Latest / State of the AI Union / CT Leong: Mastering AI Native GTM Architecture
Transcript
- 0:05Hello everybody, and welcome to this week's episode of the State
- 0:08of the AI Union. I'm your host, Laura, and join
- 0:11with me today is CT Leong of the Go to Market Architects.
- 0:15CT do you want to introduce? Yourself.
- 0:17Thanks for having me on the show, Laura, and I am.
- 0:21Over the past 10 years I've spent a lot of time building two
- 0:25different SAS companies, one venture funded and one without
- 0:29to date went from zero to over $5,000,000 in closed revenue
- 0:34across companies. Now I'm running an agency
- 0:37focused on applying AI to help founders to build their GTM
- 0:41operating system so that their pipeline can run without them
- 0:45being involved 24/7. So great to be on the show.
- 0:48Well, that's awesome. I'm sure lots of founders are
- 0:51excited to hear what that means, like having the pipeline run
- 0:54without them having to pay attention to it.
- 0:57But first, let's actually define what a go to market architect
- 0:59is. Tell us more about that role.
- 1:02Yeah. So this role is, you know, Clay
- 1:05kind of claims credit for coining this term GTM engineer.
- 1:12And I see like so many people on LinkedIn today talking about GTM
- 1:16engineering and you know, no, not to diss on them or anything,
- 1:22but 80% of it is a lot of just like diving into clay and
- 1:26working clay tables. I think the GTM architect,
- 1:31however, is able to combine like deep knowledge of go to market
- 1:39both in terms of company building, in terms of being able
- 1:42to stitch together different tactics, different workflows
- 1:45over and above your clay table and apply that to generate real
- 1:51pipeline. And it has both a strategic and
- 1:55an operational function. So it comes with overall GTM
- 1:59strategy. So being able to clearly define
- 2:01your ideal customer profile or your Icps, being able to align
- 2:05your positioning, messaging and so on.
- 2:08But then also being able to take that down to the brass text, the
- 2:13exact specific channels that you're going to be using, the
- 2:18workflows that you need to run. What does that mean for Rev OPS?
- 2:21What does that mean for sales enablement?
- 2:24What does that mean for your SDRS and your A ES?
- 2:28And stitch that all together into a functioning system that
- 2:33takes a lot of the manual work out of running GTM teams, so
- 2:40that that's how I see it anyway. Yeah, I, I like to joke that,
- 2:44you know, I was finally an engineer after 20 years.
- 2:48You know, anyway, one of the things that you talked about was
- 2:52how you know you are second time founder and your last company
- 2:56was Engage Rocket where you know, you built the the golden
- 2:59market team. And I'm sure that's where you
- 3:02realize, you know, how important pipeline is or or was it's it's
- 3:07a very near and dear to my heart pipeline.
- 3:09This topic about it. But I wanted to ask you, what do
- 3:13you think you misunderstood about Outbound very early on and
- 3:18what are the things you realize about what it should be?
- 3:22Good question. So I first started doing
- 3:25outbound more than 10 years ago. When it was still fine to kind
- 3:31of buy one domain, one e-mail address and buy a list from
- 3:38somewhere, Chuck it into a tool like at that time I was using
- 3:42mail shake and I got it on a one of the apps Sumo sales like
- 3:48lifetime deals. I think it's still like it's net
- 3:50account is still still alive. And it was really just kind of
- 3:56spray and pray, you know, and you know, I, I tried different
- 4:00tools. I I tried plenty when plenty was
- 4:04new, like they went like when they were in MVP mode.
- 4:08It was really crappy, very duct tape together.
- 4:15And to me that was outbound, right, Just like getting lists
- 4:18and just blasting the list and, you know, working the numbers
- 4:22and everyone reads this. It's it's from predictable
- 4:25revenue. Aaron Ross, Jason Lemkin.
- 4:30And that was the Salesforce playbook for like 1520 years
- 4:34ago. I think that has really changed
- 4:38though. Just even in the last 18 months,
- 4:43it used like and part of it hasn't changed like in the sense
- 4:48that like Outbound cannot exist in a vacuum.
- 4:50Like as an early stage founder at that time, like I thought,
- 4:55let's just hammer outbound as a single channel and you know, we
- 4:59can grow the business from there.
- 5:01And one of the things that could be, well, it took me a long
- 5:03time, but founders as much smarter than me quickly realized
- 5:07that you need to combine that with a slower channel like
- 5:12content marketing, brand marketing.
- 5:14You need to have marketing as air covers so that the sales
- 5:18conversation becomes easier, so that your conversion rates to
- 5:22like a discovery call become easier and and and much better.
- 5:27So I'm seeing outbound now as something that has evolved
- 5:32towards kind of just in time targeting for potential buyers
- 5:39in your market. So you have your entire Tam.
- 5:42I think there still should be some effort to try to contact
- 5:45your entire Tam at some regular cadence, but the real effort
- 5:49should be spent on who are those that are silently raising their
- 5:55hand in your time to say, I'm ready to buy or I'm looking for
- 5:59a tool like yours. And that's where the, the
- 6:03engineering, the architecture, whatever you call it like that,
- 6:06that's where like outbound is becoming more and more focused.
- 6:09And it, it, it drives value for both ends, right?
- 6:12Like you create value for the person who needs your solution,
- 6:15like right now, today. And of course you you derive
- 6:19value for your own company as well.
- 6:21Yeah. So if I'm hearing you correctly,
- 6:22it's like, you know, in the beginning what what we thought
- 6:26about outbound was more spree and prey, right.
- 6:29Let's just like sort of you know, send blast everybody an
- 6:33e-mail or you know send them some digital campaign.
- 6:35And now we are moving really more towards an ABM or ABT type
- 6:42of marketing and and outbound strategy where you know we're
- 6:46really more focused on the signals that people are giving
- 6:48out versus just giving out signal to everybody that might
- 6:52be willing to listen. Yeah.
- 6:55And it's, it's, it's really just about being able to connect the
- 6:59outreach a lot better to the rest of the machine.
- 7:02Like it's not that outbound is ever broken and it's it's just
- 7:07not connected. Right, right, right, right.
- 7:09Awesome. OK, so when we so you started
- 7:13this company to help companies with their go to market motion
- 7:16and you know, the the, the setup, the setup of the systems
- 7:20and, and all the architecture around it.
- 7:23What are some things that you might look at first when a
- 7:27company comes to you and says, hey, we need help with go to
- 7:29market? It's a great question.
- 7:31I just had one this morning that I was talking to and I use this
- 7:37framework. It's it's a little bit hokey.
- 7:38It's a little bit not very software Y because it comes from
- 7:44Alex Hormuzi, who's not typically seen as a guru in
- 7:48software space, but he has this mantra, which I really, I, I
- 7:53really believe makes a lot of difference.
- 7:55Like it's just more better than you when the first thing you do
- 7:59when you look at your go to market is what's working today
- 8:04and what's stopping us from 10 acing that?
- 8:07So if it's content, if it's SEO, like you know what I publish on
- 8:13LinkedIn, say 2 * a week, right? And it generates 1-2 meetings a
- 8:19month. Like what's stopping you from
- 8:21publishing 10 times a week? And you know, I mean, like
- 8:25hypothetically speaking, I know there are a lot of different
- 8:27limits and all that kind of stuff, but just really asking
- 8:30that question, what's working today and why can we not do more
- 8:36of it? Because as most founders and,
- 8:39and go to market leaders and revenue leaders, like we, we are
- 8:42very attracted to new. Like we go to new straight away.
- 8:44Like what's the latest tool out there?
- 8:46What's the, what's the best like NAN workflow I can install to to
- 8:5110X micro, whatever it is, right?
- 8:52But really we should be looking at like the more because that's
- 8:56working already. Like that's not that like not
- 8:59not break what's working. Let's, let's, let's juice it
- 9:02like 10 times, right? And then you start realizing,
- 9:05OK, there are limits, right? Oh, I, you know, posting 10
- 9:08times a day, like LinkedIn is going to, they're going to,
- 9:11they're, they're going to Nix my account for whatever, right?
- 9:14So then what's the limit? Like let's, let's get to the
- 9:19limit in a way that doesn't kill us or doesn't like cause long
- 9:24term damage to our architecture. And then we figure out like,
- 9:28what's better? How do we do this better, right?
- 9:32And better can mean a lot of things like I'm working on a a
- 9:36landing page right now that's converting at .8% right to to a
- 9:40demo. That's horrible.
- 9:42But it's also great news because I don't need to do anything
- 9:45else. I just need to figure out this
- 9:46landing page is getting 1000 like thousand hits a month of
- 9:52traffic. If I double this conversion rate
- 9:58like instantly the business double s like nothing else.
- 10:00Like I just fixed this landing page which has like with some
- 10:03hokey headers and, you know, a broken form on it.
- 10:08Just fix that and you double your business.
- 10:11Like, why bother doing anything new, right?
- 10:13And only after you've maxed those out, then you figure out
- 10:16new like, oh, we've had APLG that's been working.
- 10:19It's gotten us to like 3,000,000, err, 5 million,
- 10:22whatever it is. We're, we're kind of stuck.
- 10:25We, we want to look at outbound as a, as a channel.
- 10:30Now we talk about new, right? Like, so let's look at all, all
- 10:33your, all your different channels.
- 10:35Like, OK, you've maxed them out. Let's talk about new.
- 10:37Like how do we build new and new doesn't come immediately.
- 10:40That's the, that's the hardest part, right?
- 10:41Like new, you know, you get people.
- 10:44I, I get found to send me this all day, like like I'm giving
- 10:47outbounds 3 months. And like, how have you done?
- 10:49Like have you done any outbound before?
- 10:51Like have you converted any cold traffic before?
- 10:54No, zero. I was like, OK, three months is
- 10:58not going to cut it. You're looking more at like a
- 11:01nine to 12 month journey at best, you know, and if you hire
- 11:05AVP of sales, who knows what they're doing maybe 9 to 12
- 11:09months, right? If you're lucky.
- 11:12But you know, they, they always joke, right?
- 11:13That, that your, your best VP of sales comes after the carcass of
- 11:17your first one. So you every, the first VP of
- 11:20sales always, almost always failed.
- 11:22And it's almost always the fondest fault because they don't
- 11:26know what they don't know and they're just kind of hiring this
- 11:29guy hoping that he'll, he or she would figure everything out for
- 11:33them. And it just doesn't happen.
- 11:34Like you, you, you got to, you got to have all the systems in
- 11:38place and you hire AVP of sales to maximize that rather than you
- 11:43know what I see a lot of founders do.
- 11:45Yeah. So if I'm hearing you correctly,
- 11:48what you're saying is let's do more of what works and, and then
- 11:55figure out the new, because new takes time.
- 11:57And first of all, new is going to tell us, you know, what is
- 12:00working. And then we have to do more of
- 12:01that anyway. So why don't we just concentrate
- 12:03on the things that are currently working right now And just and,
- 12:08and it's OK if, if I'm hearing you correctly, it's OK if it's
- 12:12janky, as long as it, it improves, you know, the results
- 12:15of what it is today. Yeah, more than you fix better,
- 12:19right. So it's janky, sure, we make it
- 12:21less janky, then we do better, right?
- 12:23And then only after it. We're really doing great then.
- 12:27OK, let's figure out new. Yeah.
- 12:30So, you know, Rev OPS agencies have been around for quite a
- 12:34while. There have been various flavors
- 12:36of it. And you, I know, are doing a lot
- 12:39of work with AI and you're you're an AI native go to market
- 12:44Rev Rev OPS agency. So what do you think are some
- 12:47key differences between the traditional Rev OPS and this new
- 12:51version of an AI native go to market engineering agency?
- 12:56That's a really good question because the, the, the, the
- 12:58umbrella of Rev OPS and GTM engineering is just so broad
- 13:02today. Like everyone claims part of it.
- 13:05And the reality is that they're all important, right?
- 13:09Like they're all important. The, the maybe the biggest
- 13:12difference that stuck out for me was that typically what Rev OPS
- 13:20tries to do is to make your revenue data more observable,
- 13:23right, more visible. And like, you know, how your,
- 13:26how a company makes money is a little bit more clear to all the
- 13:33actors involved. Yes, I think, I think there's a
- 13:38lot of overlap with GTM engineering or GTM architecture.
- 13:45The main difference is how AI is used in this process because if
- 13:51you know, there are many Rev OPS agencies out there who really
- 13:54are like Salesforce or HubSpot specialists right there.
- 13:57Their job is to connect all the dots within Salesforce and
- 14:00HubSpot and you know, make the plumbing work well.
- 14:03And, and that's really super important.
- 14:04You got to have that thing in place.
- 14:09I think what the aspiration is on the GTME side of things is
- 14:15how do you, how do you get that 10X improvement over just those
- 14:21platforms that you have? And that's where I, you know,
- 14:28this is something that I'm going like all in on this year where
- 14:34it's almost like the UI less agency or whatever it is, right?
- 14:41Like everything can be run through the command line.
- 14:44If your tools are MCP or API native and you're in cloud code
- 14:51and you know how the machine should work, then it really
- 14:55becomes a function of what are all the workflows that we're
- 14:59running today, which are the ones in prioritizing the ones
- 15:02that are closest to revenue and just auditing them, you know,
- 15:06like which of these are efficient?
- 15:08And, and to the earlier framework like, you know, we're,
- 15:12we're functioning like what, what can we do better?
- 15:14Like we already are doing, like everyone's already maxed out,
- 15:17like they're working 60 hour weeks.
- 15:18How do we, how do we do better on any of these?
- 15:21So like looking at like all the different workflows and just
- 15:23prioritizing them and, and then fixing that.
- 15:26And it's not hard. Like once you get to that point,
- 15:29just tackled it 11 workflow at a time.
- 15:32And, and that's what GTM in, in my ideal case, it's like Rev OPS
- 15:37gives you that dashboard and GTM engineering hopefully gives you
- 15:40the self driving car, right? You should be able to move.
- 15:44Right, right, right, right. Awesome, awesome UI.
- 15:46Less UI, less Rev OPS, everything going through the
- 15:51command line. That's really interesting.
- 15:53I spent 80% of my life in in in cursor.
- 15:56Like I don't even, I don't even look at GA 4 or any of the
- 15:59dashboards because it's all there.
- 16:01Yeah. So actually I have two questions
- 16:03about this. Well, let's start with actually
- 16:05what what these AI agents do for you.
- 16:07What are some of the the capabilities that that you've
- 16:10you've gotten your UI list minions to to run?
- 16:16OK, so there's a lot of hype about this on LinkedIn and
- 16:22everywhere else. I I'll, I'll be, yeah, I'm going
- 16:26to be as honest as possible about the capabilities, right?
- 16:28So one of the agents that I, so I like, I think of them as like
- 16:33agent teams, right? So it's almost like having staff
- 16:38on on my team. So one of them is they do, they
- 16:43do like an audit, so like just a GTM audit of a company, right?
- 16:48So there's a kind of a supervisor agent that runs the
- 16:52show. There's another agent that is a
- 16:55specialist in just Google Analytics.
- 16:59So I plug that agent into GA four with specific skills that I
- 17:03give it to be able to analyse the data that comes back from
- 17:08GA. Then there's another that's a
- 17:11paid specialist. So that paid specialist then
- 17:15plugs into the meta MCP. It pulls data from LinkedIn ads
- 17:19or Reddit ads, whatever it is and then analyze that and talks
- 17:24purely about ad spend. There's another one that does
- 17:27like outbound. There's another one that like
- 17:29for each channel, there's, there's like one agent, it's a
- 17:31whole agent teams like 7 agents and the supervisor just looks at
- 17:35this and coordinates the work right and tries to make sense.
- 17:40And if there's a conflict, try to deconflict the, the data and
- 17:44puts together this report that says, all right, here's like the
- 17:47top three things that are blocking you today in your DTM.
- 17:51And then you have like 6 different reports that old, like
- 17:55detailed reports that come with it also.
- 17:57And this applies for like pretty much anything because you, you
- 18:01like, I was doing this recently and there, there were like 4
- 18:04different sources of truth. Like there was a Google
- 18:06spreadsheet somewhere. There's like hotspot, there's
- 18:09like whatever. So like then in that particular
- 18:12case, you maybe need to build like a reconciliation agent that
- 18:15pulls information and tries to make sense of it.
- 18:18So that's kind of one example. Another example would be like
- 18:24just for outbound, right? Like they're like this is a 5
- 18:27agent stack it it's pretty linear in that sense.
- 18:31Like the first one just monitors buying signals.
- 18:34So it's a signal capture agent you look at.
- 18:37So there are a few different tools they can use like
- 18:39Triggify, Aquifer clay and all that like.
- 18:41And it just monitors things like new jobs, new company news,
- 18:46whatever. And you can design it, whether
- 18:47you want to have it as a fixed number of companies with
- 18:50specific trigger signals or you want to just monitor trigger
- 18:54signals and you pull the companies from there, right?
- 18:56So that kind of 1. Then the second agent deals with
- 18:59research and qualification because not all of the, the
- 19:04signal that comes through is going to be signal there's going
- 19:06to be a lot of noise. So you need an agent to be able
- 19:10to sift through this noise for you and qualify them and enrich
- 19:15that. And that all, all that comes in
- 19:18the third step where there's a personalization of the outreach.
- 19:21So your research agent has just qualified and done some research
- 19:25on each of the, the valid Icps that come through.
- 19:29Now you have an agent that personalizes the outreach
- 19:32message given your product marketing context, given, you
- 19:37know, you load it with certain skills around like how do you
- 19:40reach out, you know, wait, it doesn't sound AI, if that makes
- 19:43sense. So you have, you have to, yeah.
- 19:47So you build all these skills into each of these agents.
- 19:50So this agent's job is purely personalization.
- 19:53And then there's another agent that does multi channel routing.
- 19:56So there's like if you have a high score, for example, then
- 19:58you go LinkedIn, e-mail. And if you're like high urgency,
- 20:01maybe go to phone, right. And you hit the SDRS.
- 20:04If it's kind of medium agency, maybe it's just e-mail,
- 20:06LinkedIn. And so on.
- 20:08So that that's the routing agent.
- 20:09It's you can use like cheaper LMS for that.
- 20:13And then kind of maybe mid steps through, you want to insert like
- 20:16a QA agent as well, like somebody like a different model.
- 20:20So, for example, if you use like plots on it, whatever to
- 20:23generate the personalization, you maybe want to have Gemini
- 20:29like Gemini pro or flash, whatever.
- 20:31Looking at that to say, like with certain rules, like QA
- 20:34rules to say, OK, this sounds very AI.
- 20:37This is not these all pass this particular framework.
- 20:41We'll let it through. If not, like, OK, let's flag
- 20:43that and and go back to the drawing board, right?
- 20:46And then like an optional one that you can put in is kind of
- 20:48like a like a like a base layer agent which does learning.
- 20:53So this this agent's job is to log the the whole process and
- 20:58understand like what's going wrong, what's what's going well?
- 21:01And then the next time you run this, apply certain Cheat Sheets
- 21:06and shortcuts that all the different agents can learn from.
- 21:08So like those are just two different examples of what you.
- 21:11Can do Wow, I love it. Your your army of agents.
- 21:14And you know, I think what I'm hearing you say also in terms of
- 21:17best practices is, is to kind of stack micro agents on top of
- 21:21each other versus trying to build 1 giant one with
- 21:24everything because that way you can also maximize for both
- 21:26efficiency and for cost. Yeah, I, I think it's maybe less
- 21:31about cost and more about, more about how effective any single
- 21:36agent can be because you're the, the nature of it's, it's almost
- 21:41like thinking about organizational design, right?
- 21:43If you're thinking about building a Rev OPS team, you
- 21:47can't say hire like 10 people and say, all right, you guys all
- 21:49do everything. You have to split the team up.
- 21:52And it's very much the same with AI, except AI doesn't sleep and
- 21:58doesn't need health benefits. Yes, and they can do it well
- 22:02while you're while you're not paying attention to.
- 22:05Yeah. And it does a honestly, it does
- 22:07a better job than me. Like honestly.
- 22:10OK, cool. Last question then let's pivot
- 22:13here. So what advice would you give to
- 22:17a sales team that's trying to move from sort of more manual
- 22:21and traditional outbound method to a orchestrated outbound?
- 22:24What should they do first? I mean, the first thing would be
- 22:28back to that point, right? More, better, new, right.
- 22:31So what is it that you're doing today that's working?
- 22:33What's stopping you from doing more?
- 22:35And if there's what's stopping you is I need to hire more
- 22:38people. That's where you pause and say,
- 22:40OK, then we need to think about better.
- 22:42Like, what are we doing badly today that we could do better?
- 22:46And I don't know, 60 percent, 70% of the time I, I see this, I
- 22:51don't get a good answer from GTM teams because their data is
- 22:55broken. So more often than not, one of
- 22:59the first things you need to do is just go fix your data, right?
- 23:01Like, and this is a lot less hard than it used to be in the
- 23:06past because now like literally you could just, you could just
- 23:09download a bunch of CSV files from every like all the
- 23:12different systems you're using, Chuck it into plot code and say,
- 23:15hey, go make sense of this, right?
- 23:16And you'll get something back. And even even with that very
- 23:21crappy process, if you are a bit smarter about it, you'll get
- 23:25much better output, of course. And so fix your data is a lot
- 23:29easier than it used to be. The second thing is start
- 23:32documenting your processes. Like a lot of sales teams are,
- 23:35well, a lot of teams in general and product teams, engineering
- 23:38all equally bad at documenting their processes.
- 23:41Again, today, there's no excuse. Like the moment I ship anything
- 23:46in cloud code, like I just have a file inside that says #1 log
- 23:50it to log it to memory, right? And so just keep that as a
- 23:53markdown file like what happened, right?
- 23:55And what did, what did we learn? Like I have like literally, I
- 23:58have a folder called Cheat Sheets for all the different
- 24:01APIs that I'm running and all the different quirks that they
- 24:04have. Like, oh, this particular Jason
- 24:06needs to be nested within that. And like, Oh my gosh, like
- 24:09there's all these tiny little things that you never know until
- 24:11you get into it. But human beings cannot remember
- 24:15all of that. But chucking that into AI, Oh my
- 24:17gosh, like just just do that. Only then do you then start
- 24:21saying, OK, which are the biggest bottlenecks that we have
- 24:24and how? I don't know, like I'm 80 to 90%
- 24:29sure and maybe I'm very biased about this 80 to 90% sure that
- 24:32you can throw AI into any of these bottlenecks and just
- 24:35instantly make it better, like almost instantly, right?
- 24:37That's that's my, that's my belief.
- 24:40Wow. OK, All right.
- 24:42Amazing. So figure out what's working
- 24:45now, do more of it, and then fix what you can Make it better.
- 24:49And then and then let's see whether we can throw AI into
- 24:53doing more new things. Cool.
- 24:56All right, So city, you moved from Singapore about two years
- 25:00ago to the Bay Area. And tell us a little bit about
- 25:05what's been different about starting a company in Singapore
- 25:09versus the Bay Area? Well, you, you know this pretty
- 25:14really deeply as well. I think there's just so many
- 25:20different, so many different areas that are different.
- 25:23Like I in terms of just market size, if you're selling anything
- 25:28to do with software, anything to do with SAS knowledge type
- 25:33products, even consulting, really, you cannot have a, you
- 25:39cannot have a better market than the US.
- 25:41And it's, it's a little bit like saying I want to star in a
- 25:46Hollywood movie, but I want to stay in Singapore.
- 25:50You know, it's, it's almost impossible to like and if you
- 25:54want to be part of that change, you have to be living and
- 25:57breathing that on a day-to-day basis in the Bay Area.
- 26:01So I, I'm very long on AI. Like I used to kind of go back
- 26:08and forth about like whether it's a bubble or not, but I'm
- 26:12very long. I'm very bullish on it right now
- 26:14still. And you just have to be here.
- 26:19And then when it comes down to the nuts and bolts of like
- 26:22company building, this was one of the struggles that I had
- 26:25building company in Singapore where you have essentially a
- 26:30really small market and everyone that is super capable is pushing
- 26:37to win that market. So in a like Singapore as a
- 26:40country is really rich, as you know.
- 26:42And but the real bias of software these are you have
- 26:48government and enterprise, like large enterprise, that's about
- 26:50it. And if you list them out and
- 26:53it's like 50, like 50 possible buyers that are in Singapore of
- 26:59any kind of scale, right? I mean, in, if you're selling
- 27:03into the US market, like I go on clay or go or go on whatever.
- 27:08And you know, I routinely come out with like Tams that are like
- 27:135000 thousand, no founders running like tattoo parlors,
- 27:18like software for tattoo partners.
- 27:19And they're like 20,000 of them were out, right?
- 27:21What I mean there's just, yeah, the scale is just so much
- 27:26different that it doesn't like there's just no comparison.
- 27:29And from a, from a work and business perspective like that,
- 27:34that's there's just no question. Yeah.
- 27:36What do you think maybe. Well, I guess one of the one of
- 27:39the counters to that is, is I think many people say, well,
- 27:41Singapore is, Singapore itself might be small, but Southeast
- 27:44Asia is very large. Asia itself is very large.
- 27:46China itself as a market is is, you know, huge.
- 27:49And while Singapore itself may be small, you know, there's
- 27:53access to a lot of these other economies.
- 27:57What do you think? Maybe the community, the
- 28:00Singaporean community or you know, society can do to maybe
- 28:04help make it healthier for young entrepreneurs that are trying to
- 28:08access these other regions, maybe access the scale.
- 28:14Yeah, I love that you asked that because you know, if you go back
- 28:19say 10, maybe a little bit more like 15 years ago now, there was
- 28:24a prevailing thesis that Southeast Asia was going to be
- 28:27the next like bricks or next like big region because there
- 28:33was like 600 million people, kind of rising middle class,
- 28:39very low Internet penetration, all that kind of stuff.
- 28:42And so like every, like there was a lot of venture money
- 28:44flowing into the Southeast Asian ecosystem.
- 28:49And I mean, just two years ago, I think it was Lightspeed that
- 28:53published. It's almost like an apology for
- 28:56like why that thesis was wrong. And a lot of it came out because
- 29:02the markets are just so culturally different.
- 29:06And the, the labour substitution is so much less acute because I
- 29:17was talking to another founder you like I was, he was telling
- 29:21me about the CRM that was like 10 bucks and did all this stuff,
- 29:2510 bucks per person per month. And it could do all these
- 29:29things. And I thought that's amazing.
- 29:31Like, why aren't you using it? I was like, dude, like I, I, I
- 29:34pay my guy $200 a month. I'm not going to spend $10 to
- 29:39give him a CRM to, to not use, you know, so that, that was kind
- 29:45of one light bulb moment for me. And the other light bulb moment
- 29:47was just like spending 24 months trying to break into Indonesia,
- 29:51into the Philippines and so on. And just realizing how different
- 29:55the, the buying behaviours were in all of these different
- 29:58markets. And just because you've made it
- 29:59in a market like Singapore has, I, I don't think that policy
- 30:03makers understand enough about how much each of these different
- 30:08economies function differently. And that that doesn't even count
- 30:12by going into N Asia, you know, into like Korea, Japan, China,
- 30:18China as a whole. Is it, it's spectacularly low
- 30:22spend on SAS for it's size. And you start wondering why.
- 30:28And you know, India for another one, right?
- 30:30Like you, there's a lot of SAS companies that come out of
- 30:33India, But if you ask any of them like where they make their
- 30:37money, those that are making money, most of them are out here
- 30:40in the United States. And that's just a function of
- 30:47just yeah, I guess it's parties per per capita GDP.
- 30:51Maybe party is is just a buying behaviors are different against.
- 30:55You know, someone told me this once, like all SAS is workflow
- 30:58SAS because essentially you only build a software to make a
- 31:02workflow more efficient if that workflow is consistent.
- 31:05And I think because there's this heterogeneity of workflows in in
- 31:11Asia, like with all the different cultures, like it's
- 31:14very difficult, very, very difficult to have that
- 31:19consistent workflow that works well for every single person.
- 31:22And so all kinds of customization need to come in
- 31:24play. And then that just drags out
- 31:27your product hill until the point that it's unfeasible to
- 31:31scale. So it becomes expensive
- 31:34basically. Yeah, it's super like capital
- 31:37intensive and like now it's even worse, right?
- 31:40Like you have a whole bunch of vibe coders out there claiming
- 31:42they can do whatever. And like those of us in the know
- 31:45know, right. They're they're not going to be
- 31:47able to recreate a sales force. But to someone in Asia who's
- 31:51maybe pissed off about sales force and this guy comes along
- 31:55and says, hey, I can build this for you at a tenth of the price.
- 31:58They're like, you know, why not? I'm giving a shot, OK?
- 32:02So, so maybe one of your advice too for young entrepreneurs is
- 32:05spend some time in the Bay Area because that's going to open up
- 32:08their their opportunities. Yeah or yeah, or even even like,
- 32:13I mean, you're talking about international entrepreneurs,
- 32:15like outside. Yeah, yeah, yeah.
- 32:17Yeah, I would say like, location may be met, like matters a lot,
- 32:22but even before, like, yeah, I don't know, like there's no easy
- 32:27way to skin this cat. Like I wish there was.
- 32:28I wish there was. Yeah.
- 32:30There's AI would say like if you want to be where the action is,
- 32:33that just like you have to be there.
- 32:36You have to be here. OK, OK, good.
- 32:39Awesome. OK.
- 32:40So last question is what is 1 food you really miss?
- 32:44One Singaporean food you really miss that you haven't really
- 32:47been able to get out here? I would say is your talk way
- 32:57it's it's really hard to get out here.
- 32:58So these like for listeners who don't know, like these little
- 33:01fried dough sticks. Like a Chinese fried doughnut
- 33:05basically. But yeah, like a Chinese fried
- 33:07doughnut. Yeah, I I ate that a lot as a
- 33:11kid. I got a bit fat from it but.
- 33:16With soybean, with soybean or with porridge, rice, pork.
- 33:21Plain man, like I just, I just take it and just I had learned
- 33:26to take it with soybean later on, like the Hong Kong style,
- 33:28OK. But like, yeah, that's something
- 33:31that just we can't get easily here.
- 33:34Yeah, awesome. All right, CT, thanks so much
- 33:38for being on the show. Any last words before we end
- 33:40here? No, I enjoyed it.
- 33:41Thanks for having me on the show and I hope this is useful for
- 33:45people listening in. You can hit me up, find me on
- 33:48LinkedIn or whatever it is. I'm happy to share more about
- 33:51all this crazy stuff that's going on in the GTM space today.
- 33:54Awesome. Thank you.
- 33:55And to all our listeners, thanks for being on.
- 33:58And if you like this episode, drop us a comment, hit
- 34:00subscribe. Otherwise we will see you next
- 34:03time. All right, bye, bye.