust,
42:00and we have some sort of
42:02observability. We've got some dashboard
42:04somewhere. We can see that people are
42:07following the systems, doing things
42:08right.
42:09And then it becomes what Gerber used to
42:11call management by exception. Like I if
42:14everything's a green light, I don't need
42:16to know about it as the manager. I just
42:18need to know when things go wrong or are
42:20going wrong.
42:21>> Right.
42:21>> Let's bring that to my attention and
42:22then we can problem-solve that.
42:25So you've got to have this sort of
42:26observability. So assuming that's there,
42:28you observe it, you build trust, then
42:30you kind of jump to the next layer of
42:33the stack. You start to abstract away
42:35all of the the crap. You don't download
42:37CSVs anymore and pull them into Google
42:40Sheets and delete the top two rows and
42:42change the formatting. All of that stuff
42:45just gets done automatically. And once
42:46you're doing that,
42:47we move up a level. I've
42:49Maybe this is a good way to explain it.
42:50I've had this sort of
42:52chart in my head for 20-odd years.
42:55And it goes It's very easy. Five words.
42:56But data, information, insight, action.
43:01I'm going to cheat. Expected business
43:02outcome. It's not one word. But let's
43:04just call it outcome. Right? And you you
43:06move You want to move up that stack.
43:08>> Yeah.
43:08>> Scripts were a wonderful way of going
43:10from data to information. Google Ads
43:12scripts. That was the automation piece.
43:15We take this data and we
43:17massage it in these ways and we format
43:19it like this and we output it over
43:20That's automation, not to be confused
43:23with AI. Just that AI's really good at
43:25writing the automation, at writing these
43:28little things to go talk to APIs.
43:32Then you go from information to insight.
43:35And that's where a lot of people would
43:36start with AI.
43:38Here's a bunch of data and you ask
43:40questions of it and it pops back with
43:42insights.
43:43To begin with, most of them are from
43:44planet average. They're a bit meh.
43:47But the more Again, the more context you
43:50add, the more information you add about
43:51that client, about their budgets, about
43:53what's happening on the website, about
43:55when tracking broke, you bring all of
43:57that to the table and now the answer
43:59gets better and better.
44:00But we're not ready to take action yet
44:04because we don't trust it. But once it
44:05starts giving us really, really, really
44:07good insights
44:09and we're saving a bunch of time, now we
44:11can move up the stack and we can
44:13actually have it take action. And I'm
44:15sure you've experienced this. Once
44:16you've got the Google Ads API working in
44:18the brain,
44:19you can chat to the brain in plain
44:21English or whatever language it is that
44:23you speak
44:24and say, "Oh, find me the best ad from
44:27the best ad group in that campaign and
44:29make a copy of it and put it in that ad
44:30group." And it will just do it.
44:33And the first few times you do that, you
44:35go open up the interface and you're
44:36like,
44:37"Holy it actually
44:38it actually did. Um
44:40I Okay, try this one, brain. I've got
44:42this list over here in this spreadsheet.
44:44Here you go, here's the link to the
44:45spreadsheet. Um can you add all of those
44:47as exact match negs to that list over
44:50Oh, it's already done it.
44:52Wow." And you just
44:54the more you see it do, the more you
44:55trust it, the more it doesn't go wrong,
44:57and you just keep
44:59doing more and more. And then, yeah, you
45:01can do things like replace software or
45:04build entire new services. I think
45:07there's going to be a huge Well,
45:10not that going to be, there is this huge
45:11opportunity right now
45:14to help your
45:17clients, those small businesses that are
45:20feeling really overwhelmed. And And
45:23Yeah,
45:24we're digital agencies, right? I'm I'm
45:26not anymore, but um
45:28Listen to this. We understand this world
45:31and
45:32small businesses look to you and say,
45:35"Well, you're the digital guy. You're
45:36the computer person. Like,
45:38you must understand this cuz we sure as
45:40hell don't in the business. We're like,
45:42yeah, we've played with ChatGPT, but no,
45:45we are behind. We don't know where to
45:46start." There's this huge opportunity
45:49there for you as the agency to help them
45:53understand it, to help them use it.
45:56And there may well be a an ongoing
45:58retainer thing there where you build
46:01something like the brain for them
46:04>> [clears throat]
46:04>> and then help them get set up and
46:06configure it all and do all the plumbing
46:08and hook it up to their
46:10email and their calendar and and then
46:12build skills for them.
46:14And say, "What is the stuff that you
46:16guys do
46:17all the time? Where does all of the time
46:19go? What are those things that you would
46:21love to do if you had the resources and
46:24the time, like you said before, but you
46:27just can't do now? Let me help you do
46:29that. Let me help you use it in the
46:30right way. And I think the tool for
46:32that, and this is why I built 80/20
46:34Skill,
46:35is going to be Co-Work.
46:37Because it's so much easier to use
46:40with almost all the power, because no
46:42small business is going to get all of
46:44their team to use Terminal. That's just
46:45never going to happen, right? It's hard
46:47enough to get all of your team inside
46:50You mentioned a team of 20, right?
46:52There's going to be people in that team
46:53of 20 that are just like, "There is no
46:54freaking way I'm ever opening that.
46:55[clears throat] I'm not going to work in
46:56there. Nah-uh."
46:58But,
46:59this huge benefits if you do. Co-Work is
47:01this beautiful sort of middle ground.
47:04But, to get the most out of it, you need
47:06skills. Hence, 8020skill.com. So, that
47:09you see it and you go, "Oh, wow." Now, I
47:11don't expect most people, or even most
47:13people in your agency, to go build their
47:15own skills.
47:17But, if you have somebody like you, like
47:19Luke, and you'll probably have two or
47:21three others that lean into this and go,
47:23"I just want to spend all my time doing
47:25this stuff."
47:26that build skills for everybody else.
47:30Now, if you can build that shared
47:32knowledge base, that shared brain, which
47:34is what I'm building at the moment. I've
47:36got a bunch of agencies beta testing
47:38this at the moment. It's due to roll out
47:40in the next week or two. And that'll be
47:42free for everybody inside Ad Astra AI.
47:44I'm not charging more for it.
47:46But, the team brain, where you have that
47:48shared context, cuz if one person's
47:50going to write up all of the notes about
47:51a client once, you don't want everybody
47:54going off and doing that. Or, you don't
47:55want those notes sitting on their
47:56computer and when they're on holiday,
47:58nobody can access it. And when somebody
48:00improves that skill, that audit skill
48:02that you've got,
48:03and somebody adds a an AI max section to
48:06the end of that, you want that to be
48:09there
48:10for all audits, if required.
48:13Every more, you don't want that new
48:14skill to be over there and version three
48:16to be over there and you've got version
48:17seven and that's just a mess, right? But
48:19if all of that's there, so now imagine
48:21taking that
48:23and building that for clients. Maybe
48:26they've got that scout internally that
48:28can build skills over there internally
48:30for their team, but maybe you, the
48:32agency, become
48:35the skill creator for them. You
48:37interview them about their processes,
48:39you get them to record looms and send
48:41you the loom and the transcript. You
48:43have a process for dropping the
48:44transcript and the loom and everything
48:46into a system and building a skill and
48:48delivering that skill back. That's
48:50hugely valuable right now.
48:53Most agencies could charge thousands and
48:55thousands per month doing that.
48:56>> Well, I've already seen some that are,
48:58you know, their retainers are pretty
49:00high, like 10 $10,000 plus retainers on
49:03some of the
49:04>> will really value it and others won't.
49:05Others will give that away for a couple
49:06hundred bucks and some people will
49:07charge five figures. It's just how much
49:10you value the work that you do, but the
49:13it it reminds me so much of when I got
49:15into Google Ads, which we we won't we
49:17won't mention the year, but it was a
49:18long time ago. Um,
49:20they want it and they need it. That was
49:23when I started the agency because all
49:25these people were coming to me for a
49:26year going, "Mike, we need you. Stop
49:28telling me like like you're just here to
49:30help and like like I want to give you
49:31money. Please, can you do this for me?"
49:33Because I was trying to teach everybody.
49:35And nine out of 10 people would say,
49:37"Mate, I don't give a how it works. Can
49:39you just do it for me, please?"
49:40>> please. Do it for me, yeah.
49:41>> Yeah. [laughter]
49:42>> Yeah. Yeah. And they it's going to be
49:44very similar with this. Like I don't I
49:45don't want to learn how to build a skill
49:47and how to connect this and API what?
49:51API what does that I don't care it
49:53stands for. Just set it up for me. I
49:55want to come in
49:57to 13 drafted replies, not a messy
50:00inbox. Also, I want
50:03>> Uh sorry to cut you off. I'm I'm already
50:05testing, hiring. Uh
50:07he's on a bit of a you know,
50:10probationary test period. Uh
50:13it's essentially an AI integrator within
50:15Grommets because from an organization
50:18and he knows Google Ads, too. So, it's
50:20like I found, you know, some that cuz
50:22like it's one thing if you know AI, but
50:24then if you don't know Google Ads,
50:25there's like a lot of training there.
50:27So, uh he knows Google Ads. He's gone
50:29through my training on Google Ads. And
50:32so, it it's it's worked out so far. But,
50:35um
50:37it's what like from an
50:38it's
50:40from an organization for Grommets. Like
50:42for me right now, and so for us here at
50:44Grommets, it's very siloed still. Luke's
50:46off doing his thing. I'm kind of doing
50:47my thing. I've shared some skills like
50:50the negative keyword thing I built, like
50:51I shared that with my school group. And
50:53I showed like, "Hey guys, if you don't
50:55have cloud code and it's too nerdy,
50:57here's how you just dump a CSV, use this
50:59skill inside of a project inside of
51:02cloud, right? Through chat." So, just go
51:03play with that, right? Now, if you're
51:05nerdy and you've got this integrated
51:08with the MCP or an AI integration, then
51:11or API integration, then here's how you
51:13would just do this inside cloud code and
51:15and automate it. So, I I kind of showed
51:17and I just shared here's the the full
51:19skill,
51:20>> Right.
51:20>> Yeah.
51:20>> right? From our standpoint, what we have
51:24found right now, and what we're dealing
51:25with is A, how do we deploy
51:28all of this to the team cuz there's like
51:29a huge training piece there.
51:31>> Mhm.
51:32>> It's would require a lot of time for me
51:34or Luke. And I don't know if that makes
51:37complete sense and that I'll I'll I'll
51:39lead into a question here then, too. Um
51:42there's probably a million questions
51:43from all of this. But, the other thing
51:45what we have found is, you know, Claw
51:48and it sounds like you're working on
51:49this with a shared brain concept, but I
51:52cannot give the API integration into
51:56Claw code to my entire team. That
51:58wouldn't make sense. I like from a
51:59security standpoint, I'm not going to
52:01give them access to the entire MCC. So,
52:03our strategists all work within sub
52:05MCCs. Now, the problem with
52:09the MCP is it does use a ton of tokens.
52:14>> Mhm.
52:15>> And so, we've ran into that. Now, Luke's
52:17working on a way of figuring that all
52:20out so it doesn't just burn through
52:22usage just when you're trying to pull
52:23like, you know, some data
52:25>> common.
52:25>> from one account. And so, I'm we're
52:28still at the stage like I'm playing with
52:30it. I've got skills. I can easily spin
52:33out audits and then just sharing that
52:35with the team, but it's still it's this
52:37siloed thing and it's not totally
52:39organized. And then I saw I'll give him
52:42a shout. I forget his last name, Daniel,
52:44who owns a Brain Labs Digital.
52:47>> Daniel Gilbert, yeah.
52:48>> Yeah, incre- I don't I don't know him.
52:50I've never talked to him. Incredible
52:51what he's built. Incredible guy. I
52:54Again, I don't know him, but I've seen
52:55him on his I follow him on LinkedIn. And
52:58he came out with this really cool post.
52:59I'll probably put that in the
53:00description of this video below, too.
53:01You guys should all check out. But how
53:04he has built I mean, this is a thousand
53:07plus person agency. So, the you know,
53:09this is not a small thing that they just
53:11accomplished, but they're using
53:13essentially Notion as the data layer, it
53:15sounds like, for their their entire
53:17organization in regards to Now, how how
53:20they hook it all up with like a thousand
53:22employees and all that, I'm not entirely
53:24sure. It's He doesn't give you every
53:25single step, but the infrastructure that
53:27he's developed within a that big of an
53:29organization was very cool to see like a
53:32true case study of it. Cuz you hear you
53:34hear all the hype
53:35>> Yeah.
53:36>> like, I automated my agency and fired,
53:38you know, half my employees and all you
53:40know, I'm not buying most of that. I
53:42think a lot of that is just like people
53:44flexing for for clicks and impressions.
53:47Um and and uh and I like his standpoint,
53:50too. He's like, "If I can get a 20%, you
53:53know, more output per employee, I'm not
53:55looking to decrease my workforce by
53:5720%."
53:58>> Not at all.
53:59>> Not at all, right? I would rather hire
54:01someone and get 20% more out of them as
54:03soon as I hire them.
54:05>> and then we'll get 30% next year, and
54:06then 50% the next year. And this what a
54:09lot of businesses will do, they're
54:10freezing hiring. Not great for juniors
54:13coming in, that's a whole different
54:14podcast. Um
54:17but yeah, if you can but I think 20%
54:19will be the table stakes efficiency gain
54:22because all of your competitors or some
54:25of your competitors
54:26will be increasing 20%, 30%, 40%.
54:29A whole bunch won't. A whole bunch will
54:30stick their head in the sand and go,
54:32"This whole AI thing, it's rubbish. It
54:34just tells dad jokes badly, and I'm just
54:37going to ignore it." And they may not be
54:39in business in two or three years' time.
54:41I think 20, 30% is table stakes to stay
54:44in the game.
54:45But yeah, Dan's awesome. Um every time I
54:47used to go through London to visit my
54:49folks, um I'd pop in and see him. When I
54:52first saw him, he had a 60-person team.
54:55Uh by the time I went back the next
54:57time, it was 200, 250. Like he is a
55:00unicorn. Very few people can grow that
55:03fast.
55:04>> Incredible. I would love I would love to
55:05talk to him. I I I might hit him up
55:07after when I when this is out. Say,
55:08"Hey, we gave you a shout-out." But I
55:10would love to talk to him because it's
55:11what he's built is just truly
55:12incredible. And you know, I I run a
55:1420-person team, and it's hard, right?
55:16You know, so I can't imagine getting to
55:17a thousand people.
55:19>> So, a couple of things on that. Um yeah,
55:21so they chose Notion because that was
55:23where the work already lived. Because a
55:25huge, huge part of this, which isn't on
55:28the LinkedIn posts, is behavior change.
55:31And trying to change the behavior of a
55:33thousand people
55:35is really bloody hard. I've never tried
55:37to do it, but I hear it's really, really
55:38hard. It's hard enough to change the
55:39behavior of 20 people. So, the less
55:42behavior change, the better. Meet them
55:44where they are,
55:46make things better, build stuff for
55:48them, and they'll come along. A few
55:50won't, and okay, that's okay. We may
55:53need to have a conversation,
55:54but most will. The model I use, cuz as
55:57you can see, I like mountains, Whistler.
55:59Um
56:01is summit scout trekkers. This is the
56:03framework I've sort of the mental model
56:05I've created for this. So,
56:07summit, that's your job. Leadership's
56:09job is to to what is the view going to
56:11be from the top of the mountain? Let's
56:13go out a year or two. There's no point
56:15going any further than that.
56:17Why? Why are we doing this? Why is AI so
56:20important to the business? Why is it an
56:22existential risk? What is this going to
56:24be like?
56:25We are not doing this so we can get rid
56:27of half of you. Like, people need to
56:29hear that because
56:31people are worried. They're worried
56:32they're not using AI correctly. They're
56:34worried they're breaking some sort of
56:35policy that they don't know about. They
56:37don't even know if they're supposed to
56:38paste client data in there or not
56:40without redacting it first. Like,
56:43why are we doing this, boss? Are you
56:44doing this so you can get rid of half
56:45the people so you can pay me the same
56:47and I do twice the work? That's what's
56:49going through people's heads. So, it's
56:50got to start with that vivid vision of
56:53this is why we're doing it and this is
56:55why we're resourcing it properly. This
56:57is why we're going to give you each
56:59Well, we're not going to count tokens,
57:01but why we're going to give you access
57:03to all of these tools and the proper
57:05training that goes with it.
57:07Then, scouts. You figure out who the
57:10scouts are in the business. Sounds like
57:12for you, with 20 people, it's probably
57:14you, Luke, and two or three others. It's
57:16generally somewhere between 10 and 30%
57:18of the business.
57:19Those are the people that are like, I
57:21would love to play with that. It's the
57:23super curious. It's the high agency. I
57:25wish there was a hiring test for high
57:28agency. I would pay a a for that test
57:31and only hire those people. If I think
57:33back, I had the very, very good fortune
57:37of hiring some incredible people. The
57:38vast majority were incredible.
57:41The ones that were
57:43those incredible people were always the
57:44ones high agency, not "Boss, I've got a
57:47problem. Can you fix it for me?"
57:49"Boss, I had a problem. I tried these
57:51three things. This was the best way. I
57:52think we should do this from now on."
57:54>> Yes.
57:54>> Awesome. Thank you very much. Right?
57:57>> So, you figure out who the scouts are.
57:59And then the scouts and someone from
58:01leadership has to be a scout. You can't
58:04delegate this. You can't abdicate this
58:06and just say, "Right, you guys are the
58:08scouts. Go figure out how to completely
58:10change our business." Dan Gilbert is in
58:12the weeds doing this stuff with those
58:15scouts on his team.
58:17And then the scouts figure out how to go
58:19from base camp to camp one.
58:21Once we get to camp one, once we've
58:22paved that path, it's now on you,
58:25leadership, to go back and get the third
58:27group of people who are the trekkers.
58:30And that's the 80% of your business who
58:32aren't going to open a terminal window,
58:34who aren't going to build a skill from
58:36scratch, but are quite happy if the
58:38paved road is in front of them. They're
58:40quite happy to go, "This way? With Okay.
58:43I
58:44Let's go." Meanwhile, the scouts are off
58:47figuring out how do we get from camp one
58:49to camp two and building that path. And
58:52then rinse and repeat. So, that's the
58:54mental model that that works. You
58:56mentioned MCPs. I know we're getting
58:58into geeky territory, but you also said
59:00the word nerdy, and I want to come back
59:01to that word cuz I think it's not if you
59:03are nerdy. I think there's too much of
59:05this that's
59:06if you're technical, you can do blah
59:08blah blah. Technical Being technical has
59:10got nothing to do with I don't think of
59:12myself as technical.
59:14I I realize the average person in an
59:16agency probably looks at me and goes,
59:17"Yeah, but you write scripts and you
59:18play in terminal. You're a geek."
59:21Probably, yes, but I don't consider
59:22myself technical. A proper technical
59:24person, a proper software developer,
59:26would look at what I'm doing all day
59:27every day and go,
59:28"You're a numpty. You haven't got a clue
59:30what you're doing."
59:31But it's creating a ton of value. So,
59:33there's that. So, I don't think you have
59:34to be technical. I think you have to be
59:35curious, and you have to be high agency.
59:38And then, just to cover the thing on
59:39MCPs, yes, I kind of use So, an MCP for
59:44if you're still playing at home and
59:45you're still listening to me an hour in,
59:46good on you.
59:47Um
59:48model context protocol,
59:50something that Anthropic created over a
59:53year ago.
59:54I think there was a very early version
59:58that is kind of morphed into skills.
1:00:00Custom GPTs,
1:00:02vaguely similar, but they're kind of
1:00:03going away. That was an early early
1:00:05version.
1:00:07An MCP is essentially a way to connect
1:00:10to something else, usually some form of
1:00:13data. So, Google Ads, thankfully, have
1:00:16finally built their own MCP, and it's an
1:00:19easier way to connect to something
1:00:22instead of writing all of this API code,
1:00:25often.
1:00:26It's a way for AIs to talk to other
1:00:29machines. However, as you said,
1:00:32tokens.
1:00:33Uh and I don't want to go too deep into
1:00:35the weeds on this and talk about context
1:00:37windows and stuff, but most Google Ads
1:00:39reports
1:00:41have a lot of data. Well, if you're
1:00:43trying to pass that data through a
1:00:46conversation,
1:00:48that is not a good way to do it for two
1:00:50reasons. One, the amount of
1:00:53words, numbers, data that you can have
1:00:56in a conversation is limited, but two,
1:00:59it's called an LLM. It's a large
1:01:01language model. We don't want a language
1:01:04model to be doing maths. It's not what
1:01:06they're good at. They're good at
1:01:07predicting the next word. They're not
1:01:09good at doing maths. Much better than
1:01:11they were,
1:01:12but they're not good at maths. They're
1:01:13good at writing code that can do maths.
1:01:17So, they can write a little Python
1:01:18script, whatever the hell that is, that
1:01:21can do maths. It's the way geeks used to
1:01:24do or still do do data analysis, right?
1:01:27They do stuff in R and Tableau and they
1:01:30build code to analyze data. Us normal
1:01:34humans, we don't do that. We open up a
1:01:35spreadsheet and we go, "What's going on
1:01:37here?"
1:01:38So, MCPs not a good way to deal with
1:01:40most Google Ads data. If it's a quick
1:01:43question, which campaign spoke spent the
1:01:44most yesterday,
1:01:46great, use an MCP. But, if you're saying
1:01:48go grab me the search term report the
1:01:50last 90 days,
1:01:51it's not good because that's a lot of
1:01:53tokens. It's a lot of numbers. So, the
1:01:55way to do it there is to use the API
1:01:58instead. So, you can go grab data. So, I
1:02:02I want to make it like
1:02:03really relevant to you cuz that will
1:02:05make it more real and concrete for
1:02:06listeners.
1:02:07>> Yeah.
1:02:08>> You've got a bunch of pods, a bunch of
1:02:09teams in different sub-agencies, right?
1:02:11So, you might create up the top here
1:02:14some sort of automation that
1:02:16every morning at 4:00 a.m.
1:02:19goes and pulls the data using the API
1:02:22and saves that somewhere, probably as a
1:02:25CSV, so that it's there and ready for
1:02:28the team to use and they can only see
1:02:30the data that they are allowed to see.
1:02:33>> Right.
1:02:33>> Whether you use We probably wouldn't use
1:02:34Notion for that, but whether you use
1:02:35Google Drive, whether you use The Brain,
1:02:38or whether you use BigQuery, who knows,
1:02:41right? But, data is stored somewhere.
1:02:43Here's the data that you're allowed to
1:02:44see. Now, you can go do stuff with that
1:02:46data. You can pull that data into a
1:02:49skill that we built and gave you. You
1:02:51can pull it into an audit. You can pull
1:02:53it into the
1:02:55negative keyword tool that we built and
1:02:57gave you and do the work on it.
1:03:00And because we're still building trust,
1:03:02the tool is going to give you some
1:03:05output. Maybe it's a file formatted in
1:03:08Ads Editor format. You can just drop
1:03:10that into Ads Editor.
1:03:12Or go to a bulk upload. Or maybe it's
1:03:14going to give you a Google Sheet with
1:03:16some ad copy. And we're not quite ready
1:03:19to have this automatically change ads,
1:03:20so we're going to copy and paste the ad
1:03:23copy from the sheet into the ad account
1:03:26cuz that's what we're comfortable with
1:03:27yet.
1:03:28But once you build the trust, you go up
1:03:30a level, and then we get to the point of
1:03:33yeah, just make all those changes for
1:03:34me. Just take that and do that there. As
1:03:36long as we've got some guardrails in
1:03:38place
1:03:39and some observability, we can check
1:03:41that it has done the thing that we asked
1:03:43it to, and we build the trust. We keep
1:03:46moving up a level, and then the team are
1:03:48going, "This is great. This is more fun.
1:03:51I'm not copying and pasting crap."
1:03:53That's, by the way, a great place to
1:03:54look for anyone listening to this that
1:03:56hasn't really
1:03:57done much with AI at this point. Where
1:03:59are you copying and pasting in your
1:04:01business? That's a great place to start
1:04:04because there's
1:04:05a huge time saving there.
1:04:07It's just integration. It's just
1:04:09computers that need to talk to other
1:04:10computers. You're copying stuff from
1:04:12Excel and putting it in Xero. You're
1:04:14taking stuff out of HubSpot and Google
1:04:16Analytics and putting it into a sheet.
1:04:18All of that stuff, start with
1:04:20automation. We don't need AI for that.
1:04:23AI can help create the automation. It
1:04:25can even help suggest the automation.
1:04:28Record that thing. I love that you use
1:04:30Loom and have been for so long and
1:04:32record everything.
1:04:35There was this wonderful YouTube. I'll
1:04:36try and find the link for you. It was
1:04:38between Tim Ferriss and the CEO of a
1:04:41company called Levels.
1:04:43And his name began with a C, but I've
1:04:44gone completely blank on that.
1:04:46It was about 3 hours long. This guy has
1:04:48four EAs just for himself.
1:04:51>> Wow.
1:04:53>> But now think of that as agents
1:04:55>> Yeah.
1:04:55>> or a brain or an AI. Like, what am I
1:04:59doing that I don't need to do anymore?
1:05:02And the question I'm asking myself more
1:05:04and more these days is why am I doing
1:05:06it? Like, what's the trigger?
1:05:07When I get this email from Stripe, I go
1:05:10and tell the brain, "Hey, can you do
1:05:12this and add that person there and
1:05:14deliver this?"
1:05:15Why am I doing that? It's just Surely
1:05:17you Hey brain, can you see this? Can you
1:05:19Yeah, I've got access to Stripe. I've
1:05:21got access to Gmail. Like, great. When
1:05:23this happens, do this.
1:05:25Great. Add it to my dashboard so I can
1:05:27see that you're doing it and I can see
1:05:28that you've done it right and you
1:05:29haven't messed up. I've got
1:05:31observability. I build trust and we move
1:05:33on to the next thing.
1:05:34>> Do you
1:05:36So, going back to like founders
1:05:38[clears throat] that need to be involved
1:05:39like you said, "Hey, you have a
1:05:4120-man team
1:05:43and you you and Luke probably should be
1:05:46a part of that, right?" And
1:05:48you you should be leading this along
1:05:49with probably a a couple of others. Now,
1:05:52one thing that I have started to see and
1:05:57I've you know, I I I can see where
1:06:01some people will catch
1:06:04the bug that this will give you, this
1:06:06like lightening of inspiration and you
1:06:09get hooked on this, right? But, I'm also
1:06:13observing and seeing what I would say
1:06:15are a lot of people playing with a new
1:06:17shiny object and they're spending now,
1:06:22you know, their full week or more than
1:06:24half of their week on it. So, there's
1:06:27like In in In the way I'm looking at it
1:06:29is
1:06:31Okay, I know this is important, right?
1:06:34So, we have to work on it. But, what
1:06:36time allocation
1:06:39makes sense for me personally or
1:06:43my Luke, the the the you know, my
1:06:46business partner to be involved because
1:06:48I see a lot of small smaller agencies or
1:06:50smaller than us, um probably even, you
1:06:52know, those that are bigger than us,
1:06:54where it's like, if I looked at their
1:06:56P&L sheet and I looked at their sales
1:06:58volume and retention, I would say, you
1:07:00know, it's probably not that smart to
1:07:03only focus right now on AI when you
1:07:07would need more sales volume and you're
1:07:09doing nothing to generate that and your
1:07:11retention's not really good and you're
1:07:13doing nothing to fix that right now.
1:07:14Now, I know learning AI is going to
1:07:18absolutely help you, right, in the
1:07:20future, but you've got to exist to the
1:07:23future.
1:07:25>> Uh,
1:07:26that's a great point and a really good
1:07:28question and I've got so many different
1:07:30answers bouncing around my head.
1:07:32Um,
1:07:32there's a line from Dan Gilbert I want
1:07:34to quote, theory of constraints and
1:07:38useful.
1:07:39Theory of constraints, right? Your
1:07:40business is a pipe
1:07:42and
1:07:43some [clears throat] bits of the pipe
1:07:44are wider and some bits are thinner.
1:07:46There is no point spending any time on
1:07:48the wide part of the pipe making that
1:07:49wider when you've got this thin part of
1:07:51the pipe because you make this part of
1:07:52the pipe wider, doesn't matter that more
1:07:54stuff can flow through there, you've got
1:07:56a thin bit of the pipe, that's the
1:07:57bottleneck, that's the constraint.
1:08:00The thing to work on in your business
1:08:01right now is the constraint. For so many
1:08:04businesses,
1:08:06for businesses out there, that
1:08:07constraint is we just need more. We need
1:08:10more customers, we need more sales, we
1:08:11need more money in the bank and that's
1:08:12why Google Ads has exploded over the
1:08:14last 20 years, that's why our agencies
1:08:16have been successful because
1:08:18it works so much better than Yellow
1:08:19Pages and letterbox drops and radio
1:08:21advertising. Like that is the best way
1:08:23to get more.
1:08:25But
1:08:26it's funny, when I was an E-Myth coach a
1:08:28long time ago and trained with Gerber
1:08:30and his team,
1:08:32everybody thought the first module would
1:08:35be let's go get more. It would be lead
1:08:37generation, but it wasn't, that was
1:08:38module five.
1:08:40You started off with mindset and systems
1:08:44and all of this stuff that prepared you
1:08:47for the point
1:08:49where leads would explode, but
1:08:52if you explode leads in most business,
1:08:53I'm sure you've done this for businesses
1:08:55in the past, right? You double or triple
1:08:58the amount of business coming in,
1:09:00what breaks? Cuz something breaks really
1:09:02quickly when that happens and you get
1:09:04the phone call going, "I don't know what
1:09:05you guys are doing, but can you turn it
1:09:06off?" Because the guys on the factory
1:09:08floor can't keep up. That's a wonderful
1:09:09phone call to get, but you just lost a
1:09:11client. But where's the constraint in
1:09:13business?
1:09:15Dan had a line in that article that
1:09:16you're going to put in the notes, the AI
1:09:19native thing, of like, "This is the most
1:09:22important thing in our business right
1:09:24now, and any time that I'm spending
1:09:26doing anything else, I feel it's
1:09:28wasted." But he's got a thousand-person
1:09:31organization. I met Ken once, his head
1:09:34of
1:09:35uh acquisitions. I won't say why.
1:09:38Um
1:09:40amazing guy, but he's got this
1:09:41incredible team around him that are
1:09:44helping run the day-to-day business. His
1:09:46job is just to sit on top of that
1:09:49mountain and figure out the vision.
1:09:51That's basically it. Yeah. He probably
1:09:54hasn't touched an account in years. He
1:09:56probably does the odd sales meeting and
1:09:58brings in some whales, but that is his
1:10:01job. For a smaller, typical
1:10:0410-to-30-person agency,
1:10:08the reason I said you need to be a scout
1:10:10is certainly
1:10:12one of you two. So, if you've got two
1:10:14business partners, if you've got a
1:10:15management team, at least somebody from
1:10:17that top level needs to be one of the
1:10:20scouts, because you can't just abdicate
1:10:22it and say, "Go figure it out."
1:10:24>> Sure.
1:10:24>> Because you have to be aware of the
1:10:27limitations of how it's going to be
1:10:28used, of what it can do now. Just to to
1:10:31build that intuition, to build that
1:10:34feeling of "I think this is the right
1:10:36path to go down."
1:10:38Roughly where are things going to be in
1:10:40six months' time. I think it's really
1:10:42important to build that intuition.
1:10:45None of us are going to be right 100% of
1:10:47the time with that, but I think it's so
1:10:49important to the business that you owe
1:10:51it to yourself and the business and the
1:10:53people that are depending on you to
1:10:55build that intuition for this stuff, to
1:10:58have a feel for it, not just read a
1:11:01bunch of LinkedIn posts and go, "We're
1:11:03going this way." Like, my team used to
1:11:05suffer from what I used to call um
1:11:07entrepreneurial whiplash.
1:11:09Which is like, imagine you're the head
1:11:11of the snake and like, "We're going this
1:11:13way." And the whole business has to
1:11:14orientate behind you to go this way. And
1:11:17then you come in on a Monday morning and
1:11:18go, "No, no, we're going this way." And
1:11:19the whole team's like, "Whoa,
1:11:21okay, now we're going over here." And
1:11:23there's this like
1:11:25And then
1:11:26oh, actually no, I don't like that way.
1:11:27No, we're going this way again. Ah, and
1:11:30it's just like, yeah, the people at the
1:11:32end of that long line are just getting
1:11:33swept out into the distance every time
1:11:35you It feels at the front of that line
1:11:38like that's easy, I just change from
1:11:39here to here. But you're you're turning
1:11:41an oil tanker, you're not turning a
1:11:42speedboat. It takes time for that
1:11:45to move around. So, I think you have to
1:11:47be
1:11:49experiencing that and resource it
1:11:51properly. Like, I am so sick of hearing
1:11:54of agencies like, "Oh, yeah, we've got a
1:11:56we've got a chat GPT account. Everybody
1:11:57logs into the same thing. They they
1:11:59share all that. Yeah, we're spending 20
1:12:00bucks a month. What what more do you
1:12:02want us to do?"
1:12:04I want you to buy a proper subscription
1:12:05for everybody on the team.
1:12:08So, that they can play and experiment,
1:12:10and they can take it home with them. One
1:12:12of the best ways to get your team really
1:12:13into AI is they take the subscription
1:12:16and go build something little
1:12:20that is meaningful to you. Something
1:12:22from home life, something with your
1:12:23kids. Maybe it's an illustrated bedtime
1:12:25story thing, or it's something to do
1:12:28with the Dungeons and Dragons game that
1:12:29you like to play on Tuesday night, or
1:12:32whatever. But, build that thing
1:12:36and then you're going to go cuz you'll
1:12:37you'll tip more time into that. You'll
1:12:39you'll spend a couple of hours in the
1:12:40evening building the thing that's
1:12:42meaningful to you, not building an
1:12:44automation for the boss.
1:12:46That's a wonderful way for the team then
1:12:48to experience this and say like, "You
1:12:50can use this. We've bought you a team
1:12:52license. Knock yourselves out. Use it as
1:12:54much as you can. Use it in the evenings.
1:12:56Please, go discover what this can do. Go
1:12:58play with some LEGO bricks.
1:13:00Because we have this amazing LEGO kit
1:13:02that we keep talking about. The good
1:13:04news is
1:13:06you can build anything that you imagine.
1:13:08The bad news is
1:13:10there's no manual that says put this
1:13:12brick here and this brick here. The only
1:13:14way you learn this stuff
1:13:16is to play with it and to experiment and
1:13:19to figure out what it's good at and what
1:13:20it's bad at.
1:13:21>> Do you see like in the future cuz there
1:13:24is a part of me, you know, for example,
1:13:26I spent 2 weeks building negative
1:13:28keyword skill, right? Now, would it take
1:13:30me 2 weeks now? Probably not cuz part of
1:13:32it was learning how to build skills and
1:13:34whatnot. But
1:13:36do you see and it's built off of, you
1:13:38know, my trainings and the way I think
1:13:41in how to do it. So, I do like how it
1:13:43works, right? But, you know, the reality
1:13:46is
1:13:47there's
1:13:48probably a thousand other Google Ads
1:13:51people out there that could build just
1:13:52as good of a negative keyword sort of
1:13:54skill, right? So, there's a part of me
1:13:56that thinks
1:13:57okay,
1:13:59you know,
1:14:00and it kind of goes back to where like
1:14:02the audit report that I
1:14:04that that's probably one of the most
1:14:06like time-saving useful things actually
1:14:08that I've [clears throat] done recently.
1:14:09And I'm like, "Wow, that's really cool."
1:14:10When When I've been looking at it more
1:14:12when I first started, it was like, "All
1:14:14right, here's all the, you know,
1:14:17work
1:14:18workflows and optimization checklist
1:14:20that we have. One by one, we got to go
1:14:22build skills for all of that." And And
1:14:24we are. We'll play around and do that.
1:14:26But part of me is like
1:14:28man, do most people need to go do that
1:14:30cuz like a year from now I could see,
1:14:33you know, Fred from Optimizer has like
1:14:35the perfect AI tool for all of those
1:14:38that's, you know, 80% there, good enough
1:14:40for most people who are like working
1:14:42within Google Ads accounts. It doesn't
1:14:44need to be this entire sort of
1:14:46ingredient playbook that your own brain
1:14:49has built. And so, I'm looking at now
1:14:52more of like like the audit thing that I
1:14:55did. Like to me, it's more like where
1:14:56exactly what can I do where I don't have
1:14:59the resources for with like my team?
1:15:02What can I just do now that I would love
1:15:03to be able to do? And I'm starting to
1:15:04create more there versus like individual
1:15:08Google Ads sort of like uh
1:15:11optimizations.
1:15:13>> We had this exact discussion in my
1:15:16mastermind group. So, I've got a a
1:15:17mastermind that's limited to 50 people.
1:15:20You'd be a great fit, actually. Um
1:15:22where we talked about this. And
1:15:26the upshot of that was kind of take the
1:15:28pressure off yourself a bit. Um
1:15:31not everything you build will get used
1:15:33by the whole team. Not everything you
1:15:34build will be valuable. But, the process
1:15:36of going through that And it wasn't 2
1:15:38weeks, right? It was on and off during
1:15:39the 2 weeks. People shouldn't hear it as
1:15:41like I didn't spend 100 hours on this
1:15:42thing. But,
1:15:45it took 2 weeks to get to the point
1:15:46where it was useful.
1:15:48But, all the skills you built after that
1:15:50would have been quicker.
1:15:52We have a choice. We could wait around
1:15:54for that year. And there will be
1:15:57incredible tools out there in a year's
1:15:59time.
1:16:00But, we'll be so far behind the curve.
1:16:03We'll be just
1:16:04grabbing things from LinkedIn and
1:16:06commenting on and downloading it
1:16:07and and stuck in the whole shiny object
1:16:10merry-go-round.
1:16:11>> Yeah.
1:16:11>> But, if you spend a little bit of extra
1:16:13time now And it is extra time. It is
1:16:15time on top of an already busy work
1:16:18week. But, this is the most
1:16:20transformative technology of our
1:16:21generation, possibly of our lives. And I
1:16:25think it is worth spending a little bit
1:16:28of extra weekend and evening time to
1:16:30understand it, to work out how to use it
1:16:32because
1:16:34it compounds.
1:16:36You learn those first few LEGO bricks.
1:16:38And then you start to understand how to
1:16:40put those together. And then you start
1:16:41to see that level up and like, "Oh, well
1:16:44if I can do that, maybe it could do this
1:16:46and this." And you learn how to ask the
1:16:49machine, "What can you do? Could you do
1:16:51this?" And you start to push it because
1:16:53this was the weird thing for me.
1:16:54I understood that my skills were going
1:16:56to compound. And like for those only
1:16:59listening, I'm sort of like drawing a an
1:17:01exponential curve with my hand here.
1:17:03I knew the skills were going to
1:17:04compound, and they start to build really
1:17:06quickly, right? Compound interest,
1:17:07you've seen that exponential curve. It
1:17:09gets really fun at the end of that
1:17:11curve.
1:17:12What I wasn't ready for is that my
1:17:14ambition
1:17:15would compound. I would start to push it
1:17:18further and further. I would start to
1:17:20ask for more and more. I would start to
1:17:22ask for things that I thought would be
1:17:24impossible. I built a CMS
1:17:28in 2 hours the other day because a
1:17:30member said, "I'm building all these
1:17:31websites for clients now. We're
1:17:33replacing WordPress with this thing
1:17:35called Astro, which I'd never heard of
1:17:37before,
1:17:38because the brain is able to build all
1:17:40of these pages in Astro. I've built six
1:17:43websites in the past year. I'm not a web
1:17:45designer, but I don't use WordPress or
1:17:47anything. I just talk to the brain. I
1:17:49need a new page. Can you build that? Run
1:17:51it through the copywriting tool. Make it
1:17:53better. Let me tweak it a little bit. Go
1:17:56build an image for that, please. Put the
1:17:57image in the right place. Now crop it.
1:17:59Resize it.
1:18:00Just with natural language.
1:18:02And
1:18:03this a few clients have been using
1:18:05Astro. Sorry, a few members have been
1:18:07using Astro to build sites for clients.
1:18:10Albut,
1:18:11now what do we do? We've got clients
1:18:14saying, "Can we make this change?" And
1:18:15like, "Okay. Well, let's let's ask the
1:18:17brain."
1:18:19Lots of people using Astro, could we
1:18:21build a CMS so that clients could log
1:18:24into this thing and be able to make text
1:18:26changes to the blog post or whatever it
1:18:28is.
1:18:30And it built it. That was like four or
1:18:31five prompts. It took less than 2 hours.
1:18:34It was
1:18:35big but but constantly
1:18:37pushing it. There's a wonderful,
1:18:39wonderful member, shout out to David,
1:18:41member of my mastermind group. He won't
1:18:43mind me saying that he's in his 70s and
1:18:46he will sit in his car while his wife's
1:18:48in the
1:18:49shop on a Saturday. He'll sit in the car
1:18:51park
1:18:52on his phone because he's not in front
1:18:53of his eight screens and he'll be
1:18:55building stuff from the car on a
1:18:57Saturday.
1:18:58He has built stuff that is going to
1:19:01change his industry.
1:19:03He's started advertising on
1:19:07Is it TV or radio? I think it was TV for
1:19:09the first time ever. Been in business 40
1:19:11years. Just started advertising on TV.
1:19:13And the TV guys were saying stuff to him
1:19:15about results and blah blah blah and and
1:19:17he knows Google Ads really, really well.
1:19:19So, he's like, "Yeah, but how do we
1:19:20measure the performance of this?" to a
1:19:22TV station. So, he goes home, fires up
1:19:24the brain and says, "This is what they
1:19:26said. This is what I want. This is what
1:19:28I want to be able to track. Da da da da
1:19:29da da." And the brain built him this
1:19:31incredible dashboard. He shows it to the
1:19:33TV guys and the TV station are like,
1:19:37"Um could could you build something like
1:19:39that for us that
1:19:42that we could use?" He's done the same
1:19:43with his accountant. He's built tools to
1:19:45show his accountant stuff and they've
1:19:46gone, "How did What did this cash flow
1:19:48forecasting that you built? How? How and
1:19:50how do we do that?" He's like, "I just I
1:19:52just ask.
1:19:53I just ask the brain.
1:19:55Like, don't be
1:19:57He said this to one of the other guys in
1:19:59the mastermind the other day. He said,
1:19:59"You're too smart for your own good.
1:20:02Just get out of your own way. Just ask."
1:20:04We tend to assume that like our way is
1:20:06the right way. Just ask the brain if it
1:20:08could do it. Ask it if it's the if
1:20:10there's a better way.
1:20:11Cuz once it has the context, it
1:20:13understands you, understands your
1:20:15business, understands your clients,
1:20:16understands the way you work, it
1:20:18understands what you built in the past.
1:20:20And if you give it enough information,
1:20:21it kind of understands what you need to
1:20:23do next. My entire to-do list, all of my
1:20:26projects are all in the brain. I can say
1:20:28to the brain, "What are the three things
1:20:30I should work on today?"
1:20:31And it said, "Well, at 9:00 you got the
1:20:33call with Austin and then da da da and
1:20:35it tells me how to structure my day."
1:20:38>> Yeah.
1:20:38>> If I wanted to.
1:20:39>> Yeah. It's uh you know, it's funny cuz
1:20:42[clears throat] we've got I'll I'll
1:20:44share a bit of a I can't share the whole
1:20:47screenshot. I'll share my screen here in
1:20:48a second but cuz it's funny we we were
1:20:51on Notion for a time but uh
1:20:54like you know,
1:20:55>> It's over engineered, isn't it? It's
1:20:56it's it's more than you need. I'm sure
1:20:58if you're a thousand person company,
1:20:59Dan, you need something like that. But
1:21:01the typical
1:21:02>> small business doesn't need a lot of
1:21:04that. Same with ClickUp, same with
1:21:05Asana. There's a lot of stuff in there
1:21:07that
1:21:07>> I'll have to I'll have my editor block
1:21:10some of this but like we've built now
1:21:12our and this is over the last year
1:21:15um
1:21:16so our own sort of operating system,
1:21:18right? And it's funny cuz now we've got
1:21:21uh
1:21:22our like this we just built Notion now
1:21:24and like we so we're getting rid of
1:21:25Notion cuz we just built Notion. And so
1:21:28now I do have I have developers. So but
1:21:31well where I see the future for us is
1:21:33like
1:21:35I there will be a period of time where
1:21:37the team is on
1:21:41Claude, whether it's co-work or if they
1:21:43want to run with code, that's fine. But
1:21:47like just as an experiment for fun, I I
1:21:51asked my one developer, "Can you build a
1:21:54campaign builder tool, right? Like can
1:21:56you just build a tool? We have an AI
1:21:59workflow process based off of like our,
1:22:01you know, research, keyword research,
1:22:03training and all of that." And sure
1:22:05enough, I did a YouTube video on it and
1:22:07shared sort of a
1:22:08a sneak preview of it to the group but
1:22:12we we have a full tool. It's not
1:22:14released yet cuz like I I do not want to
1:22:16release that until I 100% confident
1:22:19there's no bugs but
1:22:21we were able to take essentially a
1:22:23skill, an idea and now it's like a fully
1:22:26fleshed built out tool that you can go
1:22:29in and and use. And now that's turning
1:22:32into hold on a second.
1:22:34>> [laughter]
1:22:34>> There's too many ideas. We got to go
1:22:36priority here in regards to like what's
1:22:38going to move the needle for the
1:22:39organization. But I do see the future
1:22:42where
1:22:43my team for the most part will probably
1:22:45be inside what we call metrics. And a
1:22:48lot of the skills that are like working
1:22:51really well for us, we will just have
1:22:53the developers sort of build that out
1:22:55within side of our operating system cuz
1:22:58I could see I would rather have my team
1:23:00live inside metrics versus Claude and
1:23:04this and that.
1:23:05>> Yeah. And and we'll we'll we'll just
1:23:07abstract that layer away, move up the
1:23:09stack, and we'll build interfaces that
1:23:12sit over the top of all this stuff that
1:23:14are exactly the way we want them, that
1:23:16work differently. Now, I don't mean to
1:23:18minimize that at all at all cuz you're
1:23:20ahead of 99% of agencies.
1:23:22>> Please do that. I mean right yeah, we're
1:23:24>> But but but let me just take you three
1:23:25hills into the future and tease
1:23:27something that you will get next week in
1:23:30the May update of Ads to AI with the May
1:23:32update of brain.
1:23:34What I'm playing with just this week and
1:23:36I'm going to release next week
1:23:38is imagine now that software
1:23:41that can be changed by them.
1:23:43So they can get metrics and that's
1:23:46delivered as a fully functioning thing,
1:23:48but they can right click anywhere on
1:23:50that screen and say like we should have
1:23:53this thing here or it'd be really useful
1:23:55if we had these buttons here, not just
1:23:58these. Make those suggestions
1:24:01that
1:24:03obviously to begin with we got to build
1:24:04trust. There's going to be a human in
1:24:06the loop there. Maybe that bounces to
1:24:08your developer, but your developer isn't
1:24:10going to be writing that code. Your
1:24:12developer is going to come up a level
1:24:13and be the orchestrator of that and he's
1:24:16going to be basically looking at a whole
1:24:18bunch of tickets and going that's a
1:24:20really good idea. Drag that over across
1:24:22to that Claude get on with this column.
1:24:25Meanwhile, go back to the
1:24:26That's a good ticket. Claude, get on
1:24:28with this. Claude's going to build it.
1:24:30Claude's going to test it all. It's
1:24:32going to spin up the new UI. You or the
1:24:35person that created the ticket or
1:24:37somebody is going to glance at it and
1:24:39go,
1:24:40"That's really cool." Tick. And it's
1:24:42going to go into production. Does it go
1:24:44into production just for them or for
1:24:46everybody? That's up to you.
1:24:48>> Mike, I could talk to you, you know, for
1:24:50many more hours, right? But I know
1:24:51you're busy. So,
1:24:53I will end it with one more question cuz
1:24:55I see a lot of fear posting going on, it
1:24:58feels like, in regards to security right
1:25:01now. In fact, there was just that case
1:25:03It was Was it today or yesterday where a
1:25:05car rental software company it deleted
1:25:08the Claude
1:25:09went in and deleted the entire database
1:25:11like,
1:25:12you know, it sounds like they were able
1:25:14to recover it all,
1:25:16but it you know, became this huge thing.
1:25:18Uh
1:25:19I what like And I know that you're not,
1:25:22you know, you're not a lawyer, this
1:25:23isn't advice in regards to this, but
1:25:24like in regards to the security aspect,
1:25:27like here at Grow My Ads, we're not
1:25:29Nothing's being auto like edited in
1:25:31accounts or anything like that. And the
1:25:34way I have interpreted Claude, it'd be
1:25:36like if I added client, you know, stuff
1:25:39into Google Drive, right? Um now, I'm
1:25:42not, you know, if someone shared their
1:25:44like customer list with us, I'm not
1:25:46going to feed that into Claude, right?
1:25:49Um obviously that's That gets hashed
1:25:50into audience list inside Google or Meta
1:25:53or however you're using it. But um when
1:25:56it comes to the security and some of
1:25:57this fear people are hearing are hearing
1:25:59and and seeing,
1:26:01uh what like what what do you know in
1:26:03regards to like, "Hey, listen, if you're
1:26:04just pulling in CSV data or even you
1:26:07have the API integration and you're
1:26:08working with an account, it's safe
1:26:11fairly safe as long as you have
1:26:12guardrails and you're not like telling
1:26:13it to like go just automatically do as
1:26:16it pleases and make edits inside of an
1:26:18account.
1:26:19>> I think
1:26:21this is also a very very personal thing.
1:26:23Some people
1:26:25refuse to paste campaign names and and
1:26:28like just campaign level data, just
1:26:30campaign names and metrics into an AI
1:26:32window.
1:26:34I
1:26:35don't quite understand that viewpoint,
1:26:37but I understand that people have that
1:26:41belief or that way of thinking. I'd say
1:26:42more so in Europe than America and
1:26:45Australia as a rough generalization.
1:26:48I think you
1:26:50first of all just have to be
1:26:52thoughtful about it. You you have to
1:26:53have some sort of data policy as a
1:26:56business and have a way to make sure the
1:26:58people are doing that. Now, everybody's
1:27:00data policy is going to be different.
1:27:01For me, I'm a bit more fast and loose. I
1:27:04am very much of the beg for forgiveness,
1:27:06don't ask for permission
1:27:08ilk. Um
1:27:10so I'm probably not the best person to
1:27:11talk about data security, but
1:27:14um
1:27:15I think that
1:27:18as yes, guardrails are the far more
1:27:21important thing, but the
1:27:24the huge benefit
1:27:26of not blindly moving
1:27:29forward, but moving fast and using this
1:27:31stuff and learning it and seeing what it
1:27:33can do and what it can't do.
1:27:34And you have to implement it. You have
1:27:36to experiment. You can't know That's why
1:27:39it's called an experiment. You can't
1:27:40know beforehand this is going to work or
1:27:43this won't or this might be a little bit
1:27:44out of the bounds or not.
1:27:47I see far too many people moving
1:27:48incredibly slowly, being super cautious
1:27:51and super safe.
1:27:53>> Yeah.
1:27:53>> just I worry for them, for their
1:27:55careers, for their businesses because
1:27:58if you move that slowly and everybody
1:28:00else Now, I'm not saying we should all
1:28:01race ahead blindly and just you know,
1:28:03YOLO everything. That's the other
1:28:05extreme and extremes are never good.
1:28:08But moving too slowly and too cautiously
1:28:10means you just don't get the benefits.
1:28:12You're not going to get that 30%
1:28:14efficiency a year. You're certainly not
1:28:15going to double your business moving at
1:28:17a sales pace.
1:28:19I think there has to be some sort of
1:28:21middle ground.
1:28:22And a lot of it comes back to trust and
1:28:24guardrails and the appropriate backups.
1:28:27And you can give the tools
1:28:29read-only access and not write access.
1:28:32Right? I
1:28:33My AI can read everything in Gmail. It
1:28:36can download the attachments. It can
1:28:38create drafts in Gmail, but it's not
1:28:40allowed to send any emails yet. David,
1:28:43who I mentioned in my master mind
1:28:44before, his is sending emails all the
1:28:47time. He has a couple of members on his
1:28:49team who think it's weird to get an
1:28:51email from Claude.
1:28:52And they are struggling with dealing
1:28:54with that. Most of his team's totally
1:28:56fine with it because it allows the whole
1:28:59business to move faster and be that much
1:29:02more efficient. But some people are
1:29:04old-school and like "I don't want to be
1:29:06told what to do by bot." I can
1:29:07understand that, but
1:29:10my example for that has always been
1:29:11Uber. You got a quarter of a million
1:29:13people and their boss is a computer in
1:29:15San Francisco. Like, they're all being
1:29:17told what to do and where to go and who
1:29:18to pick up and where to drive and how to
1:29:19get there.
1:29:20>> Yeah.
1:29:21>> You're told by bot. So,
1:29:23you have to have a data policy. It has
1:29:24to be intentional.
1:29:26It has to match the organization's
1:29:28appetite for risk, I suppose.
1:29:30Um you could hash everything. You could
1:29:32redact everything. You could anonymize
1:29:34data. You could take that search term
1:29:36report and map every search term to some
1:29:39other term or some number or hash or
1:29:41something. Do all the analysis, come
1:29:44back through, retranslate it all back,
1:29:46and then present the report. You could
1:29:48do that, but personally, I don't see a
1:29:50problem in putting search term data with
1:29:53some metrics into a large language
1:29:56model. One, Google's already removed
1:29:58everything useful in there for privacy
1:30:00reasons anyway, but also
1:30:03these large language models are not
1:30:05storing all of that stuff in a database.
1:30:07Nobody's trawling through that and
1:30:08going, "Oh, let's find some interesting
1:30:10Google Ads data in one of these accounts
1:30:12in Anthropic."
1:30:15>> That we know of, allegedly, right?
1:30:16>> know of, allegedly, exactly, right?
1:30:18>> as if you stuck that data in Google
1:30:20Drive, right? You know, it's you you
1:30:23think, well, Google's not like
1:30:25>> Google's got pretty good security. Like,
1:30:26a lot of people are trying to hack it
1:30:27every day, and no one's managed to do it
1:30:30yet.
1:30:30>> Yeah.
1:30:31>> You put it all in a Google Sheet anyway,
1:30:33right? It's
1:30:35it's not that different. And yes,
1:30:37Anthropic and OpenAI keep your data for
1:30:3930 days, and but you can opt out of most
1:30:41of that, at least once you start paying
1:30:42them a little bit of money. So, yeah,
1:30:44again, understand the rules. If you
1:30:46really, really want, I wouldn't do this,
1:30:48but if you really want, you go get a
1:30:49lawyer, and they tell you what to write
1:30:51in your policy, and they tell you what
1:30:52you can and can't do. But, again, for
1:30:54me, that is
1:30:56going a bit overboard. But, I think the
1:30:58main thing to answer the question
1:31:00specifically is just you have to think
1:31:01about it. You have to decide
1:31:03what you as a business do,
1:31:06um and what the rules are, and make that
1:31:08really, really clear for everybody and
1:31:10your agents, increasingly, to be able to
1:31:13to follow. Had a really good thing from
1:31:15Aaron Levie at Box the other day.
1:31:17He was saying that, you know,
1:31:18let's assume everybody has an agent at
1:31:21some point in the future, right? Let's
1:31:22assume just one for now, but each person
1:31:23has their own agent. Just like we in the
1:31:25agency used to give everybody a their
1:31:27own VA.
1:31:28It gives them leverage.
1:31:30Your agent will never have more
1:31:32permissions than you do. So, each person
1:31:35at each level of the business, whatever
1:31:37they that person has access to, their
1:31:39agent can't have access to more than
1:31:42that. If you're in a sub MCC, your agent
1:31:44doesn't magically get access to
1:31:46everything. If you're plugging into some
1:31:49HR data, and you see your one-on-one
1:31:51reports and your salary, your agent
1:31:53doesn't magically get to see everybody's
1:31:55salary. And that's a really good, simple
1:31:58framing of let's just start there.
1:32:01Build it so it doesn't get more than you
1:32:03do. And then because we we've solved
1:32:05that problem in organizations for
1:32:08hundreds of years. If we understand that
1:32:10there are different levels and different
1:32:11people need to see different things.
1:32:12Let's just start there and
1:32:15iterate from there.
1:32:16>> Yeah, I've yet I've yet to see anyone
1:32:18who's like just throwing accounts to
1:32:21Claude and saying go make changes and
1:32:24and like it's I I so again, I think it's
1:32:26a lot of just
1:32:27fear posting for clicks and and debates
1:32:30and and it's fine. It works for the
1:32:32conversation,
1:32:33>> But it's it's a t-shirt I want to get
1:32:36built at or made which is common sense
1:32:38isn't.
1:32:39>> Right.
1:32:40>> But it is just common sense. You
1:32:42wouldn't just do that. You wouldn't hire
1:32:43a new person and day three on the job,
1:32:46you'd say just go do all of that stuff,
1:32:47please with no training, with no
1:32:49context,
1:32:51with no systems.
1:32:52It's it's just common sense. Just Yeah,
1:32:55don't don't do that, please.
1:32:58>> Yeah, I agree.
1:32:59Uh Mike, this was uh incredible. It
1:33:03uh you know, I can't I don't remember
1:33:05the last time I actually have talked to
1:33:06you, uh but it's hilarious cuz you know,
1:33:09we just picked up right right on this uh
1:33:11recording and uh
1:33:14I I owe you to a lot of my success. I
1:33:17would not have learned Google Ads at the
1:33:20level I did without your group, without
1:33:22your training in the early days and so
1:33:26um you know, I think very highly of you
1:33:28and uh I'm very grateful that I somehow
1:33:31crossed your path in the weird world of
1:33:34the internet uh
1:33:36all the way you in Australia and me in
1:33:37the you know, the small city of Fort
1:33:39Wayne, Indiana. So uh it's incredible
1:33:42what you've built and what you're
1:33:43building with Ads to AI.
1:33:45Uh I thought you might have been losing
1:33:47your mind many years ago when you were
1:33:49in the Facebook group always talking
1:33:51about, you know, all of the uh this AI
1:33:54stuff uh very early on and uh so I knew
1:33:58when
1:33:59um, you know, when it came to Google Ads
1:34:00and and where AI's at today, I knew that
1:34:02if there's anyone who's got it figured
1:34:04out, it's you. And sure enough, you do.
1:34:06So, anyone who's listening to this,
1:34:08go do 80/20 skill. Uh, again, that the
1:34:12the website to that will be uh, in the
1:34:14description below. And just go buy his
1:34:16course, spend the few hours it takes to
1:34:19go step one, two, three, three, four. He
1:34:21makes it very simple. A a guy like me
1:34:23was able to go through it and you
1:34:25absolutely will get hooked and you'll
1:34:28realize using uh, these tools such as
1:34:31Cloud Code is it's really not that hard
1:34:33once you get into it. So, thank you,
1:34:35Mike.
1:34:36>> I appreciate saying all of that, but I
1:34:38also appreciate you doing the work. You
1:34:40took the training, yeah, but like anyone
1:34:43could do that. Lots of people did. Not
1:34:45everybody went and implemented it and
1:34:47certainly not to the the level that you
1:34:49have implemented. Now, you're doing it
1:34:50again with all of the brain stuff. So,
1:34:52thank you. That's what every teacher
1:34:56always wants, right? Is for someone to
1:34:57actually do something with it. So,
1:35:01I'm just the the first step. You would
1:35:03have absolutely got to where you've got
1:35:05with or without me, but I appreciate
1:35:06that you have um,
1:35:08implemented everything so well and
1:35:10continue to be so curious. I think that
1:35:12is the most important thing right now is
1:35:14just being curious and being willing to
1:35:16play. It's you're doing it. Thank you.
1:35:18>> Yeah, absolutely. All right, well, I'm
1:35:20excited for how the future evolves with
1:35:22this. So, I'll have to have you on
1:35:23again, you know, uh, in the future cuz
1:35:25I'm sure there's going to be a lot of
1:35:27advancement and new things and exciting
1:35:30>> fairly slowly, doesn't it? This AI
1:35:31stuff.
1:35:32>> Yeah. So, cool. All right, thanks, Mike.
1:35:35>> Thanks, Bram. Good to see you.

About this transcript

This page contains the full transcript of Claude Will Change Agencies Forever (with Mike Rhodes) by Grow My Ads, generated from the public captions YouTube serves with the video. The transcript has 18,637 words across 2,806 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.