Andrew: Hey there, freedom fighters. My name is Andrew Warner. I’m the founder of Mixergy, where I interview entrepreneurs about how they built their businesses. Joining me as someone whose mom was disappointed in him, mom was an entrepreneur herself. He decides he’s going to quit his job, get his brother to quit his job.
And they go start a business. And his mom was worried about them and was a little disappointed. Meanwhile, they ended up doing well for themselves. They created this one company. that let people stream Bollywood music. Am I right?
Swapnil: Yup.
Andrew: What was the name of the company? How do you pronounce it? Dean Ghana. And then he, uh, they sold the business to RDO.
He went on to create a next company also with his brother Messi. Mezze is like, imagine if you could imagine if you had the ideal uh sistant and you could just message him and say, I need to fly from here to there. I have these flight, this flight requirement, and it actually did. You got the results.
That’s what matters you want to do? Am I right? You want to do that all, um, using software and you eventually did it and sold it to Amazon. Why am I looking at your face? And you look like I’m disappointed in the way that I’m explaining it.
Swapnil: Yeah. Yep. So MSI was an air power traveler system. That could plan and book travel for all business travelers and the way it worked was a lot, a very simple messaging interface. So you could check with MSCI and say that, Hey, Missy, I want to travel from San Francisco to New York by business class, and I’ll be staying closer to the downtown.
Can you find me the right, uh, for tend to stay the right flight and also some restaurants from a dinner reservation, et cetera. And MSI will use to basically do all the research and presented with only three options. And, uh, there was that there was a very high conversion from us presenting with options.
And you actually look in the camera, but behind the scenes, it was a human assisted AI platform. So we had travel agents on our team, but we used to collaborate with our app. Our platform is then deeply integrated into the entire travel ecosystem. Look for all the flights, restaurants, hotels, everything.
Andrew: And you made money from.
Swapnil: Oh B we basically licensed that platform to companies like American express travel management companies and other players who already had travel offerings to their customers. So someone like Amex started using MSCI to automate their travel concierge. And that’s how we started our partnership with them.
Andrew: The voice you just heard is Swapnil shin. He is the co-founder of that company. Um, today he’s doing the same thing. Applying it towards bookkeeping. So if you wish that you had the perfect bookkeeper, that would organize your books for you in QuickBooks, and also you could text them in the middle of the night and say, I want to know how much money we made yesterday, or last month the software will do it.
It’s called Zenni. You can see it@zenni.ai. I want to find out about how you launched this business and we can do it all. Thanks to two phenomenal sponsors. The first will host your website, right? It’s called HostGator. Check them out at hostgator.com/mixergy. The second. Is rippling. If you’re paying people rippling, we’ll do it.
And Zenni works with rippling. I’ll talk about that. And when I do the ad for rippling and tell people to go to rippling.com/mixergy Zenni for, um, excuse me, Swapnil first, how much did you sell the business to American express for? It happened about two years after launching. How much was it for
Swapnil: Yeah, it was, uh, somewhere between a hundred and $200 million.
Andrew: that is amazing within two years.
Swapnil: Yeah, it was, it was a pretty, I would say rewarding and exciting journey for us because that was a time when artificial intelligence was just hitting the market. And it was an SN days and we were able to not only build, but also scale up probably the first year, our travel as a service platform in the world.
And we scaled it to more than 10 million travelers and launched it in several countries around the world.
Andrew: And you raised, you raised very little for it. How much did you raise?
Swapnil: we had Reese close to around $11 million over two rounds of funding. So it was extremely, uh, early stage. We were, I think, post CDC at that time when we sold.
Andrew: Wow. We, how did life change for you day to day after that sale? If at all?
Swapnil: Yeah, I would say the skill that we started operating, operating at post acquisition was very different because, uh, as MSI, we were a startup, we had probably worked with somewhere around a hundred to 200,000 travelers still. Then when Amex acquired us, suddenly when we launched, we were talking about at least a million.
Card members should be using MSCI in the first few months. And then we scaled it to 10 million and then, uh, probably ends up millions. After that we launched it in several countries around the world. We grew the team significantly. The integration was super tight. So it was, I think, pretty, pretty rewarding to see a platform like MSCI being such deeply integrated into Amex.
Andrew: What about on a personal level after the sale? Were you able to finally buy a house in the Bay area or, or anything like that?
Swapnil: Yeah, well, certainly as well for booty and snail, it was probably one of our, uh, financially the biggest outcome today. We did upgrade our, our system. We moved from Sunnyvale to Los Angeles. So that was, uh, uh, a good to have. But I think apart from the monetary outcome and yeah, in the, in the sale of the house is the most rewarding part I would say was really building a team.
And affecting the lives of pretty much all the other people, people in our team in a way that it made them financially independent, probably for the restaurant, their lives. And a lot of early executives made a lot of money for the big exit. But also I would say career-wise, it actually made everyone part of a very deep success story that will live on for, I would say, decades to come.
Andrew: I remember the time we’re talking about you launch. Why 20:16 AM I right? Somewhere around there.
Swapnil: Uh, 2015. Yep.
Andrew: so around that time, uh, there were all these text-based assistance that we’re trying, like w the big one that got a lot of attention in the tech space was magic. The idea was right. You pay them what, uh, they were playing with different pricing. You’re smiling. You recognize it. what, how would you describe what magic did.
Swapnil: I think the way I would respect magic is tell us anything and we’ll get it done.
Andrew: Right. And it was supposed to be done by, by like an artificial intelligence software, something, but in reality, they were starting with real people and eventually they were going to build up to, to software. Right.
Swapnil: I don’t think they ever spoke about it being AI. Only our air power, but it was more, it got a lot of attention because they said that, Hey, you can now check with us or a simple, a text interface and we’ll get anything done for you as long as it’s legal and to us, that cause the biggest problem, because the moment you say that we, as a product can do everything you are getting ready for failure because doing everything is, is very hard.
Andrew: You came up with the idea, because I read the Forbes story where you and your wife were driving. She asked you to buy something. You said, I’ll do it. You’d look at your phone. And you say, wait, I wish I could just text message and get somebody to do this. Why do I have to go on a website and figure this whole thing out?
You go back, you talk to your brother. The two of you say, you know, we could do this. When you were thinking we could do this. Did you think all shopping in the beginning, or did you think everything the way that magic and others had gone or did you automatically zero in on travel?
Swapnil: Great question. So by the way, that was my identical twin brother Snellville was traveling with his wife and, uh, thought about this idea.
Andrew: Uh, it was he who had that experience. Got it. And he’s the one who then called you. Okay.
Swapnil: Yeah, he’s the one then will then call me. And then I take all the credit of building the company.
Andrew: He gets the credit for coming up with the idea. And then in the beginning was the idea after he had that conversation with his wife, where she wanted him to buy something, was the idea to automatically go into shopping first to do something else. How did you pick what area to focus on?
Swapnil: Yeah. So I think our initial thesis was that all of us are going to live extremely busy lives as we move on. And, uh, it would be amazing if we could have a hand behind our back, make a concierge who can help us simplify our lives on a daily basis by doing some important things for us. And hopefully if we get that experience right, then we can transform the lives of those consumers are repeater period of time.
So with that as our key vision, we launched MSCI as your shopping assistant. So we said that if you want to buy anything, Then, please use MSCI to measure your shopping assistant. But what we realized was that in the first six months of us going live, we saw that people were buying things from, I would say, 15 to 20 different verticals.
And when we started looking at the data and analytics, we realized that travel was growing at, uh, I would say triple digit growth every month. And people who spend on travel actually came back and spent on other things. And the average ticket size for travel was thousand dollars. Whereas for, for fashion and other verticals was a hundred dollars.
So when you actually use MSCI to buy travel and spend that thousand dollars with us, the trust that you established with your concierge was extremely high. And then we, we pretty early, I think we were one of the very few players who realized that building AI vertically in an extremely deep manner would be far more powerful than building something horizontally horizontal, which was, I think one of the major mistakes that most of the players there and hence weren’t able to survive, but then we repositioned ourselves as your travel assistant and started building the Adam.
That was extremely deep for the travel
Andrew: let me pause. Before we get to the AI, the first version, when it was all shopping, was it, your software was just cha routing the messages over to real people who are sitting at desktop computers and helping your customer.
Swapnil: Yes for the, for the first, I would say three to six months, uh, it would be stupid to directly build. Yeah. To do something that we don’t know what, uh, we didn’t know how consumers would be using us. So we actually use real people to simulate all those conversations. Then over a period of time, we learned that what are the most mundane tasks that they do?
And we started chipping away with using technology.
Andrew: Okay. How many people did you get to when, before you figured out it was going to be, um, travel?
Swapnil: Uh,
Andrew: how many, uh, customer service people, I mean,
Swapnil: I would say around, uh, 15.
Andrew: okay. 15. And it was you personally, not your brother who created the first version of the software.
Swapnil: Yeah. So I actually coded the entire, uh, I would say the front end, the products that were facing the consumers and snail build the entire backend.
Andrew: Okay. And so I get how you did that. Your goal was to get consumers. You told our producer. I think I went, we went out to our friends and our friends started using it. Once you got past your friends, how did you get more people to sign up for? Mezze.
Swapnil: I think a lot of them actually started spreading the word to others. And one of the key things that we did was, uh, trying to see if some of the influential people in our network who are early adopters of new products would be interested in using MSCI. And when they actually liked it, they started sharing it because these early adopters are always evangelists.
So they will talk about your products, spread the word, and kind of gather that initial setup.
Andrew: Who was an influencer? Who did that?
Swapnil: So a few of us actually, uh, some of my close friends who are, who are working at eBay PayPal, uh, then, uh, some of our investors that are advisors that, uh, we had been in touch with during our previous startups, they helped us spread the word.
And, uh, I would say then we ended up raising our first round, which was a seed round from nexus ventures. And post-bac we also had a PR firm. We did some PR around it that also had to increase the initial set of users.
Andrew: There was a, there was an article about you guys, 2015 in tech crunch and deepen the article is a list of all the competitors. I don’t know how many of them are still around. There was go Butler. Those operator. Remember operator is a ton of money, tens of millions of dollars. I don’t even know if they’re still around agent queue.
There’s an experimental feature from Facebook called M there was a bunch of stuff and you dove into it thinking these players exist. We can still win because what we can still survive.
Swapnil: Yeah. So to be very Frank, I think. Some of these started at the same time that we did, they might have this far more money than we did, but I think what really helped us when was not focused on the technology part of it go Butler’s magic. They started building like this huge customer support teams and operations team.
To support their growth, but we actually, instead of focusing on the growth, we focused on building the right technology early on. We started adopting AI machine learning neural networks to, to build the entire experience. And that really helped us grow probably an ear on the line instead of going in, in the first year itself to a level where you did a lot of you wants to do this job.
Andrew: So first it was, it was humans. It was all shopping. You zeroed in on, on travel. And then. You don’t think you ever did advertising to get customers, right? You initially went into partnerships to get
Swapnil: Yeah.
Andrew: That was, that was a key thing too. And maybe that’s why also you didn’t show up on my radar a lot because our, at all magic was constantly out being talked about you weren’t because you were powering other things.
Well, software,
Swapnil: Yup. That’s right.
Andrew: how did you get those deals?
Swapnil: So we, uh, I mean, because both me and snail have very deep technical backgrounds, we also spoke at a lot of conferences on how we were building our technology stack, et cetera. I think that increased, uh, a lot of interest in MSI and then to a level where we actually never did outbound media effort.
We had inbound interest in wanting to partner with MSCI. So we actually closed down the platform and we had initially set up five to 10 customers, and these were big customers, like big banks or companies like American express, big travel companies who wanted to use MSCI
Andrew: And they found you because you spoke at conferences, their developers saw you speak, and they said, we shouldn’t build this. Here’s a tool that we can implement. Let’s just pay for this tool instead of building our version.
Swapnil: Yes. Some, some founders slider conferences. And whenever we spoke at conferences, like VentureBeat, I still recall like data profile on MSCI at crunch, as you saw data, uh, profile on Mesos. All those, I think are things added up to the, uh, to the, uh, I would say the branding and awareness about MSI.
Andrew: Enterprise sales, especially when you’re going to be talking to so many of your customers, customers. It’s really tough. I’ve seen people here in Silicon Valley do it, who are just so outgoing, so full of friends, and I could see why they do well. Your two developers, who aren’t looking to go in glad handle and travel to all these different events and right.
You’re not putting on these big performances for, for enterprise how’d you close all those sales. They just came to you and said, we want your software because we saw you at venture beat.
Swapnil: Uh, not at all. I would say, I think, uh, whatever PR that happens around your product can. Uh, allow others to see it. And the end of the day, you should be able to send your vision, your platform, your uniqueness, your differentiator to these potential partner. Yes. And I think, uh, we have the right personalities to be able to convince this.
Like I would, I would say top decision makers to believe in what we are doing. Think of it as if you’re pitching to an investor. So it was more like an investor pitch for us than a sales pitch, because these are the partners we’re also going to help us build the product. So our first customer was in fact American express.
Andrew: express. So you got American express and they wanted to build you into what? Into their app?
Swapnil: W uh, the, the initial goal was to launch a white label app with American express. And we did that. It was called ask Amex.
Andrew: Uh, ask Amex.
Swapnil: Yep.
Andrew: Okay. And so you were basically the power behind the software that they were then using for their customers. That’s also pretty impressive that they would take that risk on you, right? Because if your startup failed, if what you did, couldn’t live up to their expectations. If you just started, uh, I don’t know, slacking off their software and their interaction with their customers would fail.
How did they protect themselves?
Swapnil: Yeah. So one of the things that Amex is great at is partnering with startups and they move also pretty fast compared to, uh, other, uh, I would say competitors that Amex has. And that allows them to innovate really fast. And honestly, they just don’t work with any startup. They work with startups who have strong teams who have strong products who are backed by strong investors, et cetera.
So they did their own due diligence. And I think for us as startup founders, whenever we start a company to do all that for us. So you always try to do your best and make sure that every especially critical partnerships like these succeed.
Andrew: And so you, you mentioned venture beat. I went and I did a search to see what articles were written about you and venture beat. I didn’t realize how extensively they co they covered you. Um, but. One of the things that they were interested in is the marriage of artificial intelligence and humans. And I see a few articles about that, that I think of one article, you said that your software could never fail because there’s always a human being behind.
If there is a search that software can’t do on its own, there’s a person to come in and help. Um, at one point, did you add artificial intelligence and what was it able to do in the beginning?
Swapnil: Yeah, that’s a great question. I think we started working on artificial intelligence, uh, in the early days, but the way we looked at the problem was that today, if a hundred percent of our. Uh, customer request ourselves by humans, what, which part of that request can be solved by machines force. So the initial part that we automated was instead of the humans going and looking and searching for flights, hotels, and dining reservations, the rewards used to do that with the same accuracy as our travel agents would, and then pick three recommendations that were extremely personalized for our customers and send them that.
So the recommendation part of our engine was first order by here.
Andrew: So I would say, I want to go to Sam. I want to go to New York. Your software would chat back at me, right? It was text-based. Where, where are you now? I say San Francisco. When do you want to go? I say that. And basically you’re doing the equivalent of punching it into a form on a site. But instead of punching it into a form on a site, you’re displaying it as a, as a search result.
You’re displaying it as text messages back to me. Right. And that’s all software does.
Swapnil: Yes. Yes. So the, the key thing to note here is that when you are chatting with Massey, we need to check with you as if. You feel that you are talking to a real human and for us to do that, we use, we build tools using natural language processing, neural networks, et cetera, to really understand that and present ourselves to you in a way that is extremely human or realistic.
Andrew: I said the natural language processing part. I get someone types in a and Y and you need to know. All right. And why in this context means New York makes sense. The artificial intelligence part, where did that come in? In the beginning? What did that do?
Swapnil: Yeah. So, so think of it like, uh, we, the way we always thought about it is like, if you actually picked up your phone and spoke to a travel agent, How would the travel agent react to the way you’re seeing, like you have the travel agent would first ask you, like, okay, tell me, where do you want to travel? Do you have any preferences?
You will say that here. I would. I typically like to sit on an aisle seat. I like to slide business when I’m flying for business or otherwise I flight economic when I’m doing the personal travel and these are my meal preferences, and these are the airlines I like to fly. And these are my reward points, et cetera.
And normally when I stay at a hotel, I like to stay at boutique hotels. I don’t like to stay at big chain hotels, and then you take all that information and you find the right matches for that particular customer. So replacing a human wine with artificial mine was actually where we use our AI and NP. It was a combination of all that that created that unique experience.
Andrew: Okay. Do you give, can you give me an example of something that would always go back to humans and you couldn’t, and at least until you sold couldn’t get software to do, but it was, it, it was repetitive and people had to keep going in and doing it anyway.
Swapnil: Yeah,
Andrew: want,
Swapnil: Yeah. So I would say that if envy sold. 60 to 70% of all requests were solved by machines. At least when it came to flight center hotels, but the remaining 30% was going to take significant more time. So I would, I would say that whenever you want to automate the first 20 to 30%, you probably can do it in six months.
The next 30 to 40% might take you an ear. And then the next 20% might take you two years after that. So the deeper you go, the more harder it is to automate things. And there are always some requests where let’s say you are people traveling together and we need to take care of them. References of all four.
Then obviously we will just have a human manage, all that instead of an AI, because AI was more fine tuned towards, uh, I would say managing travel for a particular individual instead of group travel.
Andrew: Just one thing that you did better with mezze than a real person would do.
Swapnil: Speed and accuracy and personalization. So one thing that we did using our technology was to create a living profile for each customer. We called it the, the travel genome. That means that every time you chat with us, we already knew what your preferences were from your previous chats with us. And. If you are, if you’re talking to us for the third time and you say that, Hey Messi, can you book me a flight to New York, then MSI would say, sure, we’ll be flying from San Francisco as usual.
And you will say yes. And we knew that you would like business class. We knew what your preference is for, and we’ll, we will actually give you the, those options. And that used to create an aha moment for customers.
Andrew: I do love how my assistant knows that I prefer aisle because I like to get up and go. I like that. She knows that unless there’s a dramatic price difference. I want SFO not the Oakland airport. Um, those little touches just help a lot that I just feel. Protected. I also love that in my calendar. There are certain things that are specific to me.
I like to, to just land and not know a thing about where I am or not, not even think about what hotel I go to, but no, it’s, it’ll be in my calendar and I’ll just show it to the Uber driver or, or. Punch it into Uber and go, that experience is just amazing. So you’re creating that experience. Why didn’t you go the route of competing with Expedia?
Why did you decide that what you do instead is, um, is partner with companies like American express and power, their software.
Swapnil: Because a travel concierge and not really another travel booking engine, which Expedia. Yes. In fact, we looked at Expedia or kayak as potential partners. And our goal was always to, how do we give you that concierge who can help simplify your life on a daily basis? Right? If you go back to our roots of our vision.
So that was our thesis and Amex platinum concierge is the gold standard when it comes to concierge. So partnering with them was super exciting because it directly,
Andrew: It could have been messy as the thing that we saw on television. It could have been messy that we were all interacting with for shopping, but you didn’t want to, is it because. The marketing to consumers is a skillset that’s just harder to develop while you’re building artificial intelligence.
Swapnil: No, I would say it depends upon your go-to market strategy. If we want to go to direct to consumer, then I think, uh, it will, it’s, it’s a whole different ball game where. All these little companies like Priceline, Expedia, kayak, they’re spending millions and billions on, on acquiring consumers. And from a go-to market perspective, I think it was a smarter play for us to go to consumers.
Why a partner? So it was a B2B to C model where the consumers would still be using our product. It would be powered by MSI instruct MSI as the key brand, but for us to acquire those consumers, it would be easier through our partnership.
Andrew: All right. Let me talk about my first sponsor. And then I want to come back and ask you about the previous business and why your mom didn’t want you to start it. My first sponsor is a company called HostGator for hosting websites. You’ve heard many of my guests say that. They’ve used. HostGator’s you’ve heard me say that I host Mixergy on HostGator.
If you’re looking for a hosting company, just keep this in the back of your mind. If you go to hostgator.com/mixergy, they’ll give you great hosting at a really low price in that URL. We’ll get you the lowest price they have here. It is again, host gator.com/mixergy. I keep bringing up the fact that your, your mom was an entrepreneur.
What did, uh, what startups was she involved with or she was entrepreneurial, right?
Swapnil: Yeah, absolutely. So she, I think is the one, uh, because of which we have this entrepreneur bug in us and it started at a very early age, I would say at the age of six or so, we always saw her. I would say building businesses and, uh, from scratch, like she had, like, she had a hardware manufacturing, uh, startup.
Once she ran that for several years, then, uh, she had a computer Academy that she ran for several years and then she had, uh, I would say arts and crafts video that she ran for several years. So we always saw her starting these businesses with a lot of enthusiasm running them. And she always used to tell us that.
Like having a steady job is so much easier because you always know how much you’re going to own and it’s so safe. But at the same time, we always were inspired by the work that she was doing. So at the back of our mind, we always thought that, Hey, doing something that you really love is definitely going to be super rewarding because we saw her doing it again and again.
So when it happened, like both of us were working at Yahoo, we wanted to start our first startup in Ghana. And, uh, I still remember, I told mom that, uh, I’m, I’m the one who is going to quit first and jumping it. And she was like, are you crazy? Why you’re doing that? You, you have such amazing jobs at Yahoo.
And then she asked her best friend to call me and she will make a, since you are leaving, can you at least convince me to stay at, at Yahoo? Because what if the startup feels like maybe after six months it fails in both of you will be out of job. And so, yeah, so she was always
Andrew: worry. Why didn’t you take her up on that? Why didn’t you say, you know, we are two brothers I’ll I’ll leave. He’ll stay, we’ll figure it out with when it works. We’ll both leave.
Swapnil: Yeah, because I don’t think like you have so much time when you are building startups, like every single day matters and you’ll get to kind of run ahead of time. So the more risk you take, the better of the chances that you will, uh, be rewarded and the chance of failure is so high. That you should put your 500% in it, if you want to do it or otherwise, I think it’s, it’s just increases your chances of failure.
And, and just one of us, wasn’t enough tech to be both of us,
Andrew: Where did the original idea? That company come from thin Ghana.
Swapnil: big Bollywood, Bollywood. We love Bollywood music, Bollywood movies. And when we came to the U S to, to do our masters in USC, we missed the Bollywood music iMessage. So many that we decided to. Host a little website in our own bedroom, through which we could listen. Our favorite songs when we were in college later, we started doing the same when we started working in Symantec.
And before we knew it, we told a few of our friends, but before we knew it, we had a thousand people who discovered that site out of nowhere. And then we had to move it out of our bedroom, to a hosting provider, et cetera. And that is how it started growing.
Andrew: It started as your bedroom, you broadcasting, I’m guessing you downloaded it. Is that right? You just
Swapnil: it was just our own personal library that we want. Yeah. Just for
Andrew: and it was on an ongoing basis, like a radio or was it a
Swapnil: It was interactive. You can, you could, yeah. On demand you can pick a particular movie or an album and you can go and see all the songs and then play it.
Andrew: Didn’t go to means happiness, enjoyment, fun.
Swapnil: Excitement, uh, going, going crazy with excitement and. I would say it has several emotions in it. Uh, having fun, uh, in an, I would say in an extremely excited state of there’s a lot of friends in it, funkiness in it. So it’s kind of that, that’s why the logo also short, like three people dancing on the
Andrew: And so it was, it was kind of like, it seems like a Spotify for Bollywood music. Your friends discovered it, even though you built it for yourselves, it kept going, how big did it get?
Swapnil: V when we raise that first round of funding, we had a million users using us.
Andrew: How what’d you do to get a million users?
Swapnil: Nothing. We didn’t do anything.
Andrew: word of mouth spread.
Swapnil: It just grew, it grew. I really that’s. It.
Andrew: And then what about licensing? How’d you get the rights eventually to be able to make it into a business?
Swapnil: Yeah. So we actually, uh, like, and those were the initial times when a lot of record labels in India didn’t even know how to license their music for the internet. So I remember we had like a lot of conversations with some of the parties in India to actually pay for those licensing costs. And that also made it harder for us to leave Yahoo and just do this full-time because we wanted to.
Uh, we needed money to pay for the licensing cost. So we left Yahoo when we actually raised our first round of funding.
Andrew: And so, wait, were you able to get, you were able to get. The rights to do this because they, the people had the rights we’re just new to this. And they were open to new agreements. Is that it?
Swapnil: Yeah. I mean that time it was a big, I would say novelty to be able to do this over the internet. And luckily at that time there were certain bodies who control licensing for pretty much all the music labels centrally. So we had to talk to one or two people to actually, uh, Get the licensing in place
Andrew: Who’s Rohit, uh, Bettina. Is he one of the co-founders?
Swapnil: it’ll hit back. I was a CEO based startup India that we had data.
Andrew: Ah, okay, got it. Um, and then you added radio to, uh, to kind of like, um, like Pandora, right? And was the money coming in from what? From advertising?
Swapnil: Yes. Advertising and subscriptions.
Andrew: How big did it get before you sold it?
Swapnil: We had around 10 million listeners every single month.
Andrew: Wow. We know advertising all word of mouth,
Swapnil: All word of mouth.
Andrew: mostly in the U S
Swapnil: No. Yeah. I would say 50% of it was, uh, 30% was us. 50% was India and the remaining was around the world.
Andrew: wow. And then how did the sale come about?
Swapnil: We sold it to, uh, ideal. And IDEO was again, uh, started by the founders of Skype. It had, uh, probably one of the best music products, a pretty big cult following, uh, one of the early players in the market based out of San Francisco. And they really wanted to enter the Indian market. And because we had millions of users in India using the Ghana, and we had deep expertise about the Indian marriage and in the Indian market, fascicle pretty amazing fit for them.
Andrew: And the, the site just looks so I don’t even know how to describe what I’m looking at. An old archive of the
Swapnil: Oh,
Andrew: It’s is this you creating this food? Who did the front end on this one? Here’s what it looks like. It kind of looks like,
Swapnil: that’s the, that was the,
Andrew: text search
Swapnil: that was the first, yeah, that was the first or probably the first revision I had quoted this entire UI on the site myself.
Andrew: With the tabs at the top, or you can go to all these remixes pop. This was you designing the whole thing yourself.
Swapnil: yeah, yeah.
Andrew: How does it bring you back to see
Swapnil: Yeah. I know victim, victim, I assume.
Andrew: Fair to say that at the time you didn’t have, at the time of the page that I’m looking at, you didn’t have licensing from music companies. You were just doing it on your own, downloading it yourself, putting it up on your site
Swapnil: No, I think this, this one, we had a license when you were, um, yeah. Yeah.
Andrew: But in the early versions, it was like
Swapnil: Our own music.
Andrew: from CDs that you had,
Swapnil: Yeah,
Andrew: what was Dean Ghana assistant or did Ghana assist?
Swapnil: it was, uh, A certain year that flood. So one of the problems that we had with Indian music was when you’re searching for, uh, Indian lyrics in English, does such a, would always feel, so we’ve been this assist who could assist you in finding the right song.
Andrew: software.
Swapnil: Yeah. Yup.
Andrew: That’s that was you creating it. That’s pretty impressive.
Swapnil: Yup.
Andrew: I’m going down the, uh, the internet rabbit hole of trying to figure out the early days of Tingana and how it worked out. And, um, it looks like you also are doing some partnership or something with Yahoo for a little bit. You were trying to get into content, right.
With some music news. Am I right about that?
Swapnil: Uh, yeah, we did. I think we did have a new, uh, a news product at some point.
Andrew: Right. You’re just trying to see, where does this go? Could this become like a portal for the Indian community eventually sold? How much did you sell for
Swapnil: We didn’t disclose the price on that.
Andrew: ballpark? It we’re talking millions of dollars. Is
Swapnil: Uh, I would say yeah, a few minutes. Yeah.
Andrew: Okay. Life change after that one?
Swapnil: Uh, I would say it was a good outcome, but, uh, the outcome that we had after that was
Andrew: Uh,
Swapnil: sort of bigger.
Andrew: miles ahead. All right. I want to talk about what you’re up to right now. I see a Zinzi is Zenni. Excuse me. Is the brand new product. You, you came up with it because of what what’s the problem that you experienced.
Swapnil: Yeah. So based on our own experience, we know that early-stage companies are always low on resources and high on risk. And, uh, most of them like. When we were running our first startup in the second startup, there was several different finance functions that we need to worry about. And it starts with bookkeeping, bill payments, invoicing, uh, MCL for services taxes.
And for each of them, we had to hire someone to help us do it. And it was all manual human intensive error prone. Each of them used to use several different tools. So, uh, and we had to oversee everything to make sure there are no human mistakes, et cetera. And suddenly it became, uh, extremely fragmented and complicated to manage finances.
And we saw that again and again, with both of both our startups. So we thought that I think we have to fix this problem because this problem is global. Every entrepreneur faces it and that’s fairly nonsense. And with the Navy created the first yap or what finance concierge for startups.
Andrew: But you do you do the full bookkeeping for people? You’ve a team of people. So I, I keep thinking there’s going to be some software that automatically does at least my books. I don’t know why my books are that complicated people pay for monthly membership to it, to access my courses and old interviews, advertisers.
And it’s, it’s about two dozen advertisers a year pay for ads that go within the interviews. I have random expenses, but it’s all software and people were, why is it always a headache? Why can’t I get. Uh, I want to curse so bad. They get so angry at this. All of a sudden, why can’t I know how much money I made as of yesterday.
Why do I have to wait until the month is over? And then one or two weeks for bookkeepers to do it? It’s all fricking Stripe. It’s all PayPal. It’s all basic basic stuff.
Swapnil: Yep. And that’s the problem, right? Andrew, because everyone, most of the people in the accounting world are so used to doing things manually and they don’t even start working on your books until the month ends.
Andrew: True. I, I sit in inDinero. I really like inDinero. They were going to create brand new QuickBooks. They decided instead that what they’re going to do is do actual humans doing the books in QuickBooks or whatever, or even in their software, if you want it. That’s where they pivoted. I used pilot.com.
pilot.com said, all we do is create these scripts and automations because everybody wasn’t online businesses, basically getting data from the same sources. We’re going to vacuum it. I thought great. I could always go in and see how much money I have as it is. Yes. Today, even with them, it takes a week or two.
They’re not, they’re not these old fashioned businesses. There are new businesses in this space that still can’t tell me as of yesterday, how much money I have every entrepreneur faces. I would’ve thought that you wouldn’t have, frankly, with mezze you told our producer headache after headache. You said you would have to scramble with the bookkeeper to get data before board meetings.
You’re right. Amazing. Millions of dollars from investors. You don’t have, you don’t have income coming in from random fricking sources. You still have to suffer. Why what’s the fricking difficulty here?
Swapnil: Yeah. So there are, uh, I would say several problems here and it’s a pretty hard problem. If you look under the hood fosters that you need to integrate with all the different sources that you might be using today, whether it’s Stripe, whether it’s your banks, credit cards, accounting, softwares, payroll providers, team payment.
All those sources, then look at all the data, reconcile, all that data categories, all that data, identify vendors. Look at what your revenue sources are. When are you getting that revenue? When are you supposed to record those revenues? And with each business, these one sustains. That’s my bookkeepers, or even with pilot, they, uh, they all start working on their books.
After the month ends the sec, they look at all these sources, they manually categorize transactions in a lot of cases, they will reconsent stuff. And then they will take two weeks to do all this and then close your books. But we, we knew that this is a problem, and that’s why we are building the first real time bookkeeping platform.
So we get into you daily bookkeeping every single day and offer you real-time insights into your finances every single day.
Andrew: So, and I don’t, this is what we all want. Every single person. If you’ve got a SAS company, if you’ve got recurring revenue, you want to know where are we standing? You have some ballpark already. Just tell me, are we doing a little bit higher, a little bit lower. But w I’m still curious about why others can’t get it done.
And maybe this isn’t a question for you. Maybe it’s a question for me to come back and talk to the founder of a pilot about Stripe and all these other sources are so meticulous about the way they get data out. I’m surprised that nobody else can organize it better.
Swapnil: And the funny thing is that Stripe itself doesn’t integrate with QuickBooks. So whenever you, if you are using QuickBooks, you need to pull it. Yeah. You need to pull data out from Stripe and then push it to quickly.
Andrew: I didn’t realize it. So we do that. And then I get on fricking nuts because apparently each one of these, um, entries is a what’s it called? It creates an invoice in QuickBooks. You know, this.
Swapnil: Yeah.
Andrew: This is a fricking headache, because then I want to know how many invoices are outstanding for my sponsors. And it shows me some random dude who a month ago didn’t pay his bill for $25 a month.
And now that’s showing up and it looks like we have more outstanding invoices, which we don’t. If the guy didn’t want to pay great, cancel it. Don’t invoice him.
Swapnil: Yep. So this is one thing that we solve with Zenni, where we give you a finance constraint. The filing concierge will stay on top of following up with your customers, making sure that cash collections is performed, et cetera. So you don’t fall behind.
Andrew: All right. Um, I want to talk just a little bit more about this first. I forgot. My second sponsor is rippling. Here’s why I switched to rippling. QuickBooks comes with the way to pay your people, right. Then you have to 10 99 them, if they’re contractors, you have to wire some people, if they’re outside of the country and headache, right?
So there’s all these different services. I said, I’m never going to be in the business of 10 99. And again, I have bookkeepers, I have accountants. How does it always come back to me that I have to 10 99, every contractor. And it’s kind of awkward because you have to go back to people from January and say, is this how much you made?
And they say, no, if the adjust it, and then you go back and you send them a 10 90 nines, big fricking headache. Same thing happens when you’ve got contractors in the U S has happens. If you have contractors internationally, if you have employees, there are a whole bunch of other issues that go into it. You want to handle it and you want it handled, right.
That’s where rippling comes in. What do you know about rippling? You, you work with them, right? You integrate, you suck in the data from them and you organize it in people’s books.
Swapnil: Yes. Yeah, we do crippling, uh, and we work with basically all the top providers
Andrew: What do you know about rippling? Do you know anything about them? What makes them special? I could tell you if you don’t.
Swapnil: Yeah. So, uh, I think one of the key things, uh, about our differentiators about rippling is the fact that they also do it management for you. So if you’re onboarding a new employee, they will also provision their laptop, allow you to create accounts for them across all the different services automatically, which is pretty amazing.
Andrew: Yeah, right. Especially now you’ve got people working all over the world. You want to make sure to get a computer then with all the software that you, that they need on it. Or if, or if it’s just buyer, whatever computer they want, um, let them use their own computer. Great. You still want the contractors to all have access to the same set of software and not have a situation where two different people will sign up for the same site.
We have that at our company, three different people sign up to the exact same company, three different accounts. Nobody knows what the login is, cause they it’s not organized. Rippling says you’re onboarding people. Get everything signed, all the contracts, make sure that you know where to send the money, make sure that you tell the system how you plan to pay them.
And then you give them access automatically. When they’re starting to work with you to all the software they need in one place. They just log into rippling to their account and they get to see how much money they made. They get to see where the money is. They get to decide where it goes, bank account, what, whatever.
And they get to see all the software that they have access to. This is beautiful. Listen to me, people I can talk to about rippling a lot. You can go do your own research and see why it’s amazing. If you’re paying contractors, employees. If you’ve got people who are moving from city to city, state to state, even going internationally, and you need to make sure to get them paid and get them the software you need.
You gotta check out rippling. And if you go to rippling.com/mixergy, they will walk you through this whole software and show you why it’s miles ahead. You should just understand it. Even if you’re not ready to switch today, you should know it because at some point you’re going to switch to rippling. I know it.
Um, how do you get customers now for Zenni?
Swapnil: Yeah. So today again, our entire growth is by word of mouth. We are working with more than a hundred startups today. And either our customers are referring other customers. Our, our advisors are referring customers to us. So it’s all by word of mouth.
Andrew: It’s super early though. Right?
Swapnil: I wouldn’t say now it’s super early. I would have said that an ear ago, we have already processed more than $300 million in transactions in date and are on track to process close to a billion dollars in transactions in the next 12 months.
Andrew: say you have now?
Swapnil: More than a hundred.
Andrew: Oh, wow. Okay. The reason I thought it was super early is because when you and I were looking at the site earlier, I just typed in zenni.ai and it went to a broken link. And then once I adjusted it by saying HTTPS, it took me to the site, which looks beautiful.
I’m looking at traffic sources on SEMrush and it looks like traffic didn’t really get going on the site until September 20, 20. Right.
Swapnil: Yeah, we, we have been, we have been operating in the stealth mode for the last, almost a year and a half now.
Andrew: Okay.
Swapnil: We just launched a couple of weeks ago,
Andrew: Where’s your funding from?
Swapnil: we raised a total of $3.5 million. Some our capital led the round and Silicon Valley bank. Mid-single uh, several other funds also participating.
Andrew: None of you are high-profile angel friends. None of your high profile angel friends invested. You have high profile. I feel like you’ve got really good connections in Silicon Valley.
Swapnil: Yeah, we have several angel investors who have invested in Danny, like the CEO of Poshmark Manish. Chandra is an angel investor. Gokul Radom is an angel investor. Amid signal is an angel investor and investors. So we have a lot of lucky to have a lot of amazing people backing us.
Andrew: Why didn’t you? Why, why didn’t you start with just the reporting via chat? Like, say, look, keep your bookkeeper, keep your software. We’re just going to be an add on to QuickBooks. We know you need some reports. You need some data. We’ll just go and get you that. Why did, why did you decide that you wanted to take on the whole bookkeeping and reporting operations?
Swapnil: Oh, I think for the very same reason that you mentioned before all the bookkeepers. Do stuff manually and wake up when the month ends. And we didn’t want that. We wanted to basically make sure that these bookkeepers and accountants can work hand in hand with our technology. And that had to be real time.
So we had to build this entire team from the ground up.
Andrew: And you had to do bookkeepers first, let the bookkeeper strain the software and then eventually software will do almost all of it.
Swapnil: Oh, I don’t think software would do all of it, probably for, uh, some cases. Uh, it would be able to do it, but at the end of the day, I think 70 to 80% of it can be automated, not a hundred percent. You will always need an accountant or bookkeeper or a controller to oversee things, to make sure everything is accurate.
Andrew: And the price is in line with bookkeepers. I hate using individual bookkeepers. What you guys did before at MSCI. I didn’t realize you had a part-time bookkeeper. I hate that. There, the PR person gets flaky, forget flaky. They have a real issue in their lives. And instead of thinking about how are they doing?
Are they sick? Well, they’d be sick for a month. You have to think about, well, they, I do my books. I hate that. I want to have a whole team of people. And so for the price of a single bookkeeper, you essentially, uh, at that price, we’ll do my whole books. I’m thinking of maybe having you guys do like a shadow version of my books.
Because I’m already locked in with my bookkeeping company and I, I like them enough, but man, I’d love to have instant results. really we’ll have it done within 24 hours. I could always go back in and see how much money we made as of yesterday.
Swapnil: Absolutely. We have founders who are checking their finances every single day for a lot of high growth startups.
Andrew: All right. For anyone who wants to go check it out? It’s zenni.ai. That’s Z E N i.ai. Congratulations on, uh, on all your success. I feel like this could be the biggest of all your businesses, but I also think here’s the big problem. I think that you’re going to find, tell me what, tell me if this is something that’s come up again.
I’ve been looking at this space for a long time. Switching costs are a pain in the neck. People can’t like, look at me. I’m connected to my bookkeeping company. It’s really tough to switch. Right.
Swapnil: Yeah. I mean, you will switch if you have pain points and 70% of our customers move to any from a finance firm or a part-time bookkeeper that they see problems with their current setup. And whenever you see that, that set up opportunity for, for you to consider a change.
Andrew: I think the onboarding is I’m surprised you have a hundred so fast that a hundred customers, I think your onboarding is going to be a big challenge. I mean, the, the, I think getting new customers is gonna be a challenge. Um, and you’re probably going to start to think about how do we get people in earlier, earlier.
And maybe you’ll end up doing something like you’ll LLC them, maybe you’ll buy up accounting firms and then just move their people into software. You know, you don’t see yourself doing that.
Swapnil: We definitely don’t see combining accounting folks.
Andrew: What else would you do? I could see that at some point you might offer software that requires less of a commitment to switch, where it would just be, get reporting via QuickBooks. And then we do a little bit, a little bit more. Maybe, maybe you do some basic software that will categorize things better or suck in data from Stripe.
Maybe you do like a Stripe tool, right? It’s really tough to get people to switch bookkeeping companies, but if you could get your toe in the door, Then your whole body will come in afterwards, right? People will want you to take everything on.
Swapnil: Yup.
Andrew: I could see that, or I could see something else. QuickBooks actually, surprisingly, it’s an old company.
A lot of their stuff feels really dated, but they’re very agile kind of like American express in, in recognizing the value, the value that you had, uh, with mezze. I could see that maybe QuickBooks would say, let’s just get Zenni. Let’s just be the, but no, the problem with that they have is they’re already invested in their professionals, right.
That they’ve put money in that they’re promoting on their site. Am I right? They don’t want you to replace all those professionals who are building their livelihoods on QuickBooks.
Swapnil: Yeah, I think there’s a great opportunity for QuickBooks to partner with companies like zany and we already are one of the QuickBooks partners. So if you actually sign up for QuickBooks wise, any, you get 50% off on your monthly subscription. So I think the partnership act,
Andrew: they acquire you, then they’re basically saying, use our bookkeeping company, instead of all these professionals that we’ve created a long relationship with. Right. And brought along, that could be a challenge.
Swapnil: Yeah, that can be interesting. I think, uh, if it, if not, Xeni, I think they can definitely consider acquiring players where they can take QuickBooks to the next level. When it comes to adding key features into their accounting software.
Andrew: Where this has gone, but I fricking I’m glad that you’re doing this. I think that it’s, it’s nuts that we’re all doing the same QuickBooks thing. I think it’s nuts. The people who are automated and everything, how many come, you know what? I bet you. What is it a us Airship? I forget the company that I met in, in, uh, I got to find out, I bet you, the people who are doing robot food delivery companies have human beings who are doing their books.
They will allow a human being to go bring their fricking burger from the store to their house. Right. They want the robot delivery thing, but they have a human being. Who’s doing their books and they’re willing to wait 20, 20 days after the end of the month for their books to be done. This is. This is a great opportunity for you.
Um, I feel like I’m much more excited than you are about this. You’re much calmer and saner about it, and maybe I’m overwhelming you with my enthusiasm for what you’re working on.
Swapnil: No, no, I totally agree. Totally agree. I think this is a huge opportunity though. The marketers global and the problem is seen with pretty much every startup that we have spoken to every entrepreneur that we have spoken to. And, uh, I do foresee people changing to yap out solution sooner than later in the future.
Now there weren’t many alternatives. Uh, so the moment they knew about it, they switched within 30 days. Like sales cycles are super fast, so you’ll be surprised as to how fast people are ready to switch from our traditional finance form to something. Exactly.
Andrew: They trust you with their books right away. They wouldn’t just say, look, do you see what I just did earlier? I said, what if I create a second QuickBooks account? And then I give you access to that. And then they don’t do that. They just say, take over my current QuickBooks.
Swapnil: Yeah, we, we normally don’t work. Uh, in fact, if someone suggests us that we normally don’t recommend that we always like to be the, the, the primary source for all these customers. And now. If you were looking at that first five or 10 customers, it would have been an issue of trust. But now that we have already worked with more than a hundred customers, that is hopefully behind us.
Andrew: All right. I’m glad to see that, that you’re doing this. Um, Um, looking at me, I’m getting back to business mode. I’m very calm. Now. I’m not excited about your business. I think your, I, if you go back and listen to the last 10 years, whenever somebody is willing to take on this problem, I get super psyched.
And then eventually I get disappointed because they fail at it and then they take on it. Don’t get me wrong. They end up doing well. They do boil by their customers, but the, the dream of. Better bookkeeping just done up to the minute is never, it’s just never seems to be achieved. And I feel like you’re, you’re going to get there.
You’ve got the background to do it.
Swapnil: Yep. So I think Andrew, you might have spoken to many players in this field who want to do this, but have you come across a team that has spent the last five years building AI powered platforms
Andrew: The pilot.com team do that. Don’t no disrespect to them. I think they’re
Swapnil: no, I think, yeah, they are a phenomenal team. They are one of the early movers. So hats off to them.
They came from a different background. They are, they have deep technology experience. I think where we differ is the part where we have spent the last four years building an air powered travel as a service platform where travel agents collaborated with machines to build this unique experience. Now we took the blueprint from that and applying it to the finance vertical, where we have finance experts collaborating with our.
For technology to create this unique experience for our startup customers
Andrew: That’s why I think you’ve got, you’ve got the ability to do this where others had failed.
Swapnil: in this cross. Yep. We are confident about executing on this as for plants.
Andrew: there’s no way this is, I would rather a human bringing book, my food, then, then organize my money. I’ll tell you why. Again, I’m spending way too much time. I get to flip that this. The level of problems that entrepreneurs put up with, how many entrepreneurs do you talk to? Who don’t know how much money they have in the bank?
Or if they do, they want to fricking email from someone on their team who will put it together. Cause there’s no software that does it. Didn’t Jason, Calacanis on his podcast. Talk about that, that he needs somebody on his team to tell them how much money do we have in the bank? How much, this, how much that.
Send them via email we’ll Schroeder. The guy’s like a mental patient because he must have really failed badly before starting startups.com or had a bad pair. He talks to me about how every day he needs to know his numbers via email. It’s a human being like an assistant on his team that puts it together for him.
Right. Why does that have to be why? And here’s the problem with it? Human beings always make mistakes, right. And especially when it comes to numbers, they’re bound to make mistakes. And they take time. Why, why? When all my data is all digitized, it’s all in a system already wise, whatever you and I are saying will be transferring by software.
Why is it that, that I can have that done for my finances? All right. Um, I’m really going nuts here. I’m ranting and raving. There’s, there’s no value anymore in the podcast. People should have just tuned out before I went crazy, but this, this needs to be solved and it’s a solvable problem.
Swapnil: Absolutely. We are excited to take this problem head-on and solve it hopefully sooner than later,
Andrew: All right. Thanks so much for being on here. you’re very patient to see me. I’m looking at your faces. I, as I get all wound up and you’re like, This is not the world that I play in, but I, but I like you, Andrew. Uh, uh, and I want to thank the two sponsors and listen, if you want sanity, when it comes to paying your people and giving them the tech that they need in order to work with you.
If you have a team go to rippling.com/cash Mixergy, talk to them, I’m telling you they will blow your minds with what’s possible. Do your research. You’ll see how great they are. And if you do the website hosted, go to hostgator.com/mixergy.