Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
The Serial Mobility-Tech Founder | Brijraj Bhuptani @ Spry
635 Plays2 years ago
Brijraj has explored the mobility space from multiple perspectives. His first venture Ridlr was an app to help people navigate public transport. He started Spry which is a full-stack digital health platform for physiotherapists to help their patients lead pain-free lives. Spry is a great example of a made-in-India product and Brijraj is targeting the US market as its primary revenue stream.
Know about:-
- Solving for Indian traffic; building Ridlr
- Introducing mobile ticketing for unreserved category
- Competitive differentiation: using software
- Customer acquisition strategy
Recommended
Transcript
Introduction to the Founder Thesis Podcast
00:00:00
Speaker
Hi everyone, I am Pritaraj, I am the CEO and co-founder of Sprite.
00:00:16
Speaker
Mobility is the lifeblood of an economy. Everyone and everything around you is the result of it. And in this episode of the Founder Thesis Podcast, your host Akshay Dutt talks to a mobility entrepreneur, Brijraj Bhupthani.
00:00:31
Speaker
Brijraj has explored the mobility space from multiple perspectives.
Brijraj's Entrepreneurial Journey Begins
00:00:35
Speaker
His first venture was Rydler, which was an app to help people navigate public transport, and Rydler went through a journey of pivoting from a B2C product to a B2B company before it got acquired by Olac Apps.
00:00:49
Speaker
Next, BridgeRaj started Spry, which is a full-stack solution for physiotherapists helping them to ensure that their patients get back normal mobility. Spry is a great example of a product that is built in India for the world as BridgeRaj is targeting the US market as their primary source of revenue. Listen on and if you like such insightful conversations with disruptive startup founders, then do subscribe to the Founder Thesis Podcast on any audio streaming app.
00:01:26
Speaker
I grew up in Mumbai for the most part of my life. In fact, the first 22 years of my life, I was in Mumbai. My mom and I lived together. So we are, she's from Mumbai as well. I've grown up with my maternal parents and they were always mom-makers. So they moved here during the early influx of people moving from Gujarat to Bombay as businessmen. So early life was very different for me to give you some context. My parents separated when I was one year old. So that's why I grew up with my maternal grandfather. And what was your grandfather's business?
00:01:50
Speaker
So they were into exports. So they used to export stationery products, sports products to Africa, Middle East, et cetera. So at that time, we depend on the likes of India, Mark, et cetera. So trade was based on more knowledge and information arbitrage. So they had a lot of information of manufacturers. They had a lot of investment buyers.
00:02:05
Speaker
And they would do the job of connecting the two, which websites do today. But that is what they would do. He also had a small pencil sharpener factory. If you remember the pencil sharpener that he used to use as kids in Jamlagar. Wow, pretty enterprising. So like you grew up seeing business in a way. My grandfather is my role model. My tell grandfather my energy. I've learned everything about life and business from him. And it's a very exciting conversation. I think when I was a little
00:02:28
Speaker
10-15 years old. I was travelling something with him. He was old, he had some problems with his eyesight. So I used to take him everywhere. And one of the conversations in the train journey was that, why do people buy stocks? I was in a Gujarati family. I came and came from a Gujarati family. And there were a lot of people in the share market, as you say. While we were a middle class family, a lot of our relatives had made some money from stocks. And I used to keep on hearing them that, oh, they are in the stock market and they made money. They did this, the stock market, etc.
00:02:53
Speaker
So I very curiously asked my grandfather as an innocent child that okay what the hell is this share market and why are we not in there? That's why my grandfather told me that yeah it's you buy shares in another company and you buy and sell and if the company makes profit you make profit. I don't enjoy it. If you want to do something then have shares of your own company.
Career Transition and Early Ventures
00:03:10
Speaker
Okay so that stuck with my mind and I decided that day that someday I'll become an entrepreneur.
00:03:13
Speaker
I have a listed company. This is my second startup. My course company did not list, but list. We sold our shares to a big company. And the second company used to actually list somewhere. So I did my diploma first in electronic engineering, and then I did my bachelor's degree in electronics engineering from Bombay University. From Thadamal Chani College, which is one of the better colleges in Bombay.
00:03:32
Speaker
So this is also interesting story that how I want to contact. So I was always supposed to be electronics guy. I went to the U.S., did my master's as well. After your B.Tech. Immediately when I graduated in 2002, market conditions were not the best right after the dot com works. So I really thought might as well study a couple more years and see what happens in the U.S. was a reach at that point of time.
00:03:50
Speaker
So I went to the US, I did my master's, and when I was about to graduate, I got two job offers from a company called Qualcomm. Qualcomm is like the Microsoft for wireless companies. I was a wireless engineer, I did my electronic engineering, wireless communication. And there are two job offers, one in RF engineering, where you actually work on the video, you work on the hardware, and the other one is software engineering. You make video on devices. I said, it's a no-brainer in India, we've been trained, right? You do what you've pursued, why do you have to change fields? It's called, akin to changing fields. It was a no-brainer, they're okay, I'll do RF engineering. I took up that job offer.
00:04:19
Speaker
Few days before joining, there were a mass layoffs in that field. And I was devastated. I said, okay, now I'll have to go back to India because how can I find another job before I graduate? And I would have my visa. Obviously, I got the other job offer to the HR next to call me. And that was saving grace for me. He said, no, we're happy to take you to software, although you wanted to do hardware. I said, okay, what choice do I have? Let's do software. It's important to stay in the US, get that H1B visa that we're in place for and figure out along the way, then I'll switch to hardware.
00:04:42
Speaker
According to the software, we were one of the first ones in that time, we were pioneering video on handset. So this was back in 2005. Qualcomm was a chip company when they started as a cross-seller value and they started selling software stacks also on top of the chips. So they were starting, so this was a very new revision. It was called the multimedia revision. Cameras, no one knew, the cameras were omnipresent today. One in 2005 was using cameras on phones, video on phones. It was not that common. So that group was like a startup. It was a very new team.
00:05:07
Speaker
We had bought over some company, we had to merge with them, we had to handle their code. I had to travel places because when a phone was being launched, video would not work. So incidentally, it turned out to be a great experience for me. I was in a big company, protected. At the same time, I got the experience of a very young and energetic team. And it was building the future of mobile. In 2005, video and camera was the future of mobile. Today, it might not seem so. 2005, it was, right? So then what next after that Qualcomm stint? How long was that?
00:05:32
Speaker
I always knew that I had to come back to India. And two years into my journey, I quit Qualcomm. One of the 0.2% people in the Qualcomm did that. And you quit to join a startup. Yep, I only applied to startups. So I got two job offers at the end of a six month search, interviews, etc.
00:05:48
Speaker
and one was in LinkedIn and the other was in Nextwave. Nextwave was developing the 4th generation Wi-Fi chips, Wi-max chips for that. Ideally, if I had the right financial advice, if someone could have told me what the future is like, I would have obviously taken LinkedIn. LinkedIn was a CTC or series B funded startup, that's UC where they went. DRockman must have been still running it at that time.
00:06:05
Speaker
He was the CEO at that time. He was the CEO and founder at that time. They didn't have a very big company, but the advertising email that came to me was a top cafeteria of free food. That was the advertising email I still remember that they sent to me. I was one of the first few Indian users as well, I remember. I said, no, software can't go away. We do software. The hardware background, we are electronics. Why would I want to work in software?
00:06:24
Speaker
let me switch to electronics and let me go to next wave wireless which was the Vimax which was developed by next generation of chipsets and the idea was to develop Qualcomm at that time and I ended up joining them for two years. That's how I switched from Qualcomm to a startup.
00:06:40
Speaker
for two and a half years. Great experience for me. Found it difficult initially, so you go from a company which is highly protective, everything is ready made. For example, I've written a piece of software and I have to go and test it. I just go down to the lab, plug my phone into the, we call it usually the BOA machine or whatever it is, the one which simulates wireless communication. And then you're ready to piece, you just play with your buttons, you test your software, you come back to the lab. You come back to the office and you deploy the software. Oh, but next year when you go to the lab, first of all, the lab is empty, there's no in the lab.
00:07:07
Speaker
There's no test harness. How do you test it? How do you simulate it? So then you have to create everything. In Qualcomm, when I was stuck in a problem in the first six months, when I was transitioning from hardware to software, I had colleagues who I can rely on. Here there's the one. There's very few people who you can have and everyone is very busy. So you have to figure out a lot of yourself. So it's paint.
00:07:23
Speaker
I used to work 14, 16 hours every day by myself in the lab late in the nights. I was determined to figure it out. So after the first bumpy three, four months, I built up everything. So they were also building video on Vimax. I built up the entire stack for them. And I became a rockstar engineer again for them. And then obviously, because in a startup, you do everything.
00:07:42
Speaker
So whenever they were presenting to the customers sprint at that time, or we were presenting in Las Vegas in the consumer electronics show, I was at the forefront of things. So while I would not be pitching,
Traffic Solution Startup and Challenges
00:07:50
Speaker
I would be hearing the guys talking, the business guys talking to the clients, pitching to them, talking to them. And it was doing the CS show.
00:07:57
Speaker
that I told like I'm done now if you start going companies I got out of in the break I got off the loop I called my uncle who is a telecom engineer and he's an entrepreneur first-generation entrepreneur even from the US in 85 to India I told him okay I'm coming back to India and I want to start up something for myself right so I'm done building things for others so that was a trigger point obviously it took me a long time after that to get started so I walked in next year for two and a half years so
00:08:19
Speaker
the first two years in the US and then during my stint and next year I told them that it was before the Lehman crash I told them I want to move to India because I was around 28, 27, 28 years old so I was also thinking of getting married and better to do it in India rather than keep on shuttling so I moved to India I started the India Centre because I was moving to India and they didn't want me to leave they said okay why don't you start an India Centre
00:08:40
Speaker
We let these two some get some development and for me and we don't lose you as well. We set up our development center in Mohali for that matter. And I should spend 15 days in Mohali, 15 days in Bombay for that case. And we build a good team. But unfortunately after the Lehman built the crash, funding tried out. And more importantly, the technology that we were building, no one picked up. So we were building. So 4G had two ways to go. It could either be LT, the long term evolution that is pretty well in today, or it was Vimax. Sprint was the only company which was opting for Vimax.
00:09:07
Speaker
Now, I didn't have the foresight when I picked up the today's skips. If I mean I do today's skips, they have asked me for vision, the trend and all that. I didn't have that. I just thought it's a cool thing to build and let's go and build it. But it would have been hard for anyone to predict, right? Whether LTE will be the winner or by Max.
00:09:21
Speaker
No, not necessarily. So if you look at the stock price, it was Spring dual lacking WiMAX and the others, right, the Singulars and the AT&T. And the stock of Singulars and AT&T definitely went up after the iPhone launch. So they were on par. AT&T, I think Apple launched with AT&T after the iPhone release, right, because the initial partner Spring was not. And those guys just took off, right. iPhone pretty much decided who will win that.
00:09:42
Speaker
iPhone decided in some extent. iPhone has decided a lot of things in the future, right? And time form would have taken by max. I think we would have been, I would have been a billionaire, a millionaire. But yeah, one fine day, they shut down operations and whatever assets they had, they just kept it on the side and they silently, evenly, softly told us, why don't you look for something else? Unfortunately, a couple of things happened at that time. My mom was diagnosed with cancer.
00:10:02
Speaker
I said that I don't want to take up anything full time. So then also it's the right time for me, not taking up full time. Not that I have to take care of our full time, but being around the right time was important. I said, why don't I do some experimentation with my own company at this point of time and figure out and if something great happens, great. So the ideas I initially had did not work out at that time. I wanted to build something in transport, in traffic communications, et cetera, but could not build that out meaningfully. I was not fully focused as well. Like a hardware thing or what?
00:10:30
Speaker
Yeah, it was a combination of hardware and software. So I'll do one thing about me.
00:10:34
Speaker
When I lived in the US, I became a very punctual guy. I was never late to meetings. And it was something one of my seniors taught me in school, that we had 10 people meeting and people showed up five minutes late. And then a typical Indian said, you might think, right, we're five minutes late. He said, all of your five minutes, 10 minutes each, five minutes, 15 minutes have been wasted. And that's the math he told me. That made sense to me. And I said, yeah, it's actually cool to be on time and not take time for granted. So when I came to India, obviously when I was working with next year also and all that,
00:11:02
Speaker
One of the things I saw was that people using traffic is an excuse. So between 2002 to 2010, I think we had the first stock market move. We had the stock market move and the number of cars on the road just went up like crazy, at least in Mumbai and Delhi, those places. So people would very conveniently use traffic as an excuse. I am late because I was back in traffic and I am late. I'll come in the first half or the second half, sir. Am I supposed to wait like the entire half a day for you to come?
00:11:24
Speaker
I've seen people in my family think the road is empty and we are late and they think we're stuck in traffic because there's no visibility. How do you verify that claim? I also used to be, because I have a stickler for being punctual, if I had to be at any meeting, I would reach in early, quite early, because I would plan it ahead or I would take the train. So I said that if I don't build a system for myself, which would give me real-time information into traffic in the cities?
00:11:43
Speaker
And I might not be able to avoid traffic, but I can at least plan my thing. But if it takes me 45 minutes, I can leave accordingly. If it's an alternate route, that's possible. Then I can take the road accordingly. So it just sounded like a good project to do. Classic Indian problem. There's a big problem to solve, but there's no in-prime place.
00:11:58
Speaker
And in US, I used to use these systems. So when I should travel between San Diego to LA, I would look up traffic information on the highway. The highway was blocked out in the interior roads. I would not travel at all. I had the optionality with me. So I would not do. So I could have built that in India while I was available in India. So then I started dabbling with that idea. But again, a very difficult problem. I did not even know where to get started. So I just got lost in the whole thing.
00:12:19
Speaker
How do I get GPS? How do I get sensor? Because there's no sensor data. There's no camera feeds. I went to the police and said, why don't you give me a camera feed and I will actually analyze this information and give it to him. He said, then there's a security problem. In 2008-09, if you remember, we had some oblaces in India. So security was also very high. So they were sensitive to that. And I was not smart enough to figure out how to break those barriers. But coincidentally, what happened is one of the companies I was working with, he was a GPS company.
00:12:44
Speaker
Yes, you're a GPS company. So the model that I had implemented was I'll go to a GPS company. I'll tell them to give your data to me. It is sitting on your server, right? You're doing nothing with it. To that time, we didn't have big data on any of those analytics, et cetera. So it is sitting on your server for no reason. I'll pay you some money. Give me the data. I'll convert you to traffic information.
00:12:59
Speaker
And that sounded very cool because you're not putting your own hardware, you're not inputting the Kpex. And I thought I'm the smartest guy in the world, right? I'm waiting a traffic system without deploying Kpex. The government should love me. The police should know me except it's never the case. But so I started working with the gentleman. He was named Alok Kumar. He was running his company.
00:13:15
Speaker
And she was also trying to scale. She ended up getting an offer from US-based retailer called Sears. So Sears is one of the biggest. Used to be one of the Amazon of the 19th century, right? They were really big. Yeah, they had that mail order catalog. I think they pioneered the mail order catalog.
00:13:30
Speaker
So whatever Amazon trust, if you took the CS playbook and look at the Amazon playbook, Amazon exactly copies that. A lot of Firebook which is the market base, then they did private labels, they did private labels, etc. The category expansion, having your own branded categories. Amazon is exactly copying that playbook in the online world, at least in the e-commerce space. So CS was obviously on the decline post-Amazon with 2010-11. They had been acquired by a private equity hedge fund guy.
00:13:56
Speaker
and the hedge fund guy wanted to transform Sears into an online company like Amazon.com. Now this new hedge fund person coming and taking it, the staff is still legacy. I don't think the UVA never happened. Financial guy coming in and taking it over and bulldozing his aid didn't really work beyond a point. So we hit some milestones with Sears.com, we hit 2-1 billion GMV, the mobile app were 2-12, but it never scaled and Amazon was a rocket ship, right? It was moving at a much more faster pace than all of us could ever imagine.
00:14:42
Speaker
So tell me about that journey of starting up.
00:14:46
Speaker
from idea to go to market, to product, market fit. Ideally, when you're doing a business, you have to be disruptive. I was not that kind. And that's something I've changed for the last 10 years, doing my entrepreneur journey, something that go all in. I wanted to do it, but not want to. So it's like the guy standing at the edge of the swimming pool, wants to swim, but also don't be scared of jumping. So during this two and a half years, my mom recovered from cancer, as I mentioned.
00:15:10
Speaker
I got married, I had taken my first home loan, buying a house in Bombay, that time was, even today is expensive, but that time also it was expensive for someone with my background. I did that. Yeah, it's a once in a lifetime thing. Yeah. And then I paid off the home loan also, whatever little bit, $4 saving. Then we decided like everything is taken care of and we didn't have kids at that time. This is the only time I can potentially start. So how long will I keep waiting? If I have to do it, I have to do it now. Very inspired by million dollar, I think the boxing movie million dollar baby right there.
00:15:39
Speaker
in which he says that you need that one shot in life. And I thought that I'll regret it if I don't have that shot in life. So I've done everything possible to prove myself to the good employee. We'd call comm, we'd next wave, and we'd see us for the matter as well. And this burned my conversation with my grandfather and all that. So if that has ever come to fruition, this is the time to do it now. You can't wait. Also, another trigger moment happened at that time. So as I said, I wish to do these things on weekends, night, and Twitter was getting popular at that time. This was the day when Palasayp Thakri died in Bombay.
00:16:09
Speaker
Historically, it used to happen that a politician had passed away or something happened. The city used to go buzz up, they would be shut down, so there would be lock-ups, lock-downs everywhere. That day, the city was actually quite peaceful. There was no activity on the road, there was no protest on the road, but people were generally panicking. I was on Twitter at that time because
Brijraj's Traffic App Development
00:16:26
Speaker
we had all come back, when all of us had come back home, we were not working in the office at that time, we would all come back home fearing that there would be backlash on the road.
00:16:32
Speaker
actually nothing was happening and I saw I was seeing on Twitter that there's a set of people who are panicking and there were a set of people who are saying that there's no don't worry take your own time don't panic go home but take your own time don't there's nothing on the road but this information there was an information they don't know each other right so I started becoming the bridge between the two I was taking whatever these guys are saying retweeting them and to these guys tagging these guys they don't worry and all of a sudden I like 600 followers in 15 minutes I started this offline thing and then
00:16:56
Speaker
Then within two days it became like 2000 followers just doing this. And the monsoons happened after that, where I started continuing with the monsoons unit, Bombware and Nightmare. People just don't know which road to go, where are the potholes. Then I said that people are engaging on traffic content. This is what I always wanted to do. So why don't I build this social traffic app? I already have data coming from, I will figure out the model to some extent of data coming from GPS companies in the mobile app. So I think those two, three things happened and I was working, as I said, in the background. I said, now the home load is done. This is the only time.
00:17:23
Speaker
but the responsibility is only increased. If you want to start up, do it now or never. So I decided to quit CS nightly and then I started out. So that's how the trigger could happen. I was lucky that I got funding immediately because I had done some work prior to this. So I took a rupee check from the angels which also further motivated me to say, okay, let's take this leap now. And that's why I took it. When you say you had done some work prior to this, what does that mean? When I was working in CS, I used to keep on doing something. I used to keep building the stack. I would code at night.
00:17:49
Speaker
building the traffic information stack, how the map would look like. I would also talk to a cab company on weekends that if they want to send GPS data. So the work there I started with Alokumma and I would continue doing that on nights and weekends and holidays for that matter. So there was some scope, I would go and present my solution to a traffic police every now and then. So you can say I was akin to doing two jobs but I was not.
00:18:07
Speaker
only money in the second year. I was just experimenting and wanting to build traffic information. So I was building the stack in the background and it turned out to be a very good platform. I built it all by myself at that point of time. I used to code at that time. A couple of cap companies at that time also interested in sharing data, which made me a national product from day one. So they were cap companies, easy caps. In fact, Mr. Rajiv, which I'm very grateful to him because he was the first one to give me an opportunity. And he said, okay, why don't you take my data? And he was present in five or six cities.
00:18:49
Speaker
And what was the product idea that you would aggregate data from Cap companies and couple of other sources and
00:18:57
Speaker
Yeah, basically, GPS was the, anyone who had GPS on the vehicle would be, and the guys would be, so if you have a GPS, but you're moving on the highways, or you're not parking, having a car is useless to me. But the cab companies who would GPS in those days, the radio taxis, I used to call them, right? The radio taxis, I used to call them, they were always moving on, they were paid to move on the road, to the main road of the world, the easy cabs, the mega cabs, the tap cabs of the world at that time, right? So we figured out a very smart business model. I said that, so
00:19:23
Speaker
Ola was happening at that time. Uber was happening at that time. Uber revolution was happening at that time. But these guys were not tech companies. They were not tech companies. And Uber was getting eye-popping valuations at that time. So everyone wanted to view Uber.
00:19:34
Speaker
Then I told everyone, all of these campaigns, he said, I'll build you a mobile app and maintain it for free for you, right? And in return, you give me a GBH data. So it was a mix of both. I didn't know. A mobile app, as in for the customer to book a cab, that mobile app. Everyone at that time thought just building a mobile app is enough, right? There's a lot more which goes into building an Uber and all that. But luckily for me, there was people that were not aware, no one invested in building own tech teams. None of these guys, which I talk about, except for Meeru to some extent.
00:19:59
Speaker
actually invested in building tech teams. So all of them were very willing to work with me. And I did a good job, whatever requirement they had. Did I help them grow their business? No, I never signed up for that. I just signed up for being the IT partner and the mobile app partner, which I continued being. I never, I thought I'll grow my business on the back of their data. And that is my trick. That is where my tricks with cap companies also started. I eventually got acquired Ghola.
00:20:20
Speaker
So I had met Dhola, CEO of our shows at that time in exchange for data, except that was my first encounter with him. So that's how we started. And that was the scale of it all, right? We were getting like 10 million data points a day across India. We were giving traffic information in 15 cities. And this was the time when Google was not giving traffic information. Google Maps were fairly okay. But that started much later. You were the first ones. And the engagement and the information on the app was brilliant. So today also the Twitter handle that I talk about,
00:20:49
Speaker
Only the Mumbai version of the Twitter handle has 3.5 million followers. It's not been updated for the last few years. It grew to 3.5 million organic followers without a single paid follow. So you can imagine the scope. The mobile app that we had developed for traffic information was the one of the first in India to get 200,000 daily active users.
00:21:04
Speaker
But what would the mobile app show? Would it show? Could you put your destination, get navigation and then also see traffic like same as what you see in Google maps? Yeah, you should do that exactly, right? So we were using Google maps as a backend layer, but just a map. The traffic efficient on top of it was overlayed by us. So we would color the road, radio, low.
00:21:23
Speaker
depending on how the cab would be moving and our interpretation of that. And then we also had data coming in from traffic police sources. So we had people sitting in traffic police control rooms where they have cameras. If an accident or something happened, we would overlay that on top of it. It was a portable accident. So it was a combination of data coming from GPS, data coming from the traffic control unit and data coming in from users because the users were engaging on Twitter, the users were engaging on the app. The users were also reported. There's an accident here, there's a jam here. Users were also reported.
00:21:48
Speaker
Okay. Were you also collecting users GPS data from the mobile phone? What Google Maps does? Yeah, we were. And we were not very successful in cracking that because users were not keeping on for a long time because batteries were not at the best at that point. Battery levels are not the best people afraid. Google on the other hand benefited because the older drivers and Uber drivers would keep the mobile on all the time. They would plug into the device, right? So they would get a lot more and they were using Google Maps. They would get a lot more data at that point of time than us.
00:22:17
Speaker
So that is what benefited them as well. So eventually, when we started with a very reliable and secure source of data, Google's data was not good. But over time, because Ola and Uber started using Google data, it became more reliable and scalable as well. And what was the monetization? Yeah, that is where the company failed. So we could have monitored. So initially, we started selling traffic information as a value added service on telecom operators.
00:22:37
Speaker
So we tied up with Loop mobile, with Vodafone. If you remember, you used to get SMS alerts. So people would subscribe for traffic. So people would subscribe at one time. You put your source and destination where you would travel on a daily basis. And then 9am in the morning and 6pm in the evening, we would send you traffic information. So these were like 30 rupees a day services. But very low monetization because 70% of the money in the telecom operator would keep for themselves. You would only get like 30% after tax or 22% or whatever it is.
00:23:01
Speaker
but at least it got us some monetization going and we had a big base of this like on loop mobile and water phone. We could not correct Airtel at that point of time, but these are the two carriers which we were working very closely. But what happened in 2000, I think 13, and there was a crackdown on this one entire vast industry, right? The entire vast industry was rekt down upon by TRAI. And this industry got wiped out overnight. So a lot of these guys, the ATM, money on mobile, all of these guys were making money from vast, started seeing declining revenues, right? And we were one of the smaller guys, we saw an impact of that.
00:23:29
Speaker
And more clearly, that was not the future, to make money. I believe this was because there was misselling happening and customers were not even aware that they were getting charged. Yeah, 70-year-old women were subscribed to job alerts. Someone who did not want advice would be getting advice. So there was no control on how one would opt-in and opt-out of a service. And it was the discretion of the circle-level manager or the city manager to figure out who would become a subscriber to become a subscriber. So if you have a good relationship with the circle manager, your product does well.
00:24:00
Speaker
We didn't have digital payment but this was the first form of digital payment site where your bill is integrated into a telecom and that's what a lot of people capitalise on. So I think we're selling crazy fraud, right? Beyond bond there was no organic usage of any product to be honest. So that's what I think and there was a crackdown and I think rightly so if you think from the customer perspective was the crackdown correct? Possibly yes. So yeah this crackdown happened though double opt-in etc. So there was a lot of friction introduced along the way to pick up a website.
00:24:25
Speaker
And this also coincides with smartphones and smartphone apps coming out. So then obviously, this industry wasn't. So we lost that monetization model. On the app, we had paid features. So if you want alternate routes, if you want traffic alerts from police, you would have to pay for that. But then Google started giving all of that up for free, right? So we had to make that free. And the third option we explored is why don't we work with car companies, OEMs, et cetera. We started doing some work with Map My India, which just recently, and I'm very good friends with them for the matter. And they were sourcing traffic information from us.
00:24:54
Speaker
But no car companies ever launched a connected car in India. So for traffic information, you need connected cars. Static cars work, static maps work on anything, right? So even till date, so we are very hopeful. We did some trials with some big car companies, BMWs, et cetera. But none of them ended up launching connected cars because they did not find the infrastructure for wireless communication, reliable infrastructure for wireless communication. So in the West, this all happens by FM manual. It doesn't happen on 4G or 3G, whatever it is. The FM manual for this was never allocated in India, for that matter.
00:25:23
Speaker
and that collected car dream also didn't plan out. So then we could not figure out a way to monetize this set, even if it scales, et cetera. And then you can't do an advertising play because it's a niche audience. There are between 15, 20 million households who use cars. And then with that, you're fighting a behemoth of Google, who has a distribution advantage of Android, supply advantage from all over. It's going to be very, very difficult. And that is because they always wanted to be a B2C company at that point of time.
00:25:45
Speaker
I had at that point reinvented myself and said, no, screw all of this. I want to be our B2B company. Today we would have been golden because everyone in the world, beta, Zepto, beta, just Viggy, beta, Delivery, beta, Flipkart, all want these maps and reliable
Rydler's Innovations in Public Transport
00:25:56
Speaker
information. We could have been that layer. So we had some initial conversation with some of the e-commerce company. But in my mind, I said, it was a lot in flexibility and what I want to do. The charm was that I want to be a B2C company and that job local comes. That was my wish list at that time. And I didn't pursue this opportunity.
00:26:11
Speaker
this what you're talking about the current opportunity this is what i think a company like next billion does right like
00:26:17
Speaker
Some of them are Xola and Gojek. So they're building all of it together. Because today, what is happening also when I was a CTO? There's no viable alternative for Maps. Map Mindy is good now. But a few years back, there was no alternative for Maps. And Google has a shared monopoly. If something were to happen to Google Maps, a lot of the businesses that we see today, the Swigis, the Ullas and all, will be suffering immensely. And so it makes sense. Next billion is doing that. Map Mindy is also making rapid strides under Rohan to actually get to that layer. So we need.
00:26:45
Speaker
As I said we were in early start if we had made some progress we had so much data from open street maps so we had data from traffic information we had a lot of that information going on a place and then obviously the likes of delivery and all that were coming up at that time.
00:26:57
Speaker
So if we had trained it around, right? And today we would have been a very valuable company. So today the monopoly would, they would have been an alternative for many other players in each other's monopoly. But we didn't. I think that was, it's important to stick to your goals, but I thought, no, we want to be a B2C. If you stick to outcomes, then that's what becomes like. What happens when you stick to outcomes, not to the opportunity. So we decided that we are able to drive usage, right? And using, building good apps, working with government, getting data, unorganized data, and converting it to usable information is a strength. Why don't we do the same for public transport?
00:27:23
Speaker
because when I was running the mobile app also, traveling mobile app, a lot of the queries that used to come for us go for public transport. When will my bus come? Especially when trains are late. When is the train coming?
00:27:37
Speaker
said that a public information is also broken. There are so many different agencies like Delhi Metro, the Delhi Transport Corporation in Delhi, in Bombay there's BSD, there's Western Village, Central Village, Mumbai Metro. All of them are operating in their own silos, similar to CS. No one talks to each other, but I as a user and dependent on their coordination to make sure my journey is a little more seamless and I don't have that friction. So why don't we bring all that information together and show it to the user so we take that
00:28:02
Speaker
siloed nature of the industry away from the user and they were unified due to them. So that's when we started pivoted to Ditra and became a public transformation plus a traffic app. That's what we thought. If you have to monetizing ads or any other way, then at least get to a base which is much bigger than what traffic would offer. So before that, it was not called the ride list.
00:28:19
Speaker
Now RIDLA was the new incarnation of the app. So the idea was that public transport is a RIDLA. So in India, in Bombay, so there is a BHT, there is Nui Mumbai buses, there is Viral Mumbai buses, there is Thani buses. Western Railways, Central Railways, there is Mumbai Metro, right?
00:28:34
Speaker
If it's a riddle, if I want to go from here to Washi, electric, three, four modes of transport, how do I solve? So I thought it's a riddle. And the one to solve this would be a riddler. And I think it was Batman in those days. Nothing with the Batman villain. But riddler was how that's the idea of a riddler came along. And we wanted, the learning was also keeping a traffic line makes you very gen, very specific to traffic and all that. Now we are going through much bigger traffic plus platforms. I wanted a more generic name for the platform rather than something which sticks to that particular initial domain that we targeted.
00:29:00
Speaker
I always thought there was something to do with ride. I always used to frontiers ride them. So the goal was, again, like a Google Maps competitor, except it also integrates traffic and public transport modes to help you map out how to reach from point A to point B within urban areas. You would only be focused on urban areas. There's no point in this beyond. No, we live in all cities, even in tier two cities, like Mangalore, where the buses depend on buses. In fact, in tier two cities,
00:29:28
Speaker
The only form of mass transit is buses. There's no trains, which are the fairly expensive to go from there. So it was actually a very popular product, not tier two cities as well. The problem in tier two cities was that the information was not organized structure. So the USB that we mastered at that point of time, working with agencies, working with the transport operators, agencies, et cetera, and converting that to structured information, converting that unstructured information to structured information.
00:29:53
Speaker
The way Google used to work in the West is they would standard data sources from where they would pay for this information. The agencies would do all that work for them. India, no one was going to do it for them except for one or two, like Bangalore or whatever. So that was the biggest strength that we used to, we had mastered the model of working with them and scaling this model for other cities, depending on the Southern city, Northern city, Western city, it didn't matter. We figured out a way for the ground team to figure out how to do that.
00:30:13
Speaker
So, say someone in Bangalore could map out a query that I want to go here and then you would get that person's location through their GPS and then you would tell them, okay, there is a bus leaving after 30 minutes from this stop, this is the nearest stop, the bus will... Precisely. Okay. Same thing which Google Maps does today. So, we have built the entire routing engine?
00:30:34
Speaker
I forgot the name, but it was a multimodal routing engine, right? So if you have to take two buses, three buses, if you take a bus and train combination, all of that was available on the app. Okay, okay. Though I think probably buses is something which Google Maps doesn't really cover well in India even today, like in terms of the scheduling, timing, that's very chaotic still. It's not a Google Map problem, as I said, it's a data problem, right? So the data is not structured. If I have to go, I'll tell you the example in Mumbai.
00:31:00
Speaker
If there are 27 depots, every depot manager has his own schedule. So there is a central copy of schedule, but then the real copy is the depot manager, and he's the one who manages it. So you have to figure out the right source of data and get those schedules from them. And he does it on the basis of ground feedback. There is too much traffic around the bus there, right? There's a conductor available here, do that. So he does it on the basis of ground feedback. So you have to fetch the right source of information. Google with its technology platform and all, not a ground team. It's difficult for them to call it such an unstructured environment. Did you start earning through ads then?
00:31:29
Speaker
No, we didn't. We didn't. So I think another interesting thing came up at that time. So it was when we started getting, so this was another app which hit 100,000 scale very quickly. We launched in cities, right? So the idea was either doing ads or doing something else. But the Paytm Uber success actually gives another direction, right? So Paytm and Uber, it has been a partnership. And if you remember, that was the first inflection point for Paytm and Paytm went to the next level.
00:31:51
Speaker
Yes, and PTM was the only way to pay digitally on Uber. No, there was no cash at that time. Uber was not, I think there was no cash. So PTM was the only way to pay because... Yeah, either PTM or credit card, right? No, credit card was also not the credit card. They blocked him. They needed two-factor authentication, right? And the only way to deduct was PTM.
00:32:10
Speaker
because of monopoly of sorts. So we said that you can book a red bus, you can book a long distance ticket on buses, on mobile or on website. You can book airline tickets. You can now book a cab. Why is it that public transport which we are catering to is not available on a mobile phone? The problems are equal, right? I have to stand in line all the time while buying a pass. I have to haggle with the conductor for change. I have to waste so much time. Why shouldn't this be available on mobile? And then you see it's a difficult problem.
00:32:34
Speaker
Because you can't just, it's not an API call where you can say, okay, I've reserved a seat. There's no reservation. It's not a reserved category. So if I have a printout, I could take 10, 0 copies of the printout and I'll give it to 10 of my friends and the copy can be, and if I'm going to traveling a Metro, how do you verify this? Because no one, if you scan it, the gate should know, right? This is a repeat copy, whatever. And there was no, no one was doing this anywhere in the world. And then again, my tech part of me, the hardware part of me got curious that actually I can solve this problem. And we were the first ones at Riddler to introduce a mobile ticketing for the unresolved category for buses and Metro.
00:33:04
Speaker
And that became a very popular product that started having GMB, we started having transactions with the platform, GMB through the platform. But tell me how you figured this out. Like Metro, I can understand maybe at the point of entry, the person could use the Riddler app to scan or something. Was that the way? Yeah, but how do you know that the scan has been one way, right? If the scan has not repeated, how do you know that the scan should repeat? So that's the difficult part. So very interesting solution and something I'm very proud of.
00:33:29
Speaker
We put Wi-Fi chips inside buses at the entry of the bus and exit of the buses. Wi-Fi chips have become very cheap, right? There's like 250 rupees at the bus. So even if you have to put in 250,000 bus, 4,000 buses, it's not very expensive. So what happens is when you enter the bus, your mobile app, you turn on the Wi-Fi on your mobile app, right? And the mobile app and the Wi-Fi, the Wi-Fi chips talk to each other, it activates the ticket.
00:33:51
Speaker
And when you exit the bus, it deactivates the ticket. So the life cycle of the ticket, which is during the beginning of the bus, at the end of the bus is the only life cycle the ticket is active. Then the ticket disappears from the phone. So that way we ensure that there is no fraud. And the ticket can be identified as an identifier. For example, the Wi-Fi which we put inside the buses had an identifier which was correlated to the bus number of that. So if I'm buying a bus number 66, the two have to marry and talk to each other before their ticket can be validated.
00:34:15
Speaker
When the conductor comes, you have to show a validated ticket. The conductor punches the details of the ticket on his machine and the reconciliation happens.
00:34:22
Speaker
This ticket purchase was made seamless and the reconciliation was made seamless for the agency. It's not that you have to go to a separate table to look up the reconciliation information. One question on the bus thing first before we come to Metro. You are expected to buy the ticket before boarding. It's not like you can board and it'll deduct from your wallet automatically, not like that. No, you buy a ticket. So mostly the use case for us was passes. 20-30% of the users were not passes. So you pass, pay for that pass once in a month, once in a week, etc. Or you pay for the beginning of the day. And then you just validate your pass. You marry the two informations.
00:34:51
Speaker
and then you can also do tickets but ticket is cumbersome because every 10 rupee transaction you have to go to PTM, add money and do that. But the biggest use case for us is passes. So we want the reason people switch from tickets to passes because pass buying in buses is a very tedious experience and really why? Yeah, you have to go to some office and... You have to go to Depo, right? You can board the bus from 6,000 different locations but you can buy a pass only from 25, 30, 50 locations, right?
00:35:15
Speaker
the disparity there. Whereas in the case of a train or a metro, the place where you buy the ticket from, the place where you board from and the place where you buy the bus and pass-through is all the same.
00:35:27
Speaker
If you like to hear stories of founders then we have tons of great stories from entrepreneurs who have built billion dollar businesses. Just search for the founder thesis podcast on any audio streaming app like Spotify, Ghana, Apple Podcasts and subscribe to the show.
00:35:48
Speaker
So tell me the Metro solution, what did you do for Metro? In Metro also, we put a QR code reader instead the gate. So we cut open the gate that the turnstell that you see, right? And then obviously there was a validation circuit. Now, how does one gate know that all the other, this gentleman has already passed me or this gentleman is supposed to exit. So that's where we have a mesh network that we created every Metro. The Metros typically have a fiber network.
00:36:07
Speaker
which is used for communication of the signaling information. We piggyback on top of the fiber network. So if one turnstile sees that Akshay has entered at station A, he informs all the other turnstiles which are relevant that Akshay will be exiting from this particular turnstile.
00:36:21
Speaker
If Akshay exists from one of the samsar, he's validated and he's allowed to go through. If you've broken a rule, if you've jumped a gate and you've not entered properly, then he's not allowed to exit all. And once Akshay has entered, no other gate will be allowed entry of Akshay also. So we managed. So it was the first actually blockchain implementation. Everyone was a ledger. Everything you turn still was a ledger.
00:36:38
Speaker
And that ledger had information on all the passengers which had gone in and come out, etc. It worked beautifully. It today works. And then China did it. Now a lot of countries in the world are doing it. But very proud. So today if you see Delhi Metro, Mumbai Metro, a lot of the metros now have that mobile ticketing solution. It's a win-win for both, for customers and for the agency. For the customers, you don't have to stand in line. You get everything on the phone.
00:36:59
Speaker
For the agency, you take away the cost of that token, the token which gets lost, which gets damaged, et cetera, et cetera. And you avoid lines at the metro station. You can use the real estate to advertise more. So it was one of the solutions which obviously benefited a much bigger part of the ecosystem. And this was what? Again, you buy a ticket in advance for past users or for debit from wallet. So you can buy a ticket. Basically, you have to buy the ticket before you reach a turnstile. People would buy it on the way to the metro station, ride the metro station if it's not underground.
00:37:28
Speaker
You have to do that and you buy your ticket, you open your phone and just scan on top of the gate. And the gate opens if the ticket is valid. Here there was no pass holder concept, right? I don't think. No, there were passes then. So you could buy a pass where you have to make funds. And every time you did a transaction, the value from the pass would get detected. So this led to what kind of revenues then? We got to a 4 to 5 million dollar revenue from a zero revenue company. More importantly, we had close to 100 million GMB flowing through the platform. And we had to give money in two ways.
00:37:56
Speaker
One is on a per transaction basis, that every time a ticket used to be bought on the platform, we used to make one or two percent. And the initial installation fees, because the hardware, in public transport agencies, especially when it's government funded, there's a big question on who owns the platform, who owns the hardware. And they always wanted to own the hardware, because that's how there's lack of, that's how they are far more accountable. So we used to charge them for the initial deployment. And then as an operational cost, we used to charge them on a per ticket basis.
00:38:21
Speaker
So we did this for BST, we did that for Navi Mumbai, we did this for buses in Burgham, all of Punjab State, Delhi Metro, Bombay Metro. A lot of the prominent agencies in India were starting to work with us. So we started with mobile ticketing as a way, we caught into the foot in the door. And then obviously after the whole acquisition we did a lot more, we started becoming a full stack technology company rather than just mobile ticketing company. Why did you get acquired?
Riddler's Acquisition by Ola
00:38:47
Speaker
As I told you, my passion, my reason to start this was not business race. It was a personal pain point where I started. This was a cool system to build and I started building without. There was no thought process like we have today, right? Where will the revenue model come from? Where will the models come from? What is the future? We were always talking about the mindset which I started with. Let's start something and we'll figure out along the way.
00:39:07
Speaker
But that is not how it works in the world today. That's what I've realized. You should be clear on why you want to go to it, otherwise it becomes a struggle all throughout. More importantly, I realized that given the margins that we are running, it's never going to be a big business by itself. It will be 100, 200 crore business at best. Because even if you do the best technology platform, and if everyone is using your app, on a 10 rupee ticket, you can only make 20 pesos. You cannot make more than that.
00:39:29
Speaker
And because of government restrictions, political pressure, the socialist nature of some states, the 10 rupees will never become 100 rupees. Ideally, economies of demand, if you input the service and all that. The other and the major reason I saw that, there was no inclination from any of the government authorities, anyone to actually promote public transport. That doesn't happen. The public transport is going to die.
00:39:48
Speaker
The only places in the world where public transport companies have done well exceeded is where there are restrictions of car usage. Because Singapore is a company which comes to mind. In London, you cannot take. If there is a very high tax on you going to the central part of London, that is what will encourage people to use public transport. If you don't do that, public transport by itself will never make money. And if it doesn't make money, then the losses will essentially keep on flowering and you stop putting money back in there, which results in a very negative viral loop. And I saw that most parts of India were actually following that, except for Delhi metro.
00:40:14
Speaker
which is a price crown. No other public transport agency in the world was being treated in India was actually being treated fairly. In fact, if there is a riot, the first thing you break is a bus and we are very happy with that. So if the industry has to grow and then you have to invest into it, you have to be ready to do it. At the end of the day, bus, auto, taxi, all are competing for space. It's a space problem. It's not a mobility problem.
00:40:33
Speaker
If you don't give them roads to travel, then they will not be able to travel. They will not be able to go to bus. I've gone to buses. I've gone to outside metro stations after railway station. The shared auto rickshaw drivers block the bus entry. They don't allow the bus to go through. So that people switch to shared auto. And that's your local union or the local Kunda supporting them. And there's nothing the bus agency can do, right? We've been talking about dedicated lanes for buses for the longest possible time.
00:40:53
Speaker
Delhi tried it also, which was hated by everybody. The car lobby, right? If the state government relies so much on the petroleum taxation which comes from petrol, it's impossible that they would promote non-vehicular traffic. Are we possibly going to see a future in the next 10 years where public transport is going to proliferate it? And also as a user, I saw that users, as soon as I had access to data, I used to look at cohorts of customers, LTV. Most of my users used to churn after a few months. There's nothing I did. But when I used to call them, I said, I don't even take the bus anymore. Dude, I don't have a problem with your app.
00:41:22
Speaker
It's not that the app is bad. It's just that I don't think I want enough money now to afford a rickshaw or a bike. I have switched to another mode of transport. So it was a mode of transport which was used by people coming from a village who did not have enough means at that point of time or was an out of a job. Figuring out his life, as soon as they figured out something about life, they moved to another mode of transport. So there's no way for me to monetize the customer. Then he wouldn't care about Riddler or anything of that sort. I had the fortune of looking at this objectively, the deal. Because when you do a deal, it's all on the founder and I was the decision. My investors were very kind enough to let me make the decision. But I had four offers at that time.
00:41:51
Speaker
Two for acquisition and two for funding and the investment for big companies also. So, it was not that I was forced to make that acquisition, I was forced to sell out the company. I could do it on almost the terms that I wanted and I could also weigh the pros and cons of this with a future in mind that you've already done it for six, seven years and you want to keep doing it because you've done it or you want to actually think you have to eventually go and execute investors, you have to make sure employees are liquidated. How much have you raised by then at this time? We have raised around nine, eight to nine million dollars, something around that.
00:42:20
Speaker
Like series A, series B, I guess. We linked up to series B. And then when I was out in the market to look for series C is when the acquisition offer came around. So the idea was not to get acquired. The idea was to look for series C funding. So the idea was that this was the right time. And I didn't exit. I just said that this is an existing investor. I continued building Riddler for the next three years along with Ora. And Riddler did grow during that time as well. So the idea was not to look at it in an exit, but the idea was to make sure that it makes starting businesses.
00:42:45
Speaker
So what was Ola's proposal? Where would Riddler reside within Ola? So Ola was very kind enough to always let Riddler operate independently. Obviously the business had to start being sensitive. You could not do projects to satisfy or just get numbers for the heck of getting numbers. So that was the only alignment which Ola got us done. So V for the first.
00:43:02
Speaker
For the most part of Riddler's existence, Riddler did continue to operate out of Bombay. While Ola was in base in Bangalore, there was never any push. The only thing we had to go for Bangalore was budget approval. So if we thought there was a project which was viable, we should do it. What Ola made possible for us was, first of all, we got a strong balance sheet. So now we could start bidding for bigger projects. Typically projects in buses were like you have to invest in the Capex and you require the money over five years. We at the startup never had that kind of money.
00:43:27
Speaker
But Ola with its balance sheet and BankConix was able to help us get that kind of network. And the other one, Ola had a sales team across the country. So we had to talk to bus operator in Punjab who was not reachable. Ola sales team was essentially it. So we used those two things of Ola and obviously learning the business part of it. So the proposals used to, we used to wait it. So we used to obviously say, this is the amount of spend.
00:43:46
Speaker
and this is what we make at the end of it. They were financing targets and we tried to beat both of them, but it was very grateful of them not to interfere in the 8-way operation. I could not continue in Ola, Redla for a long time because I was asked to take a bigger role at Ola, but Redla continued to operate independently during that time.
00:44:00
Speaker
What projects did it do? It continued to do the same thing
Expansion and Challenges Post-Acquisition
00:44:03
Speaker
only? No, we actually became a full-stack ticketing company. So we started as a mobile ticketing company and when we got acquired by Voila, we were only doing mobile ticketing for transport agencies. But then we said, why not do the entire thing? So even paper ticketing, card ticketing, credit card ticketing, so the entire ticketing stack. So first you are integrating on top of a system integrated and just providing the mobile layer. Then we went deep in the stack.
00:44:22
Speaker
and started doing that. So, this would be a SaaS or this would be a like what was the model here like a one-time charge and then percentage of sale price or what. Yeah, so it depends on agencies. So, in the case of metros which are treated enterprises daily metros they had a lot of money.
00:44:38
Speaker
So that would be a Capex plus OpEx, right? They had all the money to give. Governments always funded their metros. They always had extra money. They needed. So they would, and they had the swagger as well to buy out a project. We won't work with anyone. We wanted hardware, right? So they would buy out a project and then give us Capex plus OpEx. Buses on the government side wanted things, but they did not have the money to invest in Capex. But they are recurring operational revenue. So then you invest operationally. You invest one time, and then you recover the money over time. And the third was the assembly, which I call private operators in the country. There's a lot of, most of the states, many states in India don't have
00:45:06
Speaker
public transport or government transport. So their private operators run. So they are like a fleet of seven, eight buses, five, seven buses. So they were like SMB for us. So we would give the hardware and software to them. No Capex, only OpEx, but we ensure that their service tend to end. What would happen typically before us is people would go sell hardware to them and disappear and their Capex would just be sitting out. So you'd say that you don't have to worry about anything, all the software and all the hardware, it will all decide on the cloud. We'll have service engineers who will come and service if your hardware goes back, but then you do it. So it was three different models, depending on the nature of the industry.
00:45:32
Speaker
The last time I was in a bus was probably 2003, 2004, when ticketing was the conductor tearing out a piece of paper. How does digital ticketing happen in buses now? This I'm talking of the local buses, not intercity.
00:45:48
Speaker
Your intercity undeserved tickets thing happened that way. So they have handheld devices now. They have handheld devices, which are mini computer of sorts. And where there is preferred information about the routes, fares, et cetera, et cetera. The first generation of handheld devices came offline. So you would not. So basically when a conductor wants to issue a ticket, you enter the source destination. They have been taught the interface and enter the numbers and the ticket get printed. The fare is mentioned. You take the ticket and print it.
00:46:12
Speaker
And then at the end of the day, you go to the, therefore you plug in the device into the machine or laptop and that information gets downloaded. And that's how the reconciliation happens. Then the next generation, which came with started supporting online ticketing as well, because there was a big push to use debit card, which I'm using anywhere to start using on the buses. So we were the first ones to introduce debit card vestigating in Delhi Metro. So you could use a debit card offline as well as online. So then they become online terminals.
00:46:34
Speaker
where the SIM card obviously secures SIM card with PCI DSS certification. And then when you tap on it, money gets deducted and the data gets caused to bank as well as the fare collection server as well. So they still have handheld devices. The device, the handheld device. So you were providing that also.
00:46:49
Speaker
Yeah, we are drawing it. So you are working with the local manufacturer. So earlier when there was no credit card ticketing, there was a custom-made device. So we made a custom-made device which had high battery life, very high battery life, easy to operate, big buttons so that the person could bend, enter, etc. The place where you tap the card is very, we had a very big battery rate, 3600 mA, so it would last like 13, 16, 18 hours without the conduct timing to worry about discharge. Very thick paper rolls, we don't have to keep on replacing that. So that was the custom device that we made.
00:47:14
Speaker
Then the government push came for this whole NCMC cartridge, the National Card Mobility cartridge, one card for transport modes in the world. Then we switched to a very foreign standard device because those devices then needed certification compliance because they were handling magnet, correct? So then we stopped doing the hardware ourselves and we started buying all the stuff and writing our software on top of it, right? Which was, and then the backend software immediately. Some cities, most so, even today, 90% of the cities would use that custom hardware that I'm talking about. 10% of the cities are now using that very phone or that fancy device. And there we have to do a lot of work on the software.
00:47:43
Speaker
Because the device is smaller, the battery is smaller. So how do you optimize software that especially in the power consumption side, so that your battery doesn't run up. If the problems still remain the same, you have to do an eight hour shift and the battery should not drain. The printer is very small. So we had to fight a much more software challenge on those devices than a hardware challenge. Is this a reality, this MCMC, this National Mobility Card?
00:48:04
Speaker
Not, it's not the reality of the scale, but it works. So if you go to Delhi Metro Airport line today, you can tap it and tap auditing your NCMC debit card and you can use it to direct money to everything from the bank. You won't have to stand in line. Oh wow. And what I say in all issue this card, all the banks issue. All the banks. So we, yeah, that was the first. So Redla was the one who implemented this in Delhi Metro and the Prime Minister Dr. Mr. Nambudhi implemented, inaugurated. This was December 2020. Yeah. At that time. So this was the first.
00:48:31
Speaker
Okay, and this is that card which has that touchful... Yeah, they call it Wi-Fi, but it's a chip which is there, right? It's a RUPE card, it only works with RUPE. Like NFC, probably NFC technology. It has an NFC chip. So it's an NFC chip, it's based on NFC only. And then there is security, which is security data, which is provided in PCA. And so you need a certain level of security, the device has to be certified.
00:48:53
Speaker
and then the key exchange happens between the car and the device and that's how the authentication happens. But you see this becoming like you have a debit card which you can just tap on because transport is a state subject so each state will... No, I think it's not that. You are not allowed to auto debit from the bank account. So what happens is if you want to do an online transaction, again the problem of delay.
00:49:14
Speaker
So today it works as offline. It's like a pass. You go to a ticket counter, give that a 100 rupees, 200 rupees. It tops up your card for 100, 200 rupees. Essentially, it tells that. It's not linked to a bank account. Because if you link to a bank account, and if you tap at the entry gate, then the amount of time it takes to go to the bank server, find out if you have enough balance and come back, you have to pass one user in one second. In a bus, imagine it uses a local storage of the machine. That means you have to keep on topping it up all the time. Now, the other way is that if you don't have, then you have to base it on trust.
00:49:44
Speaker
And this is where India's stuff, we don't have trust. So, okay, you can say, okay, I'll go to a bank and get it. But if your bank does not have a blacklist, but there's a one right risk, there's one right risk, because I can do one right without, I can be, no one in the country was willing to take a one right risk, neither the transport agency nor the government.
00:50:00
Speaker
nor the banks. But then who takes it one day? If no one takes it one day, then you need a full-proof system. If you need a full-proof system, you compromise on user experience. If you compromise on user experience, you will not be able to use it. But this would be pretty revolutionary if you have a debit card that you can just tap and write anywhere in the country.
00:50:15
Speaker
It works in London. The transport for London takes the risk of the first ride. And obviously people are more trustworthy there. So I think it means an underlying problem of solving the underlying problem of trust. If you solve trust, then the user experience, as is the revolutionary. Why can't it happen? It should happen. But this is where the problems I see. And what happens to the transport agency? He says, why should I take the risk? It's a bank who is benefiting by distributing the cards.
00:50:40
Speaker
The bank says that it's not my job to implement security, it's your job to implement security. So there's a pull and push and no one's taking ownership. So who's taking ownership of deploying these cards? Transport agencies have been mandated that you do this card, but they don't have any pull. They think that if I do this, the banks benefit. I don't even get the benefit out of this. So the model for this has not been established very clearly. And if you establish model where both parties benefit because of this, then it makes sense. Today it's not, it's a push and pull. So one guy's pushing, the banks are pushing because they have the cloud, they have the support from the government. And these guys are coming up with all possible excuses to possibly not do it as well.
00:51:10
Speaker
So banks are pushing because they'll earn some sort of intercharge, interchange rate. If the credit card is distributed, credit card is distributed, it's good money for them. It's a low, very low. So they know transport is a captive audience. And if they do that, then they get a very big base. Then that guy uses credit card as a place. So in Hong Kong, this is how it works. Oyster is, I think, the car. Octopus is in London. Octopus is in Hong Kong. So you can use it. So transit is the way to acquire users. But then the same Octopus card, you can go and tap on other stores.
00:51:39
Speaker
and then they make the MDR from there. Now that our bank agency are not willing to share the MDR with the transport agencies, right? So then how would they run? There's no participation from them. Then why would I even distribute this card? But this Octopus card is a prepaid instrument or is linked to a bank? It was a prepaid instrument. I don't know if obviously I've not been following this for the last few years, but it was a prepaid instrument back in the days when they started out. But now I'm sure they must have linked it to the bank here. Okay, so you have to load it with money. Japan has something similar. You load it with money, then you can use it at vending machines and
00:52:11
Speaker
So you will enable the loading infrastructure if you get money. You can't get money from transport alone. It's a very low ticket. If you can make money from commerce, and then banks should be willing to share that with their transport agency, then the transport agency will be very happy. That's not happening. So what happens is both try to squeeze others to get the best bargain, and then no one is looking at user experience of promoting this product in a big way. So I don't think it's a technology problem. It's more of what is the right business model for this. And until that is figured, I don't think this technology scales right.
00:52:35
Speaker
Yeah, that's an interesting insight that transport has an enabler of commerce, which is what I think IRCTC also did for online commerce. I usually meant that I was part of Redla as part of Holayan. We used to take people to work, we used to take people to restaurants, we used to take people to shopping malls, but we never made money from Zofres. If I was a website traffic, if I was a website referral company,
00:52:56
Speaker
But I'm referring people to Flipkart, I'm referring people to Swiggy. They always give me 6-7% commission. But I'm taking trans people to buses, places, etc. I never get the money, right? So think of what happened in Gurgaon, right? DLF Metro was the one who actually, DLF was the one who built out the Metro.
00:53:12
Speaker
When you build out the metro, the price of the property goes up, you make money out of that and then you abandon the metro for some sort. So it benefits a lot of agencies. Even if you see more railways extended to Iran, the price of properties has just gone up. But what is the benefit to railways? Nothing. That's how London has actually made transport sustainable. They get planned usage fees. So if they build a metro somewhere, they get money from a billion dollars, something like the royalty, for actually increasing the prices, increasing commerce in workplaces.
00:53:38
Speaker
So when I'm doing that's the only way to make support in Prasad. None of these things are happening in India. Either you put restrictions and promote people to use it or you give them some incentive. If the things are not happening in India, it's very difficult for transport to sustain by itself. It's a theory as to why I decided to pursue the whole opportunity. Because I don't see these things happening in the near term.
00:53:55
Speaker
So you joined like from Riddler you got like an internal movement to Ola as the CTO. You call it an internal movement or you call it there was a need for me to come in and then I was it was thankful to the management controller they thought I was worthy of picking up the CTO hold and yeah I decided to take that out so I became the overnight I became from a CEO would not
00:54:16
Speaker
who had a CTO, so become a CTO of a big company. Yeah, and what was your role there? Yeah, I started off as a CTO, which obviously was to streamline tech delivery, align it more with business, but very soon I became a CPTO, which is Chief Product and Tech Officer, and six months down my regime, I also become... I was very... Amongst the whole entire whole mobile tech stack, I was most fascinated by the allocation piece and the pricing piece, which is what the heart of the system is, which category, which person, and what should be the pricing. I did most of my work there.
00:54:45
Speaker
And because I had drop insides, also became the revenue officer, chief revenue, head of revenue, right? Deciding how much incentive to give out, who, which users to get discount, which ones to not get discount, what should be the pricing for particular routes. So it decided the P&L. So in nine months, I became head of tech product and also revenue during that time. So this was essentially like a big data problem, allocation, pricing.
00:55:05
Speaker
No, it's such a behavioral problem more than anything. Behavioral and human-insized problem. You want to understand why the driver would accept or reject a ride. And what would trigger him to accept or reject a ride. Why would a customer choose your platform over the competitor platform because at the end of the day they are the same one. We have to understand this and obviously we had data science teams to actually solve the big data. So the good part about me for Ola was I could actually solve the business problem because the infrastructure work was done right. So we had a very good tech platform which is one of the best in the country. And a lot of work was done on that. So my job was not actually building that up.
00:55:34
Speaker
My job was to utilize that to maximize growth, maximize revenue, maximize margins. I was fortunate that I didn't have to build out many things from scratch, except for maybe the fraud side. Because fraud was something we were just picking up and we didn't have things in place, so we had to build those to ground up. But most of the other things were there for me. Fraud by drivers or by passengers? Everyone, driver, customers, network, everything is fraud. Because we didn't have much there, so we had to build that up, ground up. But other than that, most of the things were ready. It was a very old company.
Impact of COVID-19 on Ola and Transition
00:56:02
Speaker
Obviously, it was a seven-year-old, eight-year-old company by then.
00:56:04
Speaker
and they had built a lot of tech capabilities. So my job was to switch all of it together. So the problem was they built a lot of stuff. It was all joining it together and presenting a unified view was where my strength led and that is what the opportunity for me was. And were you there when the plan for IPO and all that happened? Yeah, that's been happening. I was there and I joined Ola when the international expansion was big. But I also there when COVID happened, which impacted the business in a big way. So we launched London
00:56:30
Speaker
where we had one of the most amazing launches. We took away 15-20% market share in a matter of few days, weeks. And then we're doing really well. We're doing well in India as well. We're growing. And then COVID happened, right? Which put brakes on the business to some extent. And obviously the IPO will get impacted. So the IPO has been in the works. I was not closely involved in that. But my unfortunate part is that my tenure also coincided with COVID where the business was one of the most impacted businesses, right? What we did was people who asked to stay home, not move.
00:56:56
Speaker
So everything I knew in the last 10 years about helping people move down the drain because people ask not to move anymore. So it was a very difficult one of the most difficult times of my career where we saw all the graphs from millions of writes, so many writes going down to zero, right? All of them actually ended up going to zero. So yeah.
00:57:11
Speaker
It was and then reviving that business after Covid was one of my biggest responsibilities and one of the most exciting things I've done. What was the reviving would happen automatically, right? Like once the lockdown is lifted? Not necessarily, right? So drivers have gone back to villages, drivers are fearful of taking passengers along, passengers are fearful of riding in the cabs. We don't know it at that time. We didn't even know about surface transmission.
00:57:33
Speaker
We didn't know, so working, giving them sanitization after every trip, trying out for sanitization, getting there. The customer base had completely changed, right? The guys, the IT workers who were primary base, like the ones who would go to offices were not going to offices. But the ones who we know, the new base was the ones who were going to buses, taking buses and rickshaws and public transport was deemed unsafe at that time, but the ones who were out stationed, it was only focused on cities, how we take care of this. The dynamics of the business has changed fundamentally.
00:58:02
Speaker
The good thing which worked enough here was that there was less traffic on the road. If there was less traffic on the road, then you can do more trips and then you can do more, whatever. But also the same thing, the gas prices went from, petrol prices went from 75 to 90, 110, right? So that was another thing we did. So a lot of things changed, balancing all of that. And then obviously the threat of Uber always looking around as to what they are up to. So considering that, no. And then every city opens in a very sporadic manner with some restrictions. So how do you comply with the restrictions the same term, not jeopardize your head? So there's a hot zone.
00:58:32
Speaker
You want to go to the driver there, you want to pick up, so you find out, okay, you don't go inside here, but if your passenger wants to go there, you pick up at this point. Then cashless became prominent because we were even exchange, we were worried about exchange of cash at that point of time. Good chunk of our business was cash businesses.
00:58:46
Speaker
Auto rickshaws are deemed safe. So we figured out an entire hotel, we figured out an entire onboarding of auto rickshaws, where auto rickshaws were drivers who could onboard themselves without having to visit Nola Centre. Because auto rickshaws were considered open, right? Again, no one knew much about COVID after the first lockdown. It was all, there was so much myth. When food delivery was considered unsafe, things obviously got back to normal once bits started going. But when we, so we had put spurious inside cabs, which automatically sprayed, disinfect the cab, the drivers were cleaning them.
00:59:10
Speaker
There were shields across. So there were a lot of things which were done. Instructions, authentication of drivers wearing a mask or not. A lot of these things had to be done to actually put... People didn't trust shared mobility at that time either. And people didn't need to travel. It was discretionary. It was not mandatory. It was not important that I am sorry. And so how do you get back people to camp? Even drivers had gone back to villages. Drivers were afraid. So whenever there's a hotspot in any area, the drivers would be scared, they should go here, they should not go here. So all of these things, we had to fight all of these battles. The only thing I did in my favor at that time was that the traffic on the roads were low.
00:59:39
Speaker
which made the experience far more easy. Sir, you would have probably had to change the app also to emphasize on hygiene a lot more and build that customer trust around hygiene. Yeah, changing the app around hotspots also. If there's a hotspot, then if the cab request comes from there, how to handle that part of it. Then hygiene was more of an information piece, but essentially communication on the hotspot went to go and into driver cancelling because of a perceived threat of hotspot.
01:00:03
Speaker
And there were all these rumors flying around because the drivers would not show up on a particular day, then communicating that offline, then it's safe, this is how it happened. So a lot of work around the app, also around the app, you have to obviously show the mask bit, compliance bit, etc. etc. We had started up a sanitization driver, so we tied up with one of the petrol companies. And the cabs were expected to go there every one day, two days, etc. to make sure that
01:00:34
Speaker
I have this very outsider perspective on Ola that it does a lot of experiments which don't really scale like I don't know if the fintech business of Ola or food delivery or now I was also reading about fast grocery that Zepto and Blinkit like going up against
01:00:55
Speaker
Do you think that is the case? That they are not focused on just the core business, they get into too many other lines? I think it's on the contrary, rather. The core business in India, if you look at what is the core business, it's the core business mobility in India. They are the closer number one player for the most time.
01:01:16
Speaker
It's a profitable business, it's unit-positive, it's avatar-positive, it's bought into a state. Now, it might not be growing due to severe reasons. As I said, the industry is not growing by itself, it's not a rule, it's not growing. The number of people travelling, the roads are not growing.
01:01:30
Speaker
you have to experiment and get to new lines of business, which is using your core capabilities. Now that's a strategy call whether you do delivery or you do financing, that's a strategy call. But I personally believe one has to do that, right? If you don't do that, then you will stall, right? And whatever I've read about economy, whatever I've read about money, it's always the one or two experiments you do which stick. Seven or eight will potentially fail.
01:01:49
Speaker
We obviously don't hear about the ones which fail a lot, but we hear about the ones which succeed. The difference between Ola, what Ola does differently, they don't throw money at the person. So I don't know if you consider the other names that you mentioned as successful because none of them have shown profits or sustainability yet. They're all in cash burn.
01:02:08
Speaker
Yeah. So then if you consider that success, then you have, we have not succeeded in burning more cash, right? You can see it, but once they become successful, you know, it's not same with fintech, right? What do you say? So fintech business from what I understand is a very good business, but other fintech company have, they really make money. I'm not sure about that yet. So I don't know the definition of success, but if you want to, if the success, the finished success, keep throwing money up, like
01:02:28
Speaker
continuously money. The point we saw at Ola was that when the thing was riddler was right so we could have done mobile ticketing in one all over India we could have but then how much money you would have made if I could get to move the needle for Ola.
01:02:40
Speaker
So the idea was that you build business in a sustainable in a manner and there are opportunities. It's not like all the problems in the world are solved. Tola did credit cards on the first ones, on the finance studies. So we started the BNPL. So what do you call BNPL rage today? Ola money postpaid. Those are things which happened much earlier than other places. So when I decided obviously it was in 2020 that I decided I'd move on. I moved out in 2021. At that time, it was a very uncertain world. Today you're seeing certainty, but there is so much uncertainty now with the other parameters. The war is going on. Then there is the Fed hike, the liquidity crunch.
01:03:08
Speaker
So I didn't want to go through that uncertainty period, especially they said if things were to not work out in mobility, not for the reasons anything would happen, but just the macro environment changes and where would I go after the age of 44, 45. I don't know the motivation to start again would be still there and go through a 10-year grind. I thought this was my time and I found a good problem statement to solve and might as well jump in. So how did you discover the problem?
Identifying New Opportunities in Healthcare
01:03:30
Speaker
Yeah, so as I told you that I really wanted to be a cricketer.
01:03:33
Speaker
But that doesn't mean I gave up on sports or cricket. I played my fair share of cricket and sports. Growing up, I represented my school, university. Even in the US, when I lived for six years, I took to cricket very well. I used to represent my city of San Diego.
01:03:49
Speaker
And I'm still a very big Lakers fan and San Diego Chargers fan. I still follow it even 12 years after moving out of the US. So being a sportsman and obviously being an outdoors guy, I always had my fair share of nickels. I've always injured some part of me. In the US, we used to play in the summers, so we had a long season. If I used to injure myself, I would just video call my sister, who's one of the best sports physicians in India, that is the problem I'm having. She used to ask me to touch some body part, risk at the start, etc.
01:04:14
Speaker
And I would do those things, icing, few exercises, and next week I would be ready to roll again. So I always wondered why is this kind of care not available to everyone. But that was the starting. But that was always the back of my mind. I never really pursued that. It was solving my problem. At the beginning of the pandemic, obviously we were all working home, so I got in touch with my sister. She was struggling a lot to manage her practice.
01:04:34
Speaker
Like it was a big transition from the offline world to the online world. Seeing patients on Zoom was, she used to run a clinic and people would come, she would make them exercise, get themselves assessed, touch them, understand what is going wrong. The paradigm is changing and it became very stressful for her. Not only the treatment part of it, but even the management part of it. So getting on Zooms, taking appointments of practice, getting on a Zoom call, taking notes when doing the Zoom call, then obviously entering them in an Excel, then sending them exercises on YouTube, getting information of their WhatsApp.
01:05:02
Speaker
patients are complaining that they are not able to do the exercise or do another zoom call. So you imagine it's Harakari, right? It's horrible. And that's when I gave her a brilliant idea. So at Riddler, we had done a very interesting project on passenger counting. Operators typically had no trust in their conductor that they were collecting the right fare.
01:05:18
Speaker
So using the cameras, you could count the number of passengers going in and out to 95% accuracy. I told you that I'll build a computer vision model for you. You upload accesses on that one. Content will automatically go to your user. Say when the accesses, they will do this in front of the laptop. And then you will get a report, right? You don't have to get 100 calls with them to ask them what they're doing. And if the report is not in line with what they're saying, then you get a call with them. So it just reduces your workload.
01:05:39
Speaker
But the report will show what? Whether they're doing the exercise right or in how much they're doing it. So any physiotherapy exercise, any exercise which involves movement of science has two things. Quantum, that is the repetitions, the number of sets, repetitions, etc. In the form of, whether it may form angles, correct alignment of the body. If you don't do it correctly, you normally risk a chance of not recovering, but you also risk a chance of injuring yourself.
01:06:02
Speaker
So the form is very crucial. So people are not recovering. They don't realize they're not doing the form correctly. But she can monitor all of it remotely. Because she cannot go to everyone's home. Everyone now cannot come to clinic to exercise with her. But is computer vision advanced enough to judge form? Computer vision cannot judge form accurately. It can judge form accurately in many cases. But you have to do two things. There are two parts of the problem. One is you detect the body movements. I detect where my hand is at every point of time, in every frame.
01:06:29
Speaker
and then I can integrate that to understand the form I can use that data to interpret the form I don't need the computer vision part to interpret the form for me
01:06:35
Speaker
Formative interpretation is subjective, but you can treat the machine what the right form is. But you can have logic. Once you have the data, and this is how my hand is going, set my hand. So if I'm moving my hand this way, I know it's moving very slow. If I'm moving this way, I know very fast. If I track my hand and see how long it will take me to make that wave, I can actually see that it's moving fast enough. So you can write data sciences algorithms on top of the body part detection to actually figure out whether it's working. So key body part detection is very accurate. So what happens, machine learning, my view of machine learning is that to do generic stuff, it's very accurate.
01:07:03
Speaker
you train the model to a generic stuff. When you ask them to do special cases, when you ask them to handle negative cases, that's when the machine learning model falls. Take them with autonomous cars. What is a negative case? I'll tell you. Autonomous cars work very well in a closed community group where the number of outlier events can be very minimal.
01:07:18
Speaker
So you can train the model to handle the outlier event. Can a truck buy into your road? No. Can someone come the opposite way? No. When you take the same autonomous car on the main roads and the highways, it takes 15, 20 years because you can't train the model for every scenario. It's a learning, it's not a discovery model. So can you train the model that a car can actually buy into your lane? You have to actually encounter, you have to find a lot of data and train that model. So many things should possibly go wrong when the human brain is faster to react and do that.
01:07:41
Speaker
But in a community where it's a, so when I say it's negative cases mean something that will go wrong. You train the model. And if this situation happens, you have to present the context of the machine. And in this context user, that's how you turn in the, unless you know all the context in the world, you cannot train the machine. Even if you know all the context in the world, how do you get all the data in the machine? So in the case of computer vision,
01:07:59
Speaker
For me, will my shoulders and my nose be different? No. We have a color complexion problem. There'll be dark problems, etc. But beyond the point, they'll be finite. But if the way I do an abduction, I bend this way. Someone else might bend this way. Those are cases which I cannot interpret because that depends on human behavior. So you don't want to get into that kind of... So you split the problem into two parts. You split the problem. And I think that's why you see companies like Voyage and all that who are deploying autonomous carvings, close proximities in a control environment, they're far more successful than the ones which are on the road today. Even the best of companies have not been there.
01:08:28
Speaker
So that's how I mean my negative case is that you cannot train for every possible scenario. I mean, you cannot train for every possible scenario, you cannot have a complete document. So we've divided the problem into two parts. Now that I've been doing it for the last year, but the idea that I'm coming back to this to how I came, so I told her then, and she said, please don't do this to me. I'm already using these four or five different pieces of software tools, et cetera, which don't talk to each other. Half my time spending integrating all of this, right? And you're giving me two more tools. And that's why I advise this case. When I am running Ola as a marketplace, as a driver, as a customer, there's a marketplace, there's pricing, there's government.
01:08:57
Speaker
All of them is operating seamlessly. When a driver comes on board, I see the license is valued, then I give him a car, the driver goes to the destination. It's all seamless. So why is all this not stitched together? And then I see that there's a big opportunity in stitching all of this. So after I look at a lot of spoken-love clinics in India, so my inclination for, obviously, physical rehab, sports and all that, and then all the clinics offered this. They were using four or five different pieces of software. In fact, one of the soft clinics in Delhi I spoke to was using Asana.
01:09:24
Speaker
to handle the patient so that they could handle a patient from one. So every patient was a project, right? And then this project management. Through her contacts, I booked a visit in South Africa, US, UK also. My contacts are usually UK. And everywhere, common feedback. Six or seven different pieces of software, all not talking, a lot of administrative work. US, the problem is far more complex because it's billing, it's working with insurance company, right? So who stitches all of this together? And then you look on the patient side, my mom was offering, she was due for a new injury at that time.
01:09:50
Speaker
any replacement at that time. And we could not get it done. We had to delay it by one minute because no operations, no physical therapy, no replacements happening to COVID. And then physical therapists would also not come home. So she had to do it all by herself. But why does she? So how do we give her, make care accessible to her the way I had accessible to me when I was injured? So when you marry the two,
01:10:08
Speaker
It seems like we've lost your deal. I was very sure the next business I build has to grow the industry.
Challenges and Opportunities in Healthcare Technology
01:10:13
Speaker
So there are 2.5 million businesses. So anyway, people want that service, but because there's so much friction, they're not. So the examples I use is slipkarts, swiggy, hola, for that matter. We always wanted caps to come delivered because they could not. That's what hola unlocked. We always wanted food from outside, but they were so broken. So physical therapy is one of them. There are 2.5 million who get impacted every year. Only half a million to billion end up availing some service to actually fix themselves.
01:10:36
Speaker
And also among the one people who actually end up using it, 70% drop of midway to the treatment because it's cumbersome to go to the clinic, stand in line, some pain I do that. And the number of people who are going to be suffering from pain is only going to increase because the lifestyle is getting more and more sedentary. The population is aging. So if I reduce the friction all the way from the end, the patient discovers that it has come to the clinic. They enter onboarding. There are lots of friction pieces along the way. The appointment piece, the intake piece, documentation piece. And then 90% of the rehab is happening outside the clinic when I'm doing it in isolation alone. I can streamline all of this delivery.
01:11:07
Speaker
then the friction reduces and my scope to actually make the industry much bigger this goal. So that is what attracted me to the problem. The other thing is can the margins be better? So yes in physical therapy and that's why we decided to work in the US first and not as much in India because in US people spend on physical therapy. It's reimbursed by insurance.
01:11:25
Speaker
It is a cultural problem. If I've been, I'm not going to call my friend or my brother to actually come and support me. I'd love to get rid of it. You can't ask your kid to do Malish. So, we've said that now that's changing in India too, right? India, people are getting needy, business done, people aren't okay to limp around, people don't want better politics. Now, I've seen today, I was happy. ICIC and Omar started replacing, reimbursing 12th video therapy sessions with them.
01:11:46
Speaker
mental illness. I think that this industry will grow and this needs a better software, right? So there are two things that are different about this industry and there's no software catering to that. One is it's a referral-based business. It's not a standalone business. You get patient from an author, you get patient from a neuro, you get patient from a... all these things are happening off facts even in the US today. Facts are paper-handed and notes. How do you streamline that referral? Second is the patient-doctor interaction year goes on for weeks and months or years. Unlike, okay, you get a medicine.
01:12:10
Speaker
and you just take a medicine and your interaction is over, you get a surgery, a surgery is done and I don't have much to do with you, right? Then you can do it. You have the interaction because you have to go through a rehab process and who is managing the pipeline or the interaction with you, who is making it seamless for them to integrate. And the third is every treatment is personalized. It's not like you take a dolo or you take a grocery, then you'll be done, right? It always depends on what your past was and what your goals are. So if I want to be an athlete, I have a different regimen. If I want to be a delivery boy, I want to be a warehouse worker, I have to do a different regimen and obviously depends on my past context. I can't really start doing this.
01:12:40
Speaker
It has personalized it. When there are so many complexities needed and the world is going to be suffering from a lot of pain, how do you make it possible? So I don't want to get in the digital clinic. I don't want to enable the physiotherapist rehab, the strength and conditioning coaches to streamline the entire care delivery, increase the lifetime value of the patients. And then on the patient side, I think that I'm doing all of this in isolation, which is not fun. No one is a rehab partner. You would have thought of a spotter of a gym or a learning partner. You've not heard of a rehab partner.
01:13:05
Speaker
So who is handling me through the entire journey and I should feel connected, giving me feedback. Like when you do blood sugar, if you're going through diabetes reversal program, you can measure blood sugar and see your blood sugar is dipping, or if you're going through weight loss program, you know your weight is under control. There's no such metric in mobility. The only thing is whether I have pain or don't have pain. Like how do I know my range has improved, my flexibility improved? So you need to give the object. And when you give users objective feedback, they will act on it because it's in their interest.
01:13:29
Speaker
So that's what seemed like the big opportunity as you see, right? It's a number of people getting impacted as 2.5 billion people in the world are impacted. US spends close to a trillion dollars on mosquito skeleton cost of productivity and actual cost both.
01:13:42
Speaker
the big industry potential to transform the industry, right, increase industry, and then therefore make more margins. And then you're part of the care deal. So you, as I said, I was not part of the outcome, right? The bus was bad, I could not afford it. Or if you recover or not, I'm equally responsible along with the doctor. I'm monitoring you, I'm reporting back. But then there's a potential of making good money as well, right, along with it.
01:14:00
Speaker
Those are the things that attracted me to the opportunity. Somehow mobility appeals to me all the time. So yeah, this is what I just thought is a great problem. Obviously I spoke to the management role, I sought a 7-1 notice video and then moved on. But yeah, this opportunity excited me and it still continues to excite me. I didn't wait for it single day. I think I left on March 31st and first every last full time on this side.
01:14:20
Speaker
Are you monetizing per transaction basis or are you selling as a SaaS monthly subscription? So we are selling as a SaaS service. So there are two parts, at least in the US. One is a subscription. So there are two parts of this platform. One is a practice management piece and the third in the second is a payment handling piece, billing piece. The practice management piece is per month, for a therapist per month, etc. is what we charge.
01:14:43
Speaker
Then the billing piece, how much money you get from the insurance company on behalf of the doctor, you take a percentage of that. So you get busy, right? In January, February, we won't make too much revenue because people have just reset the insurance plan and they don't want to use their deductibles and all that. But towards the end of November, December, people are exhausted. You are reaching the end of the insurance plan, you want to use it. So that will depend on the user.
01:15:02
Speaker
The third, which we're developing, which is monitoring the lab person remotely and all that. So that's going to be a user-based plan. Every user that we see, but that's something we're still in the early days. We have to still get clinic. As you said, in healthcare, the initial cycles are slow, right? You have to get clinical validations of these. You have to publish research papers that this actually works, and then you can start monitoring them. We are away from monetizing that part of it, but we've started monitoring the practice management and they're building the insurance piece of it.
01:15:25
Speaker
So the practice management piece would give them a starting from a like a lead gen form or a landing page where patients can fill an inquiry dinner. Yeah. So how do you, so what is happening in physical therapies that direct access of patients is increasing. So earlier, six, seven years back, you always need a referral to actually go to physical therapists. Now that's not the case. Younger ones are more aware. They go directly. If you see someone like a Coley, that was his own therapist. So you Google it and it's one of the highest such keywords. So how do you streamline that inflow?
01:15:54
Speaker
then officer referral dollar is 70% of the traffic, so how do you streamline the interview? Then the whole appointment management site, so there are multiple therapists in office, how do you manage the appointment? As part of appointment, the digital intake is done. You send the form, you ask them the goals, the problems, treatments. When you come to the clinic, the doctor then wastes 47 minutes doing that. In US, we also check for the eligibility of the patient, how much you're eligible, not only for the information, and also PT-specific eligibility, there are restrictions on physical therapy depending on the insurance plan that you have. Then the entire documentation piece of it,
01:16:20
Speaker
How do you document the journey? The progress? The deterioration? I mean, we made it very objective. We made a form in such a way that there's no need to type for the doctor. It's all click-based. You digitize every workflow. Then there's analytics as to how much time is it taking for people to recover? What are the things that are working? There's an engagement platform for the doctor. The doctor, when they're doing
01:16:38
Speaker
whenever the patient is doing something at home or the doctor not there is a dashboard for the doctor and then there is a billing engine obviously that is where the practice management piece of it comes in and the billing piece is where we have integrated the insurance companies as soon as the loads are done we submit it to the insurance company with any corrections needed we correct them we get the money from the insurance company we deposit to the bank or to the doctor then there are some patient receivables so we
01:17:00
Speaker
And then we send a link to the patient and explain to them why they have to owe the money. They'll get money from the patient and you can sell it to a doctor. So, and it all of that for them. You spoke about like that observing patient and giving him feedback and all that. So that you're saying is going to go live after a year. No, it's live already. This little video I've shown you a demo of it is quite fascinating actually. We may make a personal course there. And some of the times I can show you a demo of that actually I can send you a link. Where you do the
01:17:28
Speaker
You do your assessments in front of a laptop. You follow a guided coach. You have to develop your own content. And then a report gets generated very similar to a blood report that your range of motion is over. Now, we've started doing that somewhere else, but we have not. Obviously, we have to get the clinically validated works in all conditions, all body types before it starts being used for medical problems. It's live yet where we're devolving it. And we have a partnership with Manipal University.
01:17:49
Speaker
which is the number one allied school in India. So now they are testing this over a range of 100 to 200 patients and see if the accuracy of the results holds true across all of them. There must be a lot of treatments now. So you're creating this coaching module for each of those treatments. There must be like two, 300 different treatment plans. So there are, they have all assessments. So there are 200 different types of functional assessment depending on the body part and depending on the injury. Right. Functional assessments we call them.
01:18:13
Speaker
Why they call assessment? This is an exercise, right? Yes, there are two parts. There are assessments and there are exercises, right? So right now what you digitize is automated is the assessment part of it. Red exercise is we have not yet automated digitized. So assessment is my sister used to ask me turn the hand around so you do.
01:18:28
Speaker
touch the body, but you know the extent of the pain, right? And you know how your recovery is on the track or off track and assessments are easier to do. So my theory was that people are, it's going to be impossible for people to switch on the laptop every day and extend one of the laptop. That's very, it's not a normal user way, but if you have to do something periodically, then the chances of doing that. So we wanted to nail that use case first before you move to exercises, very focusing on assessments, another 200 and there are like 400, 4000 exercises.
01:18:52
Speaker
So there are exercises not 200, then that's 4000. So there are 4000 different exercises depending on the body part, depending on the extent of the injury, depending on your age also. So that we end the process of digitizing, but at this point we have digitized the inter-assessment protocol. So if you are a dancer,
01:19:05
Speaker
valid answer for example and you want to you want strong toes you want to assess your toes on track your legs are on track then you can do that if you're a swimmer then you need a stronger lower back right if you're a golfer you need stronger this so then there are assessment program for different cohorts based on your profession and based on your... so in these assessment program they would see a video that stretch your arm like this and then they would have to copy that yeah so it would be a guided program
01:19:28
Speaker
Basically, you were taken one by one step, so we've redeveloped the own content, right? And it's interactive, so there are two steps, step one, step two. The content would ask you to do step one. Once you do step one, then the content will move to step two. So if we have a personal coach rather than asynchronous coach, like working on YouTube, auditioning different fitness apps, would be that they do something and you're doing isolation. Here, if you're doing one prep,
01:19:46
Speaker
The other app will only start when you finish urban. It's very interactive and that's how you develop the content. So there's almost like a chatbot. So when you do that, then it'll probably say, OK, that's great. Now let's do something like that. Yeah, that's what. So doesn't machine learning need a lot of computing resources? So you do this on device, or how do you do that? Yeah, so it's a very good point, actually. I'm glad you asked this. So there are, yeah, it does. Because there are 30 frames per second.
01:20:13
Speaker
And you have to process every frame to actually start making sense out of this, right? And then you have to do key body part detection, you have to do analytics. So divide it into two parts. So one is what you do on the edge, which is under. So today we're only supporting the browser. We're not even on the mobile yet because we have to get the user experience right. And so we do any little processing on the, so there are, so there is detail, there is basic analysis and the detail analysis. So basic analysis is whether you are doing the right exercises, you're following the instruction, we do it on the end, it doesn't require too much processing. So for example, if I have to turn left hand to it, am I turning left hand doing it?
01:20:43
Speaker
if I'm sitting on chair and ring, it's very basic level of processing, something we do on the browser, which is sampling of every three seconds, et cetera. But we send this video to the back end, where we have this cloud infrastructure hosted. Then we process the actual detail. That's why the report comes to you by email. So we process the report by email and then submit it. So it's a very good point. But the job on the front end is to ensure that there is minimal noise, which goes to the back end. So if there are two people, you will not proceed. If I'm standing left or right, I'm doing the wrong exercise. So those are basic level of filters you can put. And I think machines, laptops are powerful enough to actually do at least that kind of thing.
01:21:12
Speaker
What you want to do involved, you send it to the backend where we have obviously a cloud infrastructure which does this processing. This happens pre-appointment or after the doctor? At what state does this assessment happen? Various use cases. So if you go to many ED websites, they can just get a free assessment done. So you're not sure whether you need intervention or not, right? So you can get it then without it. Second case is you are in middle of a rehab.
01:21:33
Speaker
and you're supposed to rehab at home, you've not been able to come to the clinic because you live somewhere far off or you work. You can send this assessment and the doctor can ensure. Third is you're done with your pain treatment. You've not moved to strength and conditioning, but you also want to ensure your pain doesn't come back of your limitation. So that's why we use it as a great lever for engagement and not as a way to cure it. So there are three scenarios we've identified.
01:21:52
Speaker
where people can get start of the season. For example, if you're starting a football season, etc. So you can do the different triggers of which. And that's why we leave it. We don't say that we are an assessment putter. We enable the doctor to make this customer assessment for the patient because he or she knows what the patient wants. Send it to them, interpret the report and do it. We enable the entire create, make it easy for them to create, send it to patients, engage with them.
01:22:13
Speaker
Like within a clinic, each doctor will have their own login in it where they can look at their patient's case history and make notes. And there is access control, right? So the clinic owner, the clinic owner looks at data of all patients, all doctors, and the smaller doctor will only look at it. The admin staff can only look at the appointments, reception, the finance can only look at the invoicing, all of that. So then there is, because we are now HIPAA compliant, health care is very regulatory,
01:22:37
Speaker
regulations are far more important. So we've started with HIPAA compliance, we'll be moving to other compliances later. But sanctity of data is very crucial, especially now that we are launching, we've already launched in the US and we want to do some work in the UK also, which are far more stringent. We have to be compliant from day one and make sure of these conditions from day one. And what is the pricing at? So the end-to-end package in US, we are selling at $300 per PT, 50 records per month. And on the billing part, we are charging around 6% of the overall GM.
01:23:04
Speaker
But it sounds like good unit economics. Yeah, definitely. I think economics are very good because we build a platform and we automated most of them. So in building economics can be bad if you operationalize the whole thing. If you do the entire claim submission process manually, then it's a very expensive operation. We automated the entire thing. We've integrated clearing houses, insurance companies, rule engines, et cetera, to make sure that it's not an automated process. And because our practice management and the billing talk to each other, so you don't have to have a manual layer in between to do that. If you don't do that, it becomes cumbersome.
01:23:34
Speaker
It also becomes difficult and our advantage is we focus on one physical therapy. So we are the best product in physical therapy. Are we the best product for healthcare? No. But then I think the world is moving from horizontal solutions to vertical solutions and specialize in the industry. We want to capitalize on that. Customer acquisition, how are you doing it? Like you call and send LinkedIn messages.
01:23:52
Speaker
So yeah, LinkedIn is a good way. So I'm on part of some Reddit group. So I'm part of every active physical therapy community in the world today. And I, that's how, and I had some friends in the US for my sports clinic from whom I reached out to have. And now we have a customer who are putting us to that. So it started with me reaching out to people. We didn't have a product few months back. So it was conveying the vision and beginning the buy-in and showing them screenshots of how this will look like because product is now ready. It's less than a one year old company. So we were developed and the US needs us to get product.
01:24:17
Speaker
relaying the vision, talking to them on how it will look like, how it will shape up. And then, obviously, keeping a touch point with them. Once they're interested, they knew their effort and someone else's. Now we are slowly moving towards setting up an institutional process, but mostly it was me and my co-founder reaching out to people. And now we started the mailers and all of those things. It's very early in the day to figure out, but the margins are like 70-80% on this one. So CAC is never going to be a problem on this one, I think.
01:24:38
Speaker
When did you go live? Like when did you first have your first commercial? So we went live in India after, so we started working with Therapists in India, I think in July after the second wave. So July is when we had the first free deployment in India. And then in US, we started going live in December last year, 2021.
01:24:54
Speaker
Like India, what is the pricing? I thought you were focused on US. Yeah, India is a good test market for us. The science of physical therapy remains the same in both places. It's a good test market for us to test. For example, there are some really good innovations from India which you can use in the US as a differentiator. For example, packages. In the US, it's typically been a fee for service culture. Every encounter you build, so no one thinks about
01:25:16
Speaker
How do I take care of the entire plan of care? India, on the other hand, is a cash economy. You want to lock in the customer for 10 visits. It's a package. It's something we developed for Indian physiotherapists. Now people in the US are loving it.
01:25:26
Speaker
and we marry that with the insurance delivery, that's a great thing, right? So the science is the same, I think India is a good, as I said, it's a growing market. It's not a very big market, it's a growing market. And the other good thing about India is that a lot of Indian physiotherapists have friends in the US who have either studied them in the bachelors or something like that, so they refer to us to their friends as well, right? So we are going to be a global platform. So India is going to be a testbed, US is going to be a primary revenue driver, but no reason we are not going to be expanding to Australia, not Australia, but UK after this, Australia after that. Because physiotherapy, you cannot be a vertical company
01:25:56
Speaker
also then be another vertical in the company and then you have to be global if you want to be a big outcome. We cannot afford to be both. So at some point you have to take a global stand and I've launched work with Ola in London and I understand the benefit of actually launching in the international markets on day one. It's likely more tedious and difficult but I think it rewards in the long term so that's a call you've taken. But yeah India is a good place, 100 clinics right and even when I go and tell someone in the US that I have 100 clinics that adds a lot more credibility than saying that okay you are my first customer.
01:26:23
Speaker
And what is the India pricing? It must be not in the same level. Six fifteen hundred for physiotherapists per month rupees. So you told me that your sister had that problem of remote consultation like using Zoom. So is that also built in like for video consulting? Teri Health is what we are building in. We have not built that in yet because physiotherapists is not very comfortable using Zoom. Since in-person appointments have started that telehealth consultation has been going down. So is it like the most
01:26:50
Speaker
As I said, there was a lot of fundamental problems in this sense. So these are good to have. So accounting is broken, billing is broken, appointment management is broken. So using so much friction on the fundamental layer only that you want to address that first and then obviously telehealth will be an opportunity and the appointment is less than 5% today.
01:27:05
Speaker
It's not the process, with the clinic that we're working today. I'm sure the new age ones will do more daily and then we'll cater to them at some point. Ideal customer profile today is someone who has a clinic and someone who wants to operationalize it better, use it better, and he wants to expand the business a little bit. We have a better customer service. The ideal customer profile for us, so when we get to our profile who are online guys, new age guys, the problem with going to new age guys is they don't have that much football. So I wouldn't make too much money from them today. So there's an underlying opportunity with the existing providers.
01:27:29
Speaker
They have a good footfall, they are integrated with insurance companies, make money from them and then you can actually start catering to them. You want to focus initially on startup, not be everywhere. So probably in the US, this would not be such a big ask, but in India, it might be important. Do you also do like marketing support, like helping them get customers? Yeah, we are building that out. So that's important. I think direct access is increasing, right? So we are not saying that we'll get you more people. It will happen automatically. We are saying we'll streamline your marketing intake.
01:27:56
Speaker
So basically, you get people from Google, you get people from Facebook also, so understanding all of that, posting your appointment, which is to be integrated everywhere. We're giving you a GPT-3 console where you can put in the keywords and content that's generated, you can put on your blog, you don't write the content for you because it's impossible to do that. When a patient goes, we post your Google review to them. When the Google review comes in, we post it to Google Map business page, which automatically updates.
01:28:19
Speaker
So there is so much entry, there is so much search for physical therapy, and there is so many broken processes here. We just believe that in the initial days, if we stay there too, automatically the intake will happen, and we've done manual experiments to prove that. Does it flatten out a few years? We don't know. By the time that flattens out, we'll have to see what we can do. But yeah, so we cannot promise that we are not going as a marketing company, we are going as an automation and administration company. So we don't want to take the page that we increase our users. That is difficult because
01:28:42
Speaker
Then I'm also, I don't have a loyalty to two clinics next to each other. Who would I give the patient to? I don't want to be in the conundrum tomorrow. I'm enabling all of them and I'm enabling them tools that they can do to streamline the process. It's how effectively they do use the tools and decide how well they do, right? If I think of it in the world, we would like to be like a Shopify rather than an Amazon. We would like to be a Shopify when that's a call. So because then become the conflict, whether you're on the side of the consumer, because there are two conflicting opportunities. The consumer would want to speak most of the doctor provider, so which one side is. I'm saying I'm enabling the provider to disseminate care if I give him enough tools and of course,
01:29:12
Speaker
They will be able to get care to everyone at a low cost, that's the call we are getting. It's more of a Shopify approach than an Amazon approach is what I would say for us. And we believe in healthcare, the doctor is always limited to the centre of delivery. I think we're far away from automating a doctor. We've not automated cars yet. We're not already in anything. So we're far away from automating a doctor because in healthcare there's a lot of context, there's a lot of empathy which is outside of workflows. So enabling the doctor will be a better bet at least at this point of time.
01:29:37
Speaker
the business will grow too. Like you have some projections like say by 2025 you will hit this kind of ARR. So as I said it's a 40 million dollar industry in the US itself and it's growing at a 6-7% CAGR. So when you define this industry itself should be a 550 to 60 billion dollar industry where as a part of it is our 10 billion opportunity software. So it's a 5 to 6 billion dollar opportunity just in software.
01:29:58
Speaker
Now, there are incumbents, it's a broken industry, there are long pillow operators, so in the 5 billion industry, you check up with 10% of the industry circuit, it's 500 million ARR opportunity right there. This is the US, I'm not talking about the global industry, if you agree with the ratio, it's right. So yeah, that's the way we're looking at it. And we're playing in the entire valuation, we're not just saying we're a tool company, we're saying we're doing a patient intake, we're doing a billing piece of all that. So the software opportunities close to 5 to 6 billion in five years from now.
01:30:23
Speaker
And if you take 10% market share away from that, that's not a difficult word because there's only one incumbent. It's almost a green, brownfield opportunity of sorts. And then the industry is changing. One incumbent, does that incumbent just provide administrative software or do they also look at? Yeah, it just provides documentation software, documentation software. And just with documentation, they are available.
01:30:46
Speaker
So they are not into outcomes the way you are, like that focus and outcomes. Not in the outcome. They are making reimbursements easy, nothing beyond that. And the good part for us, industry is changing. What is happening is telehealth is now reimbursed.
01:30:58
Speaker
I can get money for doing telehealth visits. I can get money for remote monitoring because there was a conflict of interest. Why would a doctor not call me remote monitor me when I don't make money for that? Others' direct access is increasing. Reimbursement models are moving more to a value-based care. So there are a lot of changes happening, paradigm changes happening in the healthcare industry, right? And it has to happen. Covid has changed a lot of things.
01:31:16
Speaker
and the budgets which we see, the national debt which we see because of healthcare spend also is also increasing. So things have to change. And we are the center of all that change. So we enable the providers to go to the leapfrog to the next level of change. So that's where we see the opportunity because people need triggers to move, right? And external triggers are far more compelling than internal triggers. So the external triggers are there and a lot of these external triggers are there which help us move there.
01:31:36
Speaker
And that brings us to the end of this conversation. I want to ask you for a favor now. Did you like listening to this show? I'd love to hear your feedback about it. Do you have your own startup ideas? I'd love to hear them. Do you have questions for any of the guests that you heard about in this show? I'd love to get your questions and pass them on to the guests. Write to me at adatthepodium.in. That's adatthepodium.in.