Bernard Leong: Astrophysics PHD to Serial Founder, Disrupting Extractive Enterprise Resource Planning (ERP) Systems & Startup vs. Family Career Decisions - E393

Bernard Leong, Founder of Analyse Asia, and Jeremy Au talked about three main themes:

1. PhD to Serial Founder: Bernard talked about his initial career in academia, specializing in theoretical physics and obtaining his PhD in Astrophysics and Cosmology, before moving into machine learning at the Wellcome Trust Sanger Institute, home of the Human Genome Project. His entrepreneurial journey started with the co-founding of Chalkboard, a location-based advertising startup that faced strategic challenges and ended in a shutdown. He also co-founded SG Entrepreneurs, which successfully exited to Tech in Asia. He also shared his experiences heading digital transformation initiatives as a tech executive, such as his digital transformation leader at SingPost and pioneering Singapore's first-ever drone delivery.

2. Disrupting Extractive Enterprise Resource Planning (ERP) Systems: Bernard shared his vision for revolutionizing ERP systems to address the inefficiencies and high costs associated with traditional ERP software. He discussed the history of ERP systems and the challenges companies face, including high customization costs and the extractive nature of the consulting model associated with ERP implementation. Bernard aims to leverage generative AI technologies like ChatGPT and Microsoft CodePilot to create a next-generation open business operating system that is 10X more efficient, productive, and adaptable than current offerings. He discussed the reasons behind his focus on the medium segment of the small and medium enterprise market and counter-positioning strategy against existing ERP giants.

3. Startup vs. Family Career Decisions: Bernard compared the life of a serial entrepreneur to that of a football team manager, emphasizing the importance of taking each challenge as it comes. He discussed the practicalities of negotiating entrepreneurship with family responsibilities, highlighting the importance of financial planning, shared goals, and open communication with his wife who’s also an entrepreneur. He also discussed what repeat founders do differently than first-time founders, ranging across distribution, mindset and adaptability.

Jeremy and Bernard also touched on the "first love startup" problem, first-principles thinking, the people who bet against his success, and the future implications of AI in business.

Join the Singapore Growth & AI Summit with Sean Ellis!

BRAVE has partnered with Causality and Sean Ellis to bring you discounted tickets for this Singapore stop on Sean Ellis’ World Tour. Sean’s book, Hacking Growth, has sold over 750,000 copies and he’s sharing the latest insights on enabling breakout growth. Use code seanbrave7 today.

(01:25) Jeremy Au:

Hey Bernard, good to have you on the show.

(01:27) Bernard Leong:

Yes, now I'm finally sitting on the other side of the table.

(01:30) Jeremy Au:

The podcaster's nightmare to be the podcast guest.

(01:33) Bernard Leong:

No, I think you have a pretty interesting setup. I actually could learn something or two from you.

(01:38) Jeremy Au:

Yeah. And I learned quite a bit from you over the years as well. So thank you for that. Well, I'm really excited to have you on the show because, you know, I've been listening to your podcast on and off and big inspiration as well for the Asia side. And obviously hung up quite a few times over the past two years. So, you know, I just wanted to kind of, hear your story. So could you share a little bit about yourself?

(01:54) Bernard Leong:

So I think the easiest way to tell my story is my name is Bernard. I, Bernard Leong, I have actually started originally as an academic. I did my PhD in Theoretical Physics from the Cavendish Laboratory specializing in Astrophysics and Cosmology. Then that was probably the one in a million for me.

And then after that, I went to work in machine learning in the Wellcome Trust Sanger Institute, which is actually the home of the Human Genome Project. And actually recently I caught up with a lot of my ex colleagues there. Most of them are now in DeepMind. Google, Microsoft, OpenAI, and it's really great to be part of that team. A lot of the open source stuff has ended up in the mRNA world. So after I came back, worked as a scientist for a while, decided that's not the path for me. And that's when I started to b Be brave. That's why I've seen a lot of your podcasts and decided to go into the startup world. I did two startups. I always have to call it one and a half.

Let me talk about the first one, which is the one that I crash and burn. And I think I'll talk about later why is one of the bravest thing I did, but I'll explain it later. So, so the particular startup was called Chalkboard. We started off pretty well. We raised money from Joey Ito, who's the founder head of mIT Media Lab, we grew it almost to about a hundred million transactions, location based advertising, except that we made some very big strategic mistakes. Some mistakes that I'm personally quite sorry for, and I did apologize to all my investors then, was we were trying to fight two wars and we were not able to do that ready to go into the U S market that early. So, it had actually a good revenue run rate, but I think we tried to overexpand and we got killed because of short cash. That's actually the pre-Grab pre nowadays people call deserve error. I think if we hang on longer, maybe we will got lucky but I think that crash and burn experience is part of the learning journey for me as well, which I will come back to later.

So the other startup, which I actually started when I came back was SG Entrepreneurs. That was with Gwen. Gwen is probably the public face to it. I'm the other owner together with Wei Chang as well. And I did a lot of the initial work, eventually I handed over to Gwen and sometime around 2014, I also assisted her in getting the sale done to Tech in Asia, which eventually ended Tech in Asia getting two very important things. They became number one in the whole region because we basically tripled their traffic. And second, they got Terence, who's one of my prodigies today in Tech in Asia.

So that is a good exit because recently we have all got our money with Tech in Asia's exit to SPH. So when I crashed and burned, I just started a family and I decided that for the next startup that I'm going to build, I'm going to be the founder and CEO. So, but then I think that at that point in time, I was a co-founder, I was a CTO, and I think one of the bad things in Asia is that people typecast you.

If you're a tech guy, you should be being a solution architect. You should be a tech programmer. So I have to basically start my route back and try to learn how to run a proper business or be the head of a business unit as such.

So I started with Vistaprint first as a product manager. Then I was about to leave Singapore and go to us to be a product manager in amazon.com but Wolfgang Bayer, who was the then group CEO of SingPost came to me and said, well, you know, I can give you a better job, but maybe less than half the pay. Would you come and join me in St. Paul's and lead a digital transformation, which I did. It was probably one of the best highlights of my career.

So I did a couple of things. One was redesigning the sand machine, which is a national icon in Singapore. I turned it into omnichannel. That means it has this mobile app. It was webbed. And then also the kiosk itself. And the kiosk is actually run on iPad. And I was also one of the few companies that Apple would come to visit me. And actually I even got a chance to do the watch earlier. I watch app earlier for SingPost for Apple. We actually gone through the design phase with them. We didn't get in, but it was a very good experience working with Apple guys. So in SingPost, I did a pop station, which I think a lot of people are using.

So one of the things that I really enjoyed that antenna is that whenever I walk around today, I see people using the products. I'm actually the one developed the products with my team, the designers, the team that built it. And of course the most legendary drone flight, which I did that gives the headline on Bloomberg, SingPost, like Amazon, that's drone delivery. So with that, that was about four years. I also ran the post office business at one point. So that was when I actually have a proper P&L and ran both a digital team, which is actually more product management side, and then running the whole entire retail business for SingPost and also with the final project was the general post office, which you see now that's designed based on the old Fullerton building the setup as such.

Then after that I decided that I needed a more regional experience I was actually because of the drone flight, I advised a lot of the post offices Out there. And also including the US postal service, who actually surprisingly came to me on advice on drone delivery. Airbus was also one of the people who invited me to advise to their C suite.

(06:25) Bernard Leong:

And then they came to me and say, Hey, we're thinking of setting up a drone services set business in the Asia Pacific, which would be interesting. So I said, well, you know, I have not done an Asia pet role, so let's do it. So it was a one and a half years constantly going to China, Australia, possibly, and also to US and Europe, because it's quite a global role. And I only spent one and a half years there. I think the only things I've done is setting up some very key strategic partnerships and also figuring out what's the strategy for China, specifically in the drone market there.

Now then, that comes to the most interesting role of my life. And I think that's possibly the missing piece that I was looking for. So when, so what people do not know, if they, if you reject Amazon once, they will keep clawing back at you relentlessly. And there was always this question in my mind that if I have joined Amazon and not SingPost, what, would be my life be maybe better? Maybe I have four, 10 times the stock options, you know, maybe I'm already a rich guy, but okay that aside, but I think the more important question was what would I have learned in an actual tech company, because in the SingPost role, one of the things was I was part of the C suite team that brought the Alibaba deal to Singapore Post. There was a 300 million dollar deal with double the market cap. You need to think of the team in SingPost that I worked for was like a startup team. We behave like a startup team. And I was the only startup guy within that team. And that was the fun part. And after that, I really have really great relationships with everyone.

Wolfgang was very helpful in my career, later on that. So Amazon came back and I decided that, well, this time round, in fact, the joke was I went to my wife and said, I signed it. You don't need to look at the contract this time. I want to do it. So I became the head of artificial intelligence and machine learning business for ASEAN, which is Southeast Asia business. I did about two years and three months. I could have gone for a little bit more, except my dad was in advanced dementia. So I joined as the Chief Digital Officer of Woh Hup both the CIO and the CTO role. That was the large, one of the largest Singapore construction conglomerate. Unfortunately, he passed away three months after that. I think I didn't regret the choice as well. And I think after two years passed, I've decided that this is finally the time. So this is the time to come back and write. I've been in that corporate wilderness part of the story for the last 12 years. So now is the time to write the comeback story. So that's where I am.

And of course, one of the side projects you probably know was Analyse Asia podcasts. I think I always use that podcast to actually train my entrepreneurial muscle, like how to keep growth hacking is a side hustle, but it actually helps me to keep my startup mind. Like I think it was actually difficult if you have gone through a very long and pretty interesting corporate career It might actually make you softer. So I need to keep that muscles going. So Analyse Asia was part and parcel of that. So that's where I am. So I'm now currently working on enterprise AI stuff, which I can talk about a little bit more, but we'll go on with the conversation.

(09:05) Jeremy Au:

There's so much to keep going. I guess we're just going way into the beginning, why did you pick that PhD in that field?

(09:12) Bernard Leong:

Yeah. When I was 14, I read Stephen Hawking's A Brief History of Time. I was really inspired. I want to be a theoretical physicist. I went from a very lousy student to actually manage to do that PhD in a top school and it was quite interesting because that's a little bit like a reminiscent of a startup journey because I wasn't in a top school, that's number one. So when I was at my A levels, I was actually even in Cali Junior College, which was not thought of as a top school in Singapore. And one of my friends who's in a better school came and said to me, my ex former classmate say, Bernard, you're not even in the top schools. You're not in the Science Olympiads. You are neither in science research program. What makes you think you can ever get to Cambridge University and do a PhD? I persisted. I found a very good teacher, a mentor. His name is Dr. Chong Siew Meng. He taught me how to read and how to think and how to really learn the whole, all the difficult math and science by itself.

I could tell you that during my NS time, if I'm not working, I'm doing math equations. I'm trying to solve physics problems for all the first year university thing. I become sharper and sharper. It's like a daily compounding in that journey. So eventually I did made it, I made it, I did exactly what it is. I just couldn't get back for the school reunion to tell the guy who told me that this cannot be done. I made it happen. So I got the PhD that I essentially want, which was actually Cosmology and Astrophysics, which is what I do. And I think that for me, that was my first one in a million hit. I think under any normal circumstances, I don't think anyone would have made it.

I, even I myself didn't think that at that point in time, I think that is probably one thing that I still think about on there. So that was the essential. I really enjoyed the subject. I loved it. I enjoy really working with the smartest people there. When I finished my PhD, I was pretty much thinking about how to actually apply things back more in real life as a friend of mine said, Hey, you know, you there's a human genome project down there. They need a lot of people like you, theoretical physicists, because I did associate somewhere in economics as well. Why don't you just go and apply your craft there?

And so the joke in Cambridge, when you have a theoretical physics PhD, you get a computer science PhD for free. The reason was because a lot of the work that I did in cosmology, we were using a sort of a technique called Bayesian Inference, which is actually doing probability calculations and sampling. And one of the essential things that you end up learning is machine learning, because you need to tease out very difficult data on such. So that was where I actually picked up all my machine learning skills on there. I even did my own programs, coded it. We even tried to make a exponential algorithm to work faster. So we use the Feynman algorithm. I propose the Feynman algorithm actually worked out because we need to negotiate with each other for distributed computing time. That was the pre club days where it's not scalable, right? And there were like seven PhD guys and I'm the post doc and we're all in a bar and everybody's like complaining to each other. They said, Bernard, you're taking on more time because you're machine learning things. So that's it. Okay, why don't we all work together, figure out an algorithm to make the exponential because that's what is causing my thing not running fast enough. So I wrote out the algo, they coded it. Okay. It's quite interesting. They actually coded it. I wrote a very basic one and it's like, nah, this is not good enough for comp science. Why don't we help you scale it? So they scaled it and we got the quote, the times down to 20, 30%.

(12:11) Bernard Leong:

I think this is very important about innovation. It's not so much about like, it's that kind of incremental things that where you're trying to solve the fundamental foundation stuff that actually you made the biggest breakthrough and then essentially we managed to solve all that problems on it. So yes the academic part taught me a lot of in theoretical physics as you will probably know from some of them whether it's Jeff or Elon they all are all physicists by training because of first principles thinking I think that is probably what being a Physics PhD is about.

(12:37) Jeremy Au:

And what's interesting is that after the physics PhD, you decided to go and explore I think two sets, right? I would say more the technology side, but also a little bit more of an entrepreneurial mindset as well. I mean, you could have become a professor, for example, right? You know, there's a common path for PhDs. What were you thinking at that point of time?

(12:52) Bernard Leong:

Yeah at that point in time when I decided to quit, I felt that Academia is becoming in Singapore it was really because of the influence of A-star and all these institutions as it's actually going to be very difficult. And I think they became very prp-US universities at that point in time. So the entrepreneur part was actually during one during when I was doing my postdoc, I was helping the Cambridge University entrepreneurs to do a conference with mIT 100k. And that was where I got to meet a lot of people from Cambridge. So the Cambridge MIT Alliance from the MIT side with MIT Sloan Entrepreneurship Center and also Harvard people as well from the Kennedy School.

So I started getting exposed to the entrepreneur side of it. I should've brought most of those expertise when I came back to Singapore and It was quite natural, I think, from that point. And I also did a small startup there and actually sold everything back to my co-founders because I felt are not fully in the game, you shouldn't be holding any equity, which is what exactly I did. And that experience taught me a lot about, you know, there's this thing called the first love problem. Like you always never forget your first love kind of thing. And because I really knew that a lot of people fail for the first time in entrepreneurship. So I was a little bit detached.

So we did set up the company. It was a biotech company, but the first thing I did was I'm going back to Singapore. It doesn't matter. I'm not here. No skin in the game. I should be fired. And then the company went on to series B, but I think it didn't continue per se. I think it sets me up ready for running ,the entrepreneur journey on that. And the way to think about it is don't think about the first love. Think about what would it be when you're with your soulmate. Yeah, so that means that will give you a lot more opportunity to get to that state that you want. It may not be the first company, even the second company, crash and burn, right? It could be the next company. Of course, there's only limited time. So that was why I thought, this is probably the time to do it. Otherwise, I don't think I have enough time to do it. Now I have three kids. And my wife is also an entrepreneur. So yeah.

(14:46) Jeremy Au:

Yeah. So, you know, that's interesting. You have three kids, you are going to be an entrepreneur again. Your wife is an entrepreneur. How are you feeling about that?

(14:53) Bernard Leong:

It's tough, but you have to first, I think the mental model should be you should never let your kids go hungry. So if I need to do any site consulting projects, I'm literally currently doing consulting and teaching in the university. So no shame. Okay. If you want an example Ryan Peterson, who is the founder of FlexPod, he was doing teaching in the university when he was creating his company. So yeah, you just have to do the side job, make the hustles happen and try to raise money and build the company. So I think all this is still the same. Your family comes first, right?

I think the other thing I started to realize is actually what is the best amount of time to dedicate to what? And I think I learned that in Amazon. So a lot of people talk about putting Amazon values specifically frugality. The highest level of understanding the Amazon near value of frugality is not resources. It's not money. It's not trying to be poor. Okay? It is actually time because you only have finite amount of time to do, to have a, there's a big demand in finite demand that you need to, so you need to prioritize. So I try to limit it within three scopes: my family, my work, my customers. So I think once you start compressing that, you start to be able to find more time to work on that. So I, that's probably the way I would think of when, once you have family and you need to redo this and over the 12 years as I've been thinking about it, I'm also trying to work through what are the different configurations? What's the best way to sustain myself while building this, you know? They always say that startup is like a plane, you know, coming crashing and you need to rebuild a plane so they can fly back up like James Bond, right? So yeah, so that is the probably the hardest part.

(16:29) Jeremy Au:

Right.

(16:29) Bernard Leong:

Yeah. And also, of course, if my wife does better than great, you know, life is easier, but obviously it's never easy for two entrepreneurs in the house. Yeah. That's it.

(16:37) Jeremy Au:

I mean, that's, you know, very few double entrepreneur households. Right. I mean, I think, I guess there is 99 and theAsianparent. so

(16:45) Bernard Leong:

Yeah. There is Roshni, you're talking about, right? Yeah. In fact, Singapore is actually have the stats have the highest number of couples. They usually in the same companies, yeah.

(16:53) Jeremy Au:

Oh in the same company?

(16:55) Bernard Leong:

In the same company, it's very common a lot, right? But actually two couples doing two different companies, it's very rare. There is for, yeah. Yeah. I think for those who know, Yuying from Esevel is my wife, so

(17:05) Jeremy Au:

Previously, a guest on the BRAVE podcast.

(17:06) Bernard Leong:

Previous, yeah. Yeah. Previously, I guess on the podcast as well. And I think we have decided that the only startup is the family. Yeah.

(17:14) Jeremy Au:

Ooh. So three startups.

(17:16) Bernard Leong:

So yeah. And in the family, we have kids, right. We want to be sure that they all do better in their lives as such. Right. And I think the way we think about things are slightly different. We have a very different set of ways of making, thinking of how startups should work. So I used to make this joke. It's a joke, okay? So I always say, you all heard of HTC, the founder this is lady who Taiwanese lady who ran, but what you all don't know is her husband is running another company called VIA, Very Important Architecture, which is responsible for a lot of semiconductor architecture, is a billion dollar company as well. So I think we are not going to be, and we never think we are couple running the same company. So let's just each go his way. Maybe she'll become more successful than me. I can be a house husband after that. Oh, you know, so I think this is the part, this is the difficult part of The thing, but of course we talk a lot of the time. I find that being her advisor is easier than just going full on on it.

(18:10) Jeremy Au:

Yeah. No, I think I read this book I think Brad Felt wrote a book about relationships and startup couples. So I remember that. And there was a time that when my wife and I were both entrepreneurs at the same time as well. And it was bonkers, but of course no kids at that point of time. I'm so very thankful for that.

(18:24) Bernard Leong:

Right? They see you day to day and they are on that, but actually, sometimes your kids can be very helpful to you too. I'll tell this really short story and actually, because I've been recommending people to read this book called The 38 Letters from J.D. Rockefeller to His Son. So it was actually quite weird. One of these days, my eldest daughter and my son asked me to read the book to them, the letters. So sometimes I think they were trying to help me because I think I was still, I'm going to start a company, but I really haven't locked down what I want to do. And I'm trying to think through things. So they made me read the letters. As I was reading through the letters to them. Rockefeller to his son, I read about how Rockefeller started his journey, trusting the wrong partners, making the business decision to acquire. How do we handle this competition? And I actually have a very deep understanding of Rockefeller because I read Titan this most definitive biography. And one of the things is Rockefeller is a pretty simple man. He doesn't need to be in demon mode. He basically worked from home three times a week.

Okay. This is the guy who runs the largest railway empire. If his wealth net worth today is probably 850 billion or even to a trillion. And his only thing to do is that three things, family, work, customers. So when I was reading those letters, I started realizing, actually, these are the things that this is what startup founders should be like, and then it started to make me really get back into the groove on that. So having kids, they may see you from day to day thinking they also see their mother day to day, their, her struggles as a startup founder. So I think if you do it the right way, or maybe I was lucky that they started to make me realize it. They start, and then my eldest daughter started to be more curious and started reading Titan and all the other books. I think she's known on to the Charlie Munger book as well. So you'll find that even kids can be very helpful to how you think

(20:03) Jeremy Au:

Yeah.

(20:04) Bernard Leong:

About being an entrepreneur.

(20:05) Jeremy Au:

What's interesting, of course, is that a lot of folks, they're having families and they have to make decision about whether to become an entrepreneur. How do you think about that? Because it's not an easy trade off, right? Like you said, you know, you put food on the table you know, this, you know, relationship negotiations.

(20:17) Bernard Leong:

I think when we did that, so the high level way I would tell everyone is to do a PNL with your other half work through the PNL and leave no expenses missed. Okay.

We actually have a pretty well documented spreadsheet that we know what targets we need to hit for the family, regardless of whether that's like the first line of defense. Then of course, savings is key. And what are the trade offs you need to make? The trade-off could be less tuition classes, less but then what do you have to supplement it back? So like, for example, my eldest daughter this year is having a a PSLE, right? So all her, the resource attention is on her. How do we deploy those resources? So I think these are things that you need to think about. What are the trade offs in a more generic way and always The way is to try to hit the targets faster than you can so then you don't get caught in a situation that you don't want to be in on that.

(21:08) Jeremy Au:

Could you share a little bit more about how to go through that accounting process with your significant other? Yeah.

(21:17) Bernard Leong:

So what we did was we took last year's data, so very data driven people, so I took the last year's data. I took all the expenses, insurance, everything, as I put it all in the expenses table, like your monthly groceries. You probably already have some average number that want to and plow it in, right? So you will probably get a sense of how much you are spending, then you work on your revenues because now both of us are working as entrepreneurs, but I'm also working for side projects as well, consulting, teaching. So those are the incoming revenues. So you try to balance against that and making sure it's zero.

So. The key to this is to be ahead of the curve to try to make sure that there are enough projects that will be there until you hit the milestones of your startup. I think most startups founders have to be very mind that before they reach series AB, they don't have a pretty decent salary. So there's certain minimum numbers and I will just keep to the market rate. I think my wife pays herself lesser than the market rate. So, the question then is how do you balance P&L back and forth and really look at it when if you are like three months ahead and you know that You're short. You should start to figure out what to do as fast as possible. Yeah. And then what are the things that you can rinse and repeat and you can actually do better on that. But I think generically, where usually things will happen is you cut costs first. It's like a startup, right?

You think about it, right? You're going to be investing in building a startup. So hence, you need to do a cost cutting. Then you just do a revenue PNL. So everything just goes down to being very good with your PNL. I think it's something that couples should do. Especially this year's PNL, I went down to every number. Literally, like I actually went to check every number. I go into all the bank statements.

Yeah, I was even thinking of how to use ChatGPT to organize the thing. So I was trying a few things on it and then just test it and just try to figure out what are the things that you have to do and what are the things that you may not need to do. And then, what is the next steps? So, and then of course there's a definite plan on there. And then there's, they probably just give it a try. If it doesn't work out, then what's plan B? Plan C? Until plan Z.

(23:20) Jeremy Au:

Yeah. You previously mentioned the first love problem in the context of startups. Right. And now you've done several since then. And so you're off to build something new now. I'm kind of curious what are you interested in?

(23:31) Bernard Leong:

So actually it's a problem that I've encountered through both as the buy and the sell side of enterprise technology products. So let me start from the buy side when I was a CIO, CDO. Every company out there are looking for someone like myself to do digital transformation for their companies. Technically, what they really do is you, it takes three years to get it done and you have to have the stomach to invest in three years, right?

So the first year I have to hire the team, I'd hire a solution architect, I have to hire programmers, I have to build maybe a programming team outside of Singapore because of the high costs here. And then subsequently second year. You have the team, you start to manage the team, build the initial products for that. And then after the third year, everything starts to move. And then at the same time, you have to manage the legacy infrastructure. So one thing that came out after my role in Warhub and actually, we actually were much faster.

And one of the things that actually made it faster was using copilots and Code Whisperer for programmers. We hired a team in Vietnam. We put a manager who works for us, Vietnamese from us, 12 years back to Vietnam to run that team. We set up that digital development team there and we first tested with Copilot. That's pre-GPT just to be, their productivity increased by 50%. So we were able to develop apps so much faster than it was, and the deployment was great. It was what the, it's probably one of my last products I do for a company, which we went to hockey stick growth. It was actually a very basic procurement system that goes all the way into the finance.

I won't share a lot, any details about it, but the thing is actually, you try to get the correct financial information at the correct time for the stakeholders. So it dawned on me that one of the other things that happen is to deal with a lot of these enterprise software. And I would, there's a class of software called enterprise resource planning software. And if you have dealt with one of them, which is the longest version, and I shall not name names here. It is still running in, I think, a 1990s to 2000s UI interface. If you don't buy that particular ERP system, it's almost like you get you will get fired for it. So, you know, the joke of you don't get fired if you are using IBM, you know, that same thing is happening.

(25:39) Jeremy Au:

Yeah.

(25:40) Bernard Leong:

So, when ChatGPT happened, so I'm going to marry this whole digital transformation ERP question together. I realized something. I could actually speed up that digital transformation by 10X. And at the same time, I could rebuild the next generation of ERP systems, or I call it business operating systems. Faster, better and customizable. Let me give you one clear problem for all ERP software, whichever company you want to name. Problem number one, just to change a very simple workflow, let's say Bernard to Jeremy to Adriel is one workflow. Okay. Suddenly we decided that actually Bernard approves. And then we want to understand to either, either or to change that workflow costs you at least 10K.

(26:22) Jeremy Au:

What?

(26:23) Bernard Leong:

And you have to pay a freaking consultant to do. And to be quite honest, when I actually privately get my solution architect to check how to do it, we didn't want to, we didn't, we actually know how to do it, but we're like, well, the CFO is not going to agree to this. So we were like, Why is this so extractive? First, the license is 3k per user, 20 percent off your entire payment, which is almost a few hundred K dollars is on is to pay them for support. And then on top of you pay a consultant to do this. Come on. You and I are in the software side, right? I built software for so many years. I'm like, this is., It's ridiculous. And then I started, of course, doing a lot of work on GPT actually. In fact, actually, now for me to code, surprisingly, remember the story that I have to write the algorithm, wrote a code, and then my com science friends say, Hey, it's not scalable. Let us help you to rewrite this into a better version. I started doing a lot of those and I suddenly realized chatGPT is a great coder. So the question is. So the high level of the idea is, can you build a next generation, open business operating system that allow that makes companies optimally efficient, improved productivity, and continuously adaptive? The continuous adaptive is what all the current ERP systems cannot do because to make one simple customization, because what they're doing is they're trying to confine you to their software framework. So all the big companies, including the ones that I used to know ,some of those in local, the big ones, they are all stuck with those systems. And they cannot change it. And they had a lot of technical debt.

So the high level of it is to try to build it, but I'm not going to target the current ERP customers. It's a waste of my time. What I would like to do is focus on the small medium market, but the medium of the small, medium market. And I have a counter positioning strategy to what their current business model is. As you can look behind, I actually have read the only history book of one of the ERP companies, the only history book. I just studied a lot of their history now. So I have a better sense of how they actually made it in the first place and really thought through what it is. So that's the essence of what the idea is.

(28:18) Jeremy Au:

What is the history of systems? I mean,

(28:21) Bernard Leong:

Yeah. It's actually interesting because it started off with people from IBM wanted to build customized software for very specific customers in financial accounting. And this company was in Germany and they were actually getting referrals. It was only six engineers in the company doing that. They only hit the U. S. market only in their 10th, 13th year I mean, anybody would know which ERP company I'm talking about, but I think this is interesting. And I think recently was it YC that actually talks about the next generation ERP that they were in one of their startup RFPs. It is an interesting problem, except that if you are not in enterprise sales, like I do, You're going to be very difficult to penetrate through. So I've even thought it true.

I think first-time founders look for product market fit. Second-time founder look for distribution, right? So I really have a pretty clear distribution strategy in my mind, how to sell this. So it's, so, so I think it's a question of how do you think through this. Actually, I was thinking about it for a while. If anybody had read Hamilton Helmer's Seven Powers, the most powerful, ERP company that I talk about, they have a power, it's called switching costs. So the question is how do you counter-position against them? So if you are a startup and you cannot counter position against the current big Goliath, you're in trouble So the counter position can only come from not just the technology, but the business model Which is why second-time founders need to look for distribution on that.

(29:38) Jeremy Au:

Let's talk about it because we're both serial founders, right? So, some notes here. Well, what does it be to be like, you know, so you're like second-time founders focus on the distribution, like what are the learnings that you have, because you know, you're a veteran, you're coming back in.

(29:50) Bernard Leong:

think I'm always a beginner Cause I want to be very fair to everyone out there. Just because we are second-time founders, doesn't mean we're entitled to anything. In fact, a lot more pressure on you. What I would say is the first time round, you try to look for product market fit. You try to build. Focus a lot more on the product, less on trying to, on your customers. I think when you are a second-time founder, you realize that actually, you don't need to build the best technology. You just need to build a good enough technology. The question, the first question you should probably ask, are there customers? The answer for me is I do have customers. Basically, what happened was there were CEOs of second-generation, third gen, fourth-gen business owners. They don't have ERP systems. They have very difficult digitization needs. And all are coming to me for advice. And I started listening to everyone and I realized actually that's what you really need, but you have to do it in steps. And you cannot customize for everyone, and you need to figure out how to mass customize for everyone, which I believe generative AI can, except that there are still a few hoops that you need to jump.

So as a second-time founder, what I'll do is, okay, the technology, I've tested it. In fact, I can tell you between Gemini cloud and Open AI ChatGPT, which one gives the best code generation? So you have to solve, have some kind of mental model where you get your customers and you need to work backwards from your customers. How do you hit the buttons to scale? And the one thing I learned, and this is the reason why I thought that my role in Amazon fills up that missing piece I was looking for to come back to the startup. So if you have worked for any tech company, which I really urge everyone should think about, you probably have figure out how 50 million to 1 billion. And that is the skill that you need to get, acquire. And of course, all the relationships, which I do, I still have very good relationships with my Amazon colleagues. You have to and how to you'll be forced by very grueling targets. In fact, Amazon is like a Chinese company. It's just that it doesn't do it yet.

I was under pressure just to do like how to get from zero to 1 million for one particular AI product. And that taught me how to prioritize and how to sell fast enough while actually the product was actually not very ready for prime time. So once you can think it that way, it will become much easier to think about distribution. So to me, the second time is actually a lot of focusing on the skills that I didn't acquire the first time. So it doesn't mean that the second time founder means that you can do, you also make the same mistakes and you have to be a third-time founder. But I think for serial founders we also need to think through how to do it like you think of the podcast, for example, is a second iteration of SGE, right? Where did we go wrong? But we sold. We did acquire a certain amount of audience, right? If I were to tell you that when I was telling this podcast idea to one of the notable entrepreneurs in our region, and he told me, Bernard, you're not a hot chick. Your podcast will never be large. I'm serious. I have that quote right on the, yes. he has to end up appearing in the show.

(32:39) Jeremy Au:

Yeah. That's so funny. Yeah.

(32:43) Bernard Leong:

Just because I have done the podcast doesn't mean I fight on my podcast I can tell everyone of every 10 approaches I made on my podcast, only one come back. Those guests that you think that you can never get, I actually cold call them. I didn't even go to a referral. It's the same thing. What I say that when it says it helps me with my entrepreneurial muscle, it makes me a beginner. And I think whether you're a first time, second time or endless time founders, you have to remember you're always learning. If you're not always learning, don't do a startup job because there's always something that is pretty strange that will come in your way and you'll be like, okay, what should I do? Yeah. You're always a problem solver. So I think when you think about it, that is. Yeah, think about that. So can you believe when people tell me that it's not going to work? In fact, what the ERP, some entrepreneur when I was just trying to formulate the idea, he just told me, I'm going to put a shot on you and you're not going to make it. Great. That was what the other two guys told me the last time around.

(33:35) Jeremy Au:

I mean, I think there's a truth to the matter is that betting against a startup is pretty much the most probabilistically correct bet to make.

(33:41) Bernard Leong:

You know, it was funny because the drone flight, that day for for SingPost from Pulau Ubin and actually IMDA couldn't do it and so we partnered together and they put a bet on me and they internally lost that bet and you know what they told me, they said, we thought you cannot make it for that drone flight, but you know what, we didn't expect Wolfgang to put an entrepreneur in front of them.

(33:59) Jeremy Au:

Yeah. Oh.

(34:01) Bernard Leong:

Yeah, they were betting against me, and we were laughing over it, we popped the champagne when the flight was done. But, that was what they said to me, they said, we didn't expect Wolfgang put an entrepreneur in front of them. They think that CWAS will not let me fly. I did. I caught them on an error. And you know the error? It was on physics, first principles. Yeah. funny when people say what you learn in school doesn't apply. I think it does.

(34:24) Jeremy Au:

Do you think there's more mental self pressure for a second or third time founder, on themselves? or is it easier?

(34:31) Bernard Leong:

Yes, all the time. Yeah. It's like football managers have pressure every game, right? But you need to think of it as every game. You cannot think of it as the entire league. Yes, you want to win the Champions League, you want to win the Premier League, right? It's the same, right? Every time you go through the next cycle, it is another game. It can be in your favor, it can be stacked against you, You do not know. But what you can do is have a plan, get punched in the face, rework on the plan, and then go in again. Until, of course, it cannot be done, then, you can call it a day, right? I think the mentality should be, don't be too hung up on the first idea.

I think that was actually very helpful for me. It's also part of growing up on that. I think you will be under a lot of pressure, but you should just think, okay, let's just do this. Conventionally, people are trying to find out what I'm doing in the enterprise AI space. My focus now is very simple. I just wanna build a product first. Build the first deck, build the correct customer base. I don't want to fundraise so quickly, and even if I do, I will only be talking to Angels who want to back me.

(35:30) Bernard Leong:

I don't need to go all out yet. I think what I need is to figure out does the strategy work? Because once the strategy work, it doesn't matter. Everybody will come to you. You don't need to go to everyone. And everybody who wants you to fail will want you to fail. Everybody who wants you to succeed will want you to succeed. So, you just have to make sure you have the right capability. And please don't fake credentials. It doesn't matter. Yeah, sorry. Yeah, I have an actual Cambridge PhD. And I work in AI. Okay. I also, just like any other people out there, I'm not going to get the VCs throwing money at me. Everybody ,the best thing in startup world I like is that everybody is in an equal and zero state. Fair. That's what we should be thinking in terms of second time founders.

(36:09) Jeremy Au: Yeah, I agree. On that note, could you share about a time that you personally have been brave?

(36:13) Bernard Leong:

Brave, huh? I think it's the time where we shut down Chalkboard. So, my co founder and I, we had a pizza party, we invited everybody on there, and plus the fact that because I own SGE, it's going to be a public. I'm going to get hammered in the public news for failing. Then after that, of course the news got out, and then we failed. Couple of things. The first thing I did was every investor who invested in the company, whether even it's LP, I apologize to them that we fail. You can ask, there is actually he was one of the investors. So, true Joey's fun. I think that's important, but it cannot be something that you hang over for the rest of your life. So that's the first thing.

And you will have to deal with a lot because if you're a media person, you get a lot more public failure. On that you just have to hang a chick out and just do whatever it is. Do what matters at that point in time I gonna have a kid coming another eight months time, I have to gracefully shut down the company. Took me three years to do it because of a lot of administrative things that we didn't do when we were starting company taught me a pretty important lesson how to keep track of things. And when I was running Analyse Asia as a company, I actually was very disciplined with accounting, with everything else. So, things that look very small to you in a startup, you develop a system to deal with it. So it took me three years on there and I was not ashamed. Actually, the bravest thing is, of course to keep our face and meet all the former investors, former startup founders always be observer around but I think after about one year plus I moved on.

Yeah. Then I just thought about, I think that what is here is you have to be brave to own up that you failed. Number one, I think you have to be brave about everything else. Then I didn't even know whether I will last in a corporate job. That was a bad place on me, actually, on that. They bet and say we give you two, we give you six months and then I took the bet to 12 years. So, I already won 99.99 based on the expectation curve. Like, let me run your expectation curve. There's like, they were like five years. We can give it to five years. No, it went on and on. And then recently we had a private WhatsApp group. He's like, Oh my God, it's 12 years. How could you do this? I'm like, yeah, but I think you have to, I think people always talk about being brave about setting up venture, but I think the biggest bravery, I think that I ever felt, which I've thought about it because you asked me this question is that it's the bravery to realize when you fall and just be responsible about it.

(38:33) Jeremy Au:

You know, I think being responsible for it is hard, right? Because there's so many reasons why failure happens. There's the environment, there's other people, there's yourself, there's investors. What do you think is the right way to define being responsible for failure?

(38:43) Bernard Leong:

So when you fail, basically the first thing we did, the response, we fired all the employees is something that I took it with the rest of my life. That was once I think somewhere around 2018, I met one of the Thai engineers I fired, we took a picture together. He's actually through the experience. He became now a head engineer of a big startup in Thailand. So, people who you're fired, but because really that you're running out of cash, we have to come up with the remaining, we basically already know that where our burn rate is going to be. We just basically made a pretty conscious decision not to get ourselves into debt, but enough to pay back suppliers, pay back your employees. So that's number one, take care of people first.

Of course, going back to the investors and say, you fail that you have to own up. No choice. Some people, so a lot of the founders will say, Oh, well, they may never back me again because of the kind of Asian culture about failure, right?

I think that has changed a lot in the last 10 years. Some of the investors do ask me, Hey, when are you starting a new company? Right. I'm like, yeah, give me some more time to think about this shit. Now they are not, I don't know whether I go back to them. They will even bother. I think that's the second part.

And then the last part is make sure all the things you do to shut down the company. I told you it was three years, right? I, Probably spent probably five digits to sort out all the accounting, the whatever, and got it gracefully shut down.

(39:58) Jeremy Au:

Yeah.

(39:59) Bernard Leong:

So I guess, so, I think the late Patrick Turner from Incel once told me if you can gracefully shut down a company, you should put that in your CV. I think it's hard. Actually, everybody thinks that a starting company is easy. Starting a company is even harder.

(40:11) Jeremy Au:

Yeah. On that note, so much for sharing your journey. I really appreciate it. Let me summarize the three big takeaways. First of all, thank you so much for sharing about your early education and career. It was fascinating to hear about your experience as being inspired about space and the research of that. And beating the odds to do a PhD, but also explore machine learning and coding, and also actually your early experiences building your first few companies. I really liked the way you said it, your first love problem, you know, and how you were thoughtful about, you know, which companies you want and also which companies you eventually took on to build.

Now, secondly, for sharing about ERP and what you're building. It was interesting to hear a little bit about the history of the ERP systems that, you know, most companies use today, especially when it's super large, but also I think the challenges that they face and, you know, kind of like buying, implementing, and getting exploited extractively, and what you're looking to build to make this a different experience, right? Especially the rise of, ChatGPT and Microsoft CodePilot.

Lastly, thanks for sharing about your experience and thinking about being a serial founder. I thought it was fascinating to hear about how you compared it to being a football team manager about how you know, you have to think about the mental condition for every game and just taking each game as it comes. So I thought it was really fascinating to hear about different experiences, rightWhat's it like to build a new company again? How to negotiate with your spouse, especially when you have family and mouths to feed and about how to, you know, wind down a company gracefully as well. So these are all, I think, experiences that have become more clear over time.

On that note, thank you so much for coming on the show.

(41:37) Bernard Leong:

Thank you for having me on.