Full Transcript

Apoorv Shankar (Aina): So we had a prototype of that that didn't go in the hands of the customer. And at the same time, Ultrahuman wanted to build their own smart ring for sleep tracking. So we decided to join hands with them. Although between the, you know, when COVID happened and the supply chain of manufacturers was sort of stuck in India, we had built a small product called HandKey. And this was a, I think that was the first product which I had made along with my team, which actually shipped to customers. And it was a small brass key-like device, but it was smartly designed in such a way that it had a pocket or a cap which self-closes. And it was a, you know, a tiny smart solution for, you know, ensuring that we don't touch public surfaces and, you know, get affected with COVID as much. So that was one. That was the first small, cool product that we had done. We were featured on BBC for that too.

Dhruv Sharma: How many different iterations or prototypes does it take to, you know, ship the V1 product into the hands of consumers of Purvan hardware?

Apoorv Shankar (Aina): Yeah, it's hard to count. Just lots of them, right? Yeah. It's just lots of them. Lots of them. It doesn't, with experience, of course, you get to a place where you know that it's going to take maybe 10 or 15. But I think when you get started for the first time as a hardware founder, I think it's just, there's no count to it. And I think Dyson had mentioned once that they had to do close to 2000-odd versions, 2000 to 4000-odd versions of the product before they launched their first vacuum cleaner. Yeah. But of course, not everyone has the money to take so many iterations, to make so many iterations, but it's fun. I think hardware, I'm seeing a lot of cool hardware companies being built out of India in exciting times.

Dhruv Sharma: Would you say that, you know, the time spent between iterations, between versions, does it increase as, you know, you go from the first prototype to more advanced prototypes? Does it reduce?

Apoorv Shankar (Aina): I think if you have a journey of 0 to 100, let's say, from a percentage point of view, I think you'll do a lot of iterations from 0 to 30. And then 30 to 80, 85 is doable. You know the right part, you'll get there. And then 85 to 95 or 85 to 100, again, takes a lot of effort. That's around small tolerances, small parts and fitments not working out together when you're integrating the whole product together. So I think the initial bit and the last bit is the hardest, you know, when you have to take a product from idea to production. Prototyping, I'd say, is once you've logged in on the form factor, prototyping is comparatively easier.

Dhruv Sharma: Got it. I think we're going to spend most of our time together today talking about what you're doing now. But before we get there, what were the years at Ultrahuman like?

Apoorv Shankar (Aina): Oh, I loved the experience. I think Mohit and Vasal both were, like, had this belief that if you're serious about software, you have to build your own hardware. And I think we aligned well on how to consume our product. India is in the premium category, focused on design and R&D. And I think I loved my time at Ultrahuman. We shipped, I think at our peak, we were shipping close to 50,000 units a month from these small factories in Bangalore. And I think not a lot of people know that Ultrahuman had a big factory here in Bangalore and the rings were being manufactured up here. Was a humongous task. And I think we did contribute to a decent extent to the manufacturing system for startups here in Bangalore.

Dhruv Sharma: Was your office located in the factory?

Apoorv Shankar (Aina): Yeah. So when we were at Ultrahuman, the whole hardware team used to sit at the factory.

Dhruv Sharma: Yeah. Fantastic. And what is happening now, Apoorv? You're fresh, you're out with, you've just announced a recent raise. Tell us more about that.

Apoorv Shankar (Aina): Sure. So for the last, I think, 12 months, we've been working on a project, which is more of a human-computer interaction lab. We've been experimenting on form factors, on understanding what, how software or AI is going to look like in the next few years. And figuring what form factor works out the best to become this next big interface after phone touch screens. Because it's very evident that phone touch screens and keyboards are not the best form factors for this age of AI, wherein everything is changing so fast. And AI has context of everything you need. Soon we'll be at a stage wherein AI systems would know what you want. And they're going to just keep sharing options. You're going to say yes and no to hundreds of things a day, instead of you browsing and searching for everything. And we're already seeing that happen to a large extent in some day-to-day activities. I think we're going to have hundreds of yes and no going forward. And we are building this interface for that generation. We haven't announced the form factor yet. We are starting a private pilot soon. So that's happening. But throughout this journey, the last 12 months, we had showcased a few concepts that we wanted to validate, use cases around, and seeing what kind of use cases, what kind of technology, context-aware shortcuts, keyboard shortcut replacements, things like that, what works out well. We had showcased some concepts at CES earlier this year. And then we had shipped a product called Dune, which is a three-key macro pad which sticks to your Mac. It is a context-aware keypad which changes the meanings of the shortcuts based on which app you're working on. And we had shipped 500 units of those and spoke to a bunch of early adopters trying to figure out what works for them, what doesn't work for them. And I think all that has been just homework to get to where we are at today. And now we're focusing entirely on this device.

Dhruv Sharma: Fantastic. I'm just curious, is there a product hunt for hardware?

Apoorv Shankar (Aina): I'd say Kickstarter is more like that product hunt for hardware, wherein people can pre-order devices. And even the first Ultrahuman Ring we had launched on Kickstarter alongside the website. So Kickstarter, I'd call that place.

Dhruv Sharma: You know, Apoorv, while you're here, I think we definitely want to get a quick 101 out of you on just human-computer interaction, right? Humans and computers have been interacting for decades now. So how has the nature of that interaction evolved over the decades?

Apoorv Shankar (Aina): Sure. I think we'll probably start with keyboards, because before keyboards, I don't think there was a lot of development that was happening in that sector, or we'd remember much of that. But I think keyboards is where we started, where we were able to start talking to machines in a way. And then the mouse came, which paved way for a graphical user interface. I think you wouldn't have seen Windows, you know, having icons and a person interacting with a monitor if the mouse was not there. And I think we are operating in a similar way, honestly, at this point in time. I think we are waiting for a new hardware interface, which is designed for the age of AI. And then we'll see what all software can actually do. Because imagine if we did not have a mouse, we would still be typing on a keyboard line by line, right? And the use cases of other computers would have been way limited.

Dhruv Sharma: I think, of course, yeah, so I think the mouse was invented at Xerox PARC, if I'm not mistaken. Yes, yes, yes. An early, yeah, wow.

Apoorv Shankar (Aina): Yeah. And Apple was the one like behind that launch, and they acquired, I think, the IP around the mouse and made it into a mass consumer product. So that's there. Honestly, Apple has been behind the two most successful hardware interfaces out there, the iPhone and the mouse. But yeah, it doesn't look like the way Apple is operating. It's not like as fast and innovative as it used to be. And I would assume that the next big interface has to come either from a startup or any other company who's just been operating or has a DNA in this space.

Dhruv Sharma: I think a lot of people are going to keep an eye out for that because Apple also gets the new CEO, I think, in about a month's time.

Apoorv Shankar (Aina): Oh, yeah.

Dhruv Sharma: But speaking of iPhones, Apoorv, what's going to be the next big iPhone moment for hardware enthusiasts around the world?

Apoorv Shankar (Aina): I'm hoping it's ours. Yeah. We've also seen, of course, a few companies try to be in this space, claiming that they could potentially be the next iPhone. Humane was one, Rabbit was one, OpenAI is building its own hardware, and there's a lot of buzz around what's next and what that form factor looks like. We are of the opinion that there are two major parts or major technologies that are going to be defining what that form factor is going to look like. One is a major change is that we are using voice a lot more in interacting with computers. So that is definitely going to be there as a primary focus. In addition to that, I think devices would have to know what you are doing. They have to be actively collecting context and acting on your behalf so that you don't have to search and think of what you might need next. For example, if you're in a meeting, I mean, the device should know that I'm not supposed to show you like 10 of the notifications which are all random. The device should know that now it's time for me to transform into more like a meeting control for this person because that's what the person needs. And so in a way, voice plus context awareness is what is how we branded software going forward. And we are designing a form factor which is optimized for both.

Dhruv Sharma: Fantastic. You mentioned Humane, which got, I think the most charitable thing to say is they got mixed reviews and reception. But look, hardware is tough. You never know when a PhD is going to wake up on the wrong side of the bed and give you a bad review and there you go. But you, I'm sure, have seen all of these successes and failures in a different manner than most of us. So tell us some lessons you've drawn from things people have gotten right, gotten wrong. And how have you carried those lessons into what you're doing now?

Apoorv Shankar (Aina): I think one important thing is setting expectations as well, right? I think one problem which, I'm not sure if they directly communicate this or not, but Humane had this problem where people were comparing it with the iPhone. And it was designed, or at least it was portrayed as a replacement of the phone itself, which I think is just an extremely hard problem to tackle. I think that expectation setting itself is wrong. Because when your phones came in, it's not that your laptops were thrown away. You still have your laptops for different things. And the keyboards work for a lot of our daily tasks, but you still have the phone. So I don't think new interfaces completely replace the older interfaces. Mouse is still there. It's just that the use cases keep changing. You want to do something with your smartphone, something with a laptop. And you might not entirely be dependent on your devices, with different devices for different things. So I think that's what changes. You should think a lot of testing before you go about launching a product like this is important. And we've been sort of building on that approach. That's why we are doing a private pilot right now. We want to work with early adopters, trying to figure out what are the top two use cases where that makes sense, where you are not before going about launching. Once you've launched, it's a hand or incident of all kinds of customers, people who may not have any patience if the product does not function in the beginning. I think you have to have enough user testing done before you go about launching a product. It's a very, very hard problem to solve. And it can't happen. It can't happen without the early adopters feedback.

Dhruv Sharma: And if something goes wrong in software, you just restore to the previous version, what are you going to do in hardware? You have no option but to recall devices. That must be like the biggest nightmare ever for anyone who works in that domain. Also, take us with you for only as much time as we have, Apoorv, into the design process of these devices. How do you make those minor trade-offs? Because at the end of the day, the device has to be a certain size, a certain weight, certain dimensions. You have to fit all of the hardware into that. So talk to us about the hard technical choices and trade-offs you guys also make along the way.

Apoorv Shankar (Aina): Sure. I think generally, every product, when it starts, when the design of a new product starts, surprisingly, since I've been a design student, I know that there have been like frameworks designed in the last century of mechanical innovations, which help us think through what is feasible, what is not feasible, what are the top few things which a user would need, what is a demand, what is an expectation, what are things which are good to have. There are frameworks which are already put into place. It's surprising that not a lot of people really know about it. And I think design in a way that way is a little underrated as something to learn. It does help you in just getting to a potential right solution before even you start building anything. So I think those are a few things which people miss out on generally when they go about building stuff. On further, I think on how do we go about finalizing the form factor processes. I think there are again frameworks like people tend to start with anthropometric data of the human body because you generally design everything for the human body. So there are researches and existing data on, let's say, for example, what is the average height of a person in the US or in India. And based on that, you design, for example, automotive standards accordingly. So there are a bunch of these things you take into, you factor in before you go about choosing what form factor is going to be, what is the size of the product, etc. Then you think about just ergonomic data on whether a human's, you know, range for something. And I'm quoting automotive because I think we have very, very high standards of automotive design. Back then in the last century when we started designing automotive products, I think a lot of design process went into doing that to make it safe and friendly. And I'd say Europe has been like behind all of those standards. So it's a long process. When you start with anthropometric data, then you look at human ergonomics, and then you think about what is feasible, what technology is out there, what is not out there. And then you think about what are the use cases we prioritize versus we'll discard based on the minimum viable product that we can ship out, you know, as at the earliest. So a bunch of these things. But again, keep going on and on and on in design processes. It's something that people don't talk about enough.

Dhruv Sharma: Yeah. And Apoorv, I mean, with this, you've gotten the raise done. And tell us how you're building the team. Tell us what's going to happen at Aina over the next three to six months. And then maybe as a parting shot for any hardware enthusiasts, wearable enthusiasts, you know, listening to this, give them some recommendations that they don't know about already.

Apoorv Shankar (Aina): Sure. To start with, we are starting the pilot now of the product that's being installed. Generally, we are prioritizing access for people who are experimenting in AI themselves. They are using all these AI systems tools that keep coming out and are building their own AI agents. It's easier to get feedback from them. They are also builders in a way, understand where we are coming from too. Helps us identify the top use cases, identify where we are going wrong. And before we go about launching, we just have the right story in place. So that is where we are at. Three to six months is going to take just in finalizing the product before launch. We already have a large team here in Bangalore. We're still hiring primarily now more in SF as we think of GTN. But that's primarily it. I think we have a great hardware design, mechanical team already. And I think the software guys are also enthusiastic about what we've been doing. It's just that people have not seen enough work of them outside yet. But it's going to come soon. But we focus a lot more on design and aesthetics. And you'll see when the product comes out. So on hardware, folks who are building in hardware, I'd say that we do have a lot of people who have built in the last decade. People who have been building companies in the hardware ecosystem in India. Talk to them, reach out to them. They have a lot of gyan and connects which will be very valuable. And all of these are nice people. I don't think people reach out and ask for help enough. And you know, end up getting stuck with vendors who not grant. You know, you spend the next six months in just figuring out whether this can happen or not. Can you hear me?

Dhruv Sharma: I think we lost you for a little bit. But we heard you. All right, Apoorv.

Apoorv Shankar (Aina): Okay, cool. I was just saying that people just need to ask more questions.

Dhruv Sharma: Fantastic. Apoorv, thank you so much for coming and spending time with us, telling us what's happening at Aina. Whenever you can open a public waitlist, we'll be more than happy to share it with the TON community. And all the best to you. And we hope to have you back on the show to announce new updates from Aina. All the best.

Apoorv Shankar (Aina): Thank you, Dhruv.

Dhruv Sharma: Cheers. All right, listeners. That was Apoorv Shankar of Aina. Our next guest on the show, our second and final guest today is Abhinav Sharma. Abhinav, hi. Welcome to the Offline Network. Abhinav is building a company called hisabkitab. And I love companies that have, you know, that have self explanatory names, because it's abundantly clear from the get go, and what the company might be doing.

Abhinav Sharma (hisabkitab): Yeah. So yeah, yeah. So hisabkitab is basically an AI accounting software, as the name suggests, obviously. So we are trying to solve this automation problem into this accounting domain, which is right now a very bit manual. People are doing softwares, working on softwares. But still, it's very redundant right now, what all things are available. And being a CA from that background, it gives us an edge to understand what actually people needed.

Dhruv Sharma: Abhinav, in fact, let's just start there. As a practicing CA, give us a sense of, for people who are not chartered accountants, give us a sense of the CA's software stack and, you know, the different point tools they use just to do their work on an everyday basis.

Abhinav Sharma (hisabkitab): To give you an idea, like for basic accounting, every CA uses Tally. So that has been there since 1987. Then there are other tools for GST filings, like ClearTax, Serral, and all those tools. Then when you want to automate some of the process of your accounting, you add layers over Tally. You add, so with Voucherit and other players, which are there, which only works as a stack to your Tally. So you have to integrate those tools over it. And if at all you want to use some, let's say, a mobile app, then you need to purchase another software called Credflow or other things, LifeKeeping or something like that. So there are like, maybe like four to five apps, which a CA usually uses. And this is only with regards to like basic compliance and all those. If you want to go into audit, then there is an altogether different software. ITR filing, there is another software. So it gives around like six to seven apps. So our idea is basically to replace all those things into a single bundle. Right now, we are able to do it with four or five apps. And we are working on audit and ITR, which will soon come out.

Dhruv Sharma: All right. I'm going to, in fact, now that you bring this up, I'm getting curious, like, what is the overall size of the accounting industry in India, beyond the big four, big six? Like, how many registered CAs do we have at ICAI? And how many smaller regional firms and, you know, some of the homegrown firms? Ball pockets.

Abhinav Sharma (hisabkitab): Yeah. So if we talk about, there are like six lakh CAs in India, after those 1.5 lakh are into practice, basically signing work, ITR work, and all those things. So there are 1.5 lakh CAs into this. And if we talk about accounting and as an industry, so like around five years back, India was bought busy. At that time, they valued the industry at 5,000 crores. So since then, there has been no standard publication into the market. But presuming how the market has moved, it is right now at around 8,000 crores or something like that.

Dhruv Sharma: Got it. And we'll get, we'll dive into hisabkitab in just a bit. But in a general sense, how's the CA community adopting AI in their professional lives right now?

Abhinav Sharma (hisabkitab): We are behind very just. So the ICAI trying to bring in more courses and more things, because the general tendency has been that CAs are not much tech savvy. But that thing has been changing since the last one year, all of us, like I have realized that we were falling behind, all of us have realized that we were falling behind. So every now and then, people are doing something or the other thing. And slowly that adoption has been there. And people have been integrating AI into their regular workflows, what all things they are doing. So they're trying to automate a bit of the stuff using annotations and other workflows. And so yeah, right now, the adoption is coming, but it's in very early stage to be very honest. So it's a good point of time for any person to invent something into AI for CAs.

Dhruv Sharma: Yeah, I think one of the other views that people have in general on this is that LLMs, for instance, are great at giving you probabilistic outcomes, not deterministic outcomes, or outputs rather, because if three people were to prompt in this way, they're likely to get different answers. And in the profession of accounting, most answers are a whole number. You don't need a subjective text answer.

Abhinav Sharma (hisabkitab): Yeah, the input is defined and the output is defined. So you can't just hallucinate there. That word can't exist in this industry.

Dhruv Sharma: But at the same time, AI is also great at several other use cases, for instance, retrieval, you know, text-heavy files and hundreds of pages. So talk to us a little bit about the use cases that you're also solving at hisabkitab.

Abhinav Sharma (hisabkitab): So right now, if you want to do a bank statement accounting entry, so let's say there is a one-pager accounting bank statement. So if you'd want to do an accounting entry, you need to do it one by one. You need to select the debit account, the credit account, and that's vice versa, whatever it is. So with the help of AI, we have done bank statements get scanned with the help of OCR. And at the back end, our LLM defines that where this should go on the debit side or the credit side. So if I'm measuring in the time, if a 10-page bank statement earlier used to take two hours, right now it only takes 15 minutes. And the more you use it, the more the rules get set, the more the system gets developed, and the more it realized that where, what are your general patterns are, where you are debiting, where you are crediting, and all those stuff. And the same thing we have implied in purchase and sales also. So what you can do is you can click a photo of your purchase bill and the accounting entry gets done. You can click a photo of the sales bill. If suppose you are generating it from some other software, if you are getting a sales bill directly, and then you are booking it at the back of your accounts, all you need to do is you need to click photos and a screen appears. All you need to do is approve it and the accounting entry gets done. And we have integrated, and the next step which we are trying to do is we have integrated our MCP with Claude. So you can talk to your accounting data directly like how you chat over it.

Dhruv Sharma: Fantastic. And what's the adoption like for the MCP integration you have with Claude?

Abhinav Sharma (hisabkitab): It's in very early stages. So we have only made it live to our paid users and we are still discussing a lot of things because a new act is going to come, which is data protection and privacy laws. So right now we are in talks with our ICAI that what all things will be allowed for us to expose to Claude and not. So anyways, it is going to be the Claude of you, not ours. So we are discussing on that part with our ICAI. And once that thing gets cleared, then we'll be making it live for a huge among all the people and for outside public.

Dhruv Sharma: Yeah, we're getting one audience question, Abhinav, which is, I'm not gonna read the entire question, but it basically has to do with the relevance of the profession itself in the age of AI. So I think we just finished tax filing season. It's literally 3rd of August and this time around a lot of people just imported all of their files into an AI tool of their choice and got the draft competitions, et cetera. But people don't realize how complex accounting can typically get for businesses and other entities. So as someone who's not only a trained and practicing chartered accountant, but now also very close to AI, what do you think about the future of the profession?

Abhinav Sharma (hisabkitab): See, I would say, as you only mentioned a bit earlier, that AI gives you a probabilistic or something that kind of answer. It doesn't give you an exact answer that this is how things are supposed to be done. It has been a case with a lot of times in our GST practices also that people are drafting notices reply via this AI tools. And a lot of time it giving references of the case which doesn't even exist as of now. I understand that it will gradually it will improve a lot of time, but at the end of time, a human will be there. So I think that human in the loop will always be there. So it will run hands-on. We can run together, not entirely replacing or something that kind of.

Dhruv Sharma: This is real guys. I mean, of the many things AI is good at, of the many things AI is good at that it's actually not supposed to be good at is if you actually use this helpful litigation it will just manufacture precedents that don't in fact exist, which can get you into trouble. Great. Abhinav, you brought up Tally and I think there's this internal question that it's really hard to, and the switching costs really high to go away from Tally. So anyone who's coming and offering either replacement system or even a companion system has to think about making migration super smooth. How would you guys think about that?

Abhinav Sharma (hisabkitab): So we have made a Tally connector for the very specific purpose because in early stages of like five to six months, everyone was like, yes, your tool is very good. Everything is very good. But what about my previous data which I had been doing from these many years in Tally? As per income tax also, you are supposed to keep your eight years of data with you because at any point of time, you can get a raid or anything. So you need to have those data with you. So within six months, we built a Tally connector. So right now what happens is within 15 minutes, you can bring your all the data from Tally to hisabkitab with all the masters, ledgers, entries and everything. And at the same point of time, some people were scared to scared to like 100% shift to our platform. So the Tally connector works in other way also. So you can send the data back to Tally as well. So you can, what you can do is you can bring your old data here. Right now you are doing automation here with the tool, with all the things which we are providing. And then you can send that data back to Tally as a backup to you as well.

Dhruv Sharma: And the profession of accounting loves backups in general. Backups and backups of backups.

Abhinav Sharma (hisabkitab): Yeah, so that solves the thing that people had a big hindrance about that what about our previous data? And once they get the taste of automation, which is actually happening right now, because the friction has been from the accountants community more and less from the business owners. So the accountants were more scared about losing their jobs about from the automation and all such things. And so the resistance was a bit more from that. So right now from all our paid customers, 50% are the ones who have shifted from Tally itself.

Dhruv Sharma: Yeah, and you also brought up DPDP. I've been having a lot of people and not just in professional accounting, but in general in financial services. We're using agents to do financial services work. I've also built like a, you know, double entry bookkeeping ledger underneath their applications. Have you guys also taken that approach?

Abhinav Sharma (hisabkitab): No, not as of now, but we are looking to integrate directly. The next thing which we are thinking is bringing in treasury into us. So if we are having, people are having bank balances, then a prompt can go to those guys that you can invest this much sum into that much amount or something like that. But that thing, it's still in the pipeline. We are discussing about it. But yeah, this specific compliance is what I want to get clarity upon. So that's why a regular meeting is happening with ICAI and other bodies that what all data is allowed to be shared and what is not allowed to be shared.

Dhruv Sharma: And in your estimate, like which domain within, again, the profession, is AI likely to impact the most? Is it going to come over it, assurance, litigation?

Abhinav Sharma (hisabkitab): I believe accounting is something which is going to be affected a lot because it's anything which can be done, which is being done on a repeated basis with a similar flow is what the expertise of AI lies in. That you can define things that this is how it should be done. Because it's a universal accounting as a concept is the universal thing. So accounting can be replaced, not like 100%, but at least 80% to 85% can be done. A maker-checker process will always be there. But assurance, audit, consulting, those things are always going to be there.

Dhruv Sharma: Yeah, they're always going to be there. They also serve just a very, very different purpose. And yeah, I think one of the other unique advantages or opportunities that AI presents again for accounting specifically is you will not have to wait till end of period to reconcile. You can pretty much reconcile in real time if you just have the right systems for it. You can hourly reconciliation or even on a per transaction basis.

Abhinav Sharma (hisabkitab): Exactly.

Dhruv Sharma: Fantastic. So tell us a little bit about the scale of the business, Abhinav, and how you're thinking about growth. Tell us more about the team. Tell us where you're actually based and all those things.

Abhinav Sharma (hisabkitab): So we are based out of Surat, Gujarat. And the journey has been around like for the initial four years, I was into a regular CA practice. And then we went into building a product and it took us around 2 to 2.5 years and building a final MVP, which could be utilized. And then we are into this journey since like last Jan 2025. And this year in Jan, we were able to close the round, but the PR came late when the funds arrived. And so about the numbers, so we have like around presently 3000 paying customers with us. Last month, we did a revenue of around 27 lakh rupees. So we are trajectory on a month on month, like 20% growth every month. And if I talk about team size, so we have around like 40, 45 people, including tech sales support and our customer success and all those various kind of roles, which are usually tech company has. So that's the team formation as of now. And what we are looking at is basically to reach a 1 crore MRR by the month of March 2027.

Dhruv Sharma: And can you also talk to us about, you know, like the basic philosophy, is it going to be like AI prepares and humans review or is it going to be, what's it going to be? Like, let me not put words in your mouth.

Abhinav Sharma (hisabkitab): The way you mentioned, obviously, it is always going to be there like your AI companion. It's not going to be a replacement or something like that. So maybe like a concierge or something, that kind of thing, you can term it with your accountant, concierge, and also not replacing entirely accountants because those are the community which we are also tapping on. Chartered accountants are the best, our best use cases and our best promoters as well.

Dhruv Sharma: So Tally famously found a way to build an entire ecosystem around it, right, with the training centers and, you know, everything.

Abhinav Sharma (hisabkitab): So their strategy has been all about referral partners and channel partners. So all they do is they give referral commissions to the people who install and then those guys handle the support as well. So the big thing Tally was able to crack was support because in any tech product, support is a very big thing. So all they did is they hire regional partners who own on the first installation of the software. And then if you want to build anything over Tally, those guys used to provide support and then they used to earn additionally over it. So the distributor network of the Tally is something which helped them build what they are today.

Dhruv Sharma: Plus it became like product plus service revenue.

Abhinav Sharma (hisabkitab): Product plus service revenue for those people.

Dhruv Sharma: And speaking of product revenue, Abhinav, like what are the price points that you're offering hisabkitab at?

Abhinav Sharma (hisabkitab): Yeah, so our pricing, there are three plans, like 3000, 5000, 7000. Those are the annual plans I'm talking about. So in 7000, we provide all the AI things. In 5000, it's basic plus eBay billing invoices. That's mandatory as per compliances on all those things you have to do. 3000 is the basic entrant plan for doing your basic accounting and everything. And if you want to do add-on or you want to do some e-commerce marketplace, if you want to build cost centers, if you want to have more AI tokens and all those things. So that goes above the 7000 plans with the certain add-ons into it.

Dhruv Sharma: And how do you think about customer segmentation? Is it by, you know, is it firms by headcount? Is it something else?

Abhinav Sharma (hisabkitab): So segmentation is around like anyone who is in a business of from around like 50 lakhs to like 5 crore. Those are the target audience we are looking at. Beyond 5 crore, people move to ERP or some other software. And 50 lakhs is about 50 lakhs is something which we are able to afford is because there are a lot of billing software in the market who are like a thousand bucks on an annual basis.

Dhruv Sharma: Fantastic. Any closing thoughts, Abhinav?

Abhinav Sharma (hisabkitab): Just a thing which I have personally learned like being from a non-tech background and building a tech product. So that's the thing which people are usually scared of. I meet a lot of people. So I think you can learn all those skills over the time. And you should invest at least like first invest in yourself for the first few five to six months learning at least the basic tech so that you can have an idea about all the things. And this AI thing helps a lot in this thing that you can check what all is correct or not. You can have a second opinion on thoughts of other people whom you are working with. It may be right. It may be wrong. But at least it will open up some discussions for you. So with the help of AI, I believe that everyone can obviously right now try to be a tech founder.

Dhruv Sharma: Fantastic. Wise words. Again, Abhinav, thank you so much for coming and sharing what you're building at hisabkitab. All the very best to you.

Abhinav Sharma (hisabkitab): Thank you. Thank you so much to you.

Dhruv Sharma: Thank you so much. Thank you. All right, listeners, that is the end of TON120. See you for number 121 on Wednesday. Have a great rest of the day.