Raghu Ravinutala, Yellow.ai: Bootstrapping to $1M ARR and Scaling Globally

1Mby1M Research · 1Mby1M Case Study

Interviewer: Sramana Mitra

This interview is part of the case study-based research and entrepreneurship education work of 1Mby1M, the global virtual accelerator founded by Sramana Mitra.

Abstract

Raghu Ravinutala bootstrapped Yellow.ai to $1M ARR over roughly 2.5 to 3 years, staying profitable from year one before raising any institutional capital. The company grew a 22-person team on the strength of an early Microsoft/Azure partnership and 35 to 40 reference customers, only turning to venture funding once it needed capital to expand geographically. This staged, proof-first approach to fundraising – bootstrap to revenue and repeatability, then raise deliberately for expansion – offers a real-world case for the capital-efficient, equity-preserving scaling logic central to the 1Mby1M methodology.

About Raghu Ravinutala

Raghu Ravinutala is the Co-Founder and CEO of Yellow.ai, which enables enterprises to drive automation across customer experience and employee experience by integrating enterprise data into conversational AI for sales, marketing, HR, and IT automation. Prior to founding Yellow.ai in 2016, he spent roughly 15 years in the semiconductor and EDA industry. He built the company from Bangalore, India, growing it into a Delaware C-Corp with customers across 15 countries.

Interview

Sramana Mitra: Let’s start by introducing our audience to yourself as well as the genesis of Yellow.ai.
Raghu Ravinutala: I’m the Co-Founder and CEO of Yellow.ai. Yellow.ai enables enterprises to drive automation on their customer experiences and employee experience by integrating a whole set of enterprise data and delivering phenomenal experiences that the companies can leverage for sales, marketing, HR, and IT automation.

I’ve been in the semiconductor and EDA industry for about 15 years before starting the company in 2016. The genesis has been serendipitous. I was chatting with my friends over a break. We were talking about the Facebook acquisition of WhatsApp. One of my friends commented that “I think WhatsApp is going to monetize conversations with businesses.” I could relate to it quite strongly because I could sense the frustrations that I used to have in talking to enterprises.

I launched a mobile app where consumers can chat with businesses. I did that with my friend Kishore who used to work for Flipkart. That got traction. We had to automate it, because we didn’t have a lot of employees. That was the genesis of the story. One thing led to another and, now, we’re automating more than a billion conversations every single quarter.

Sramana Mitra: Where are you located?
Raghu Ravinutala: I’ve always been located in India and Bangalore. Now I’m making the shift to the Valley.

Sramana Mitra: What about the company? Is this a Bangalore-based company?
Raghu Ravinutala: The company is a Delaware C-Corp. It’s a US-incorporated company with customers across 15 countries. In the US, we have 40 people.

Sramana Mitra: In Silicon Valley?
Raghu Ravinutala: In Silicon Valley, about 15. The rest are spread across different states and cities.

Sramana Mitra: You’re doing a virtual company?
Raghu Ravinutala: Yes.

Sramana Mitra: Do you have an operation in India?
Raghu Ravinutala: We have a very large operation in India. We have about 600 plus people in India. A large part of our development and support is in India.

Sramana Mitra: That’s Bangalore?
Raghu Ravinutala: Yes.

Sramana Mitra: Going back to the genesis, you and your friend started the company. How did you get it going? Was it bootstrapped?
Raghu Ravinutala: We bootstrapped till our first million in ARR. In 2016 just because we started on the software and built some automation when Facebook opened up their Messenger platform, we were one of the companies who had some kind of platform ready to automate conversations on Messenger. We initially partnered with Microsoft. We were on Azure.

We got the initial set of 5 to 10 customers working with Microsoft. Most of them are still our customers even after six years. That helped us prove the entire concept of automation on top of their websites. These were all large enterprises with large consumer volumes. They also helped us generate that initial revenue. We were profitable from year one.

Sramana Mitra: Microsoft helped you get to those customers?
Raghu Ravinutala: Yes. There was this global launch of Messenger. There was enough demand for customers to explore automated chatbots. Microsoft didn’t have a lot of presence during that time. The first partner they wanted to introduce was someone who is hosted on Azure and who has some ready-made technology that we can showcase. That helped us get to the initial set of five to 10 customers.

Sramana Mitra: You got to a million ARR on the strength of those first 10 customers?
Raghu Ravinutala: No, once we got the 10, we got more. When we were at $1 million, we had 35 to 40 customers.

Sramana Mitra: You didn’t have to raise money until that point?
Raghu Ravinutala: Correct.

Sramana Mitra: How many people were you then?
Raghu Ravinutala: We were about 22.

Sramana Mitra: SaaS right?
Raghu Ravinutala: We were primarily doing a consumption-based business model. That was not that popular in the 2016 to 2018 timeframe. There were a lot of questions about the treatment of consumption-based overages. We were primarily SaaS-based with a consumption-based coverage model.

Sramana Mitra: The last question I have on that phase is how long it took you to navigate that zero to $1 million period. Microsoft was a big part of that. You were able to use those reference accounts within each industry segment. How long was that?
Raghu Ravinutala: It was about 2.5 to 3 years between the first dollar and a million in ARR.

Sramana Mitra: Then what did you do?
Raghu Ravinutala: We were interested in expanding. A lot of that ARR came from India and a little bit of Asia Pacific. We clearly saw that there was an opportunity to expand the company geographically. That’s when new platforms like WhatsApp opened up. That’s when we felt that this was a venture-scale business.

We got in touch with Lightspeed at that time. There was interest from them to be part of that journey. We raised our first institutional investor from Lightspeed India in 2018 to primarily build out the sales force. It was founder-led sales at that time. Our Series A helped us with that. That led to the expansion into Indonesia, Singapore, and Malaysia.

During that process, something interesting happened. We came across these global capability centers. We were partnering with Accenture and a few others. They introduced us to a few companies. What we saw there was we were primarily a CX automation company. Quite a few of them were interested in using the same technology for employee service automation. Roche Pharma wanted to try us out for IT support operation.

We saw some green shoots in the US and European markets with Fortune 500 companies. We were able to expand beyond a few different use cases. They became reasonably large customers. That told us that these markets were not just messaging-first markets. It has a larger scope in Western markets. Of course, these initial wins can be green shoots, but to really go and expand into US and Europe, our Series A money would not be sufficient.

Sramana Mitra: How much was Series A?
Raghu Ravinutala: $4 million. We thought of raising Series B from the Valley. Lightspeed had a good presence in the Valley. Lightspeed US was interested in investing in our Series B. A year after our Series A, we raised our Series B to help us start in the US. That’s when COVID struck. Our plans were a little bit delayed. We raised the funds, but we were still evaluating how COVID will impact the company.

We saw a lot of acceleration in the emerging markets. In 2020, we doubled down on emerging markets. At the start of 2021, we started placing some bets in the US market. We saw some good traction. That led to our Series C. We were growing 2.5 to 3x every year during those two years. Once we raised our Series C, we accelerated our US growth.

Sramana Mitra: What are the numbers on Series B and Series C?
Raghu Ravinutala: In Series B, we raised about $20 million. In Series C, we raised about $79 million.

Sramana Mitra: That’s a lot of money that you’ve raised. You’d have to build a huge company to return that investment. One of the things I’m hearing from a lot of enterprise startups that sell to the enterprises is that it has become easier to sell to the international market because buyers have become used to evaluating and buying online. Are you experiencing that as well?
Raghu Ravinutala: What we have seen in Asian markets is that with the COVID-related changes, a lot of our large enterprise deals are happening online. A lot of deals in North America also happen online, but they happen online with local field sales. We didn’t have as much success with remote teams selling into the North American markets for the enterprise.

At this time, SMB and commercial is not a market we’re able to find a sweet spot in. That is the nuance. We still went ahead and invested in local marketing in the North American market. That shifted the need significantly on how fast we were able to generate traction.

Sramana Mitra: Let’s talk about positioning. You didn’t come into a virgin market. There are quite a few automated self-service customer support companies. Talk about the space and how you differentiate. What scenarios and use cases are you winning? Where do you have an unfair advantage?
Raghu Ravinutala: This space, from the outside, looks competitive. It is competitive, but it looks even more cluttered from the outside. Having a nice positioning on why you’re relevant to a customer is super important. We defined that along three axes.

One is around the channels. A lot of our competition is either on chat or voice just because of the history around the specific channels they support. Right from the beginning, we have seen voice and chat as constants.

The second is the ability to drive autonomous transactions. Somebody calls an airline to reschedule their ticket. You could still automate that conversation by saying that you will take your data and somebody else reschedules at the backend. The conversation is still automated, but the process is not.

Coming from the Asia-Pacific messaging-first markets, here the messaging is not seen as a supplementary channel. Here, it’s a core channel. You need to get banking transactions done. You need to reschedule them entirely on chat. There’s a nice workflow agent that works along with the conversation automation. End-to-end autonomous completion and automation have been a significant capability. Maybe we are the company that processes the most number of transactions on conversations than any other company in the world.

The third part is born in a multi-lingual environment. We have the most multi-lingual conversations happening in real-time. For companies that have a multi-geography presence and users with multiple languages, our solution is differentiating. When we put all this together, what we are delivering is not a single-channel CX automation, but it’s a total experience.

Companies are out looking at that total experience. If they are looking for a virtual assistant that can handle hundred-plus tasks in a single front-end interface, we come out as a differentiated platform for that.

Sramana Mitra: Good description. Multilingual is one, and voice and text are another. Then the chat platforms. Given that positioning, whom do you see in deals? Do you see Aisera?
Raghu Ravinutala: Not a lot. We do see Moveworks and Core.ai.

Sramana Mitra: How developed are your system-integrated channels? The kinds of things that you’re talking of doing, this is the kind of thing system integrators do well, right? You have talked about customer support use cases. You talked about IT support use cases and HR support. If you went into an enterprise, you could do millions of dollars of business in each enterprise. Are you penetrated into that channel?
Raghu Ravinutala: Yes, I talked about Roche Pharma. That’s a joint win with Accenture. It’s deployed by Accenture. We have deals with Infosys. India’s largest government implementation, which is for India’s income tax, runs automated servicing for their customers and is completely handled and deployed by Infosys on top of our platform.

TechMahindra exclusively chose us for their customer’s conversational AI implementation. We are well-entrenched in the SI-based ecosystem. We also have the ability to build a good part of that within the company as well. We have a reasonably-sized onboarding and professional services team. Some of our premium customers handle the professional services integration.

Sramana Mitra: Talk a bit about the multi-lingual capability. Technically, how complicated is it to introduce a new language?
Raghu Ravinutala: The important part of the multi-lingual capability is the architecture of the model that you use from day one. There could be 50 different models that you’re potentially using, but the complexity is that a single change in the core training is somebody needs to make those changes across the 50 of them.

The core part of how we’ve designed that multi-lingual system is having a single model across several different languages which are retained based on differences and similarities. That enables a particular customer to have a set of conversations trained in one language and have it available in different languages. This is the most complex part of handling multi-lingual conversations. At the same time, we need to make them scalable. For us to add another language, it’s prebaked. It’s the same model. There is no R&D required.

Sramana Mitra: When you started this, did you use existing technology from elsewhere? It sounds like an enormous amount of work. You must have used components from elsewhere.
Raghu Ravinutala: Our first deployment was based on a Microsoft stack. This is something that I believe as well. Building a product is like building a swimming manual. You can’t build a swimming manual without swimming yourself. For the first few implementations, we were using outside technology. This helped us figure out what needed to be built that can help us differentiate in the space.

Sramana Mitra: You moved out of the Microsoft stack?
Raghu Ravinutala: We were there only for a few months. Right now, we don’t use any company’s technology stack. We are just hosted on different platforms. We don’t use any company’s stack.

Sramana Mitra: The speech-to-text is also your own?
Raghu Ravinutala: Yes, some parts of it are our own. We also use a lot of open-source models for that. We use some of the text-to-speech which we believe is reasonably commoditized. We use some components of that from different providers as well. The core NLP is 100% our own.

Sramana Mitra: Interesting. What use cases are you seeing the maximum adoption of?
Raghu Ravinutala: 60% of where we see the demand is on customer support. COVID has led to a lot of explosion of marketing and commerce use cases, especially in emerging markets. People want to buy things on WhatsApp. Somebody can select their shoes and buy Adidas on WhatsApp. We have seen an explosion of commerce use cases.

Sephora is another large customer where it is automation plus live sales experience where the consumers use WhatsApp and talk about their needs. It’s just like what someone experiences in a store. There is always an assistant that can jump in. We have seen commerce and marketing-related use cases explode. We are also seeing HR and IT automation use cases. This is especially in super enterprise companies where they have 100,000 employees. This is exclusively for companies that are across geographies with 100,000 to 200,000 employees.

Sramana Mitra: What is your relationship with companies like Zendesk and Freshworks?
Raghu Ravinutala: I call it frenemies. We integrate with Zendesk. Any transfer is handled by Zendesk. We have a lot of cases where we have replaced Zendesk as well. In some cases, we do compete with handling the same budget out there. The biggest thing I talk about is Salesforce as an investor in the company.

The way we see it is if there is a CRM that is defined or designed in the AI era, it would be something like what we are building. Software is customer relationship management. Predominantly, CRMs have stored customer data, but the relationship is managed by a human. Humans are interacting with the user and provide the actual service.

If you see how that paradigm has shifted, the software has moved on from just storing the customer data to actually interacting with the customer and managing that relationship. It is owning more and more the part of the customer relationship with the end customers. If there is a CRM that is built for the AI world, it is something that will not require a lot of human intervention. That’s the direction that the company has an opportunity to take over the next five to 10 years.

Sramana Mitra: Is all your customer base in the B2C model?
Raghu Ravinutala: Predominantly. The reason is B2C is the area where there is a scale in the number of users and conversations and the depth of requirement needing code transactions to be involved. There is a need for a larger spectrum of languages and channels in which channels want to interact. B2B is primarily lead generation and provides FAQs.

The rest of the B2B service is handled either offline or through their own proprietary software. We really started on B2C. That’s where our monetization is concentrated. Just because it’s consumption-based, you have large numbers of volumes interacting on your platform. We are 95% B2C.

Sramana Mitra: You talked about how long it took you to get to a million in revenue. Then you raised all this money. To what extent have you accelerated the revenue growth? How is that moving?
Raghu Ravinutala: From the time we raised, the company has grown somewhere between 25x to 30x growth year on year. We have never seen a year where we haven’t grown nearly 2.5x. One aspect that we are proud of and that’s driving a lot of the growth is net retention owing to the consumption-based model.

They start using us for a use case. They expand the channels and expand on the use cases. They start on chat and get on voice. We are 150% in our net retention revenue. Every single dollar a customer spends this year expands by 50% over the next year without any sales effort. That drives a lot of growth.

Sramana Mitra: I’m thrilled to hear your story. It’s a pleasure to see that you have navigated a world-class company. I’m sure you will get a lot of acquisition offers. Then it’s a question of what is your level of ambition of building an independent company. You do have an opportunity to build an independent enterprise software company here.
Raghu Ravinutala: That means a lot. Thank you. We are fully committed to go all the way. We are excited about building this. We like building. We don’t have any short-term liquidity needs.

Sramana Mitra: I don’t think you’ll have liquidity needs. In the kind of company you’re talking about, the beauty of it is, there is so much expansion and upsell within the major enterprise accounts. These are large enterprise deals. There should be no need for additional capital if you don’t want additional capital.
Raghu Ravinutala: Absolutely. There are 400 billion customer support calls every year. Companies spend about a trillion dollars. The state of automation is less than 1% of the 400 billion have any kind of automation. The situation could be flipped in 10 years. That’s the opportunity. The second biggest thing is none of them have any structured representation, which is a gold mine. By the nature of automation, you’re getting structured data.

Sramana Mitra: You can model them. You can do product recommendations. Thank you for your time.

Key Takeaways

• Yellow.ai bootstrapped from zero to $1M ARR over roughly 2.5 to 3 years, reaching profitability from year one, before raising any institutional capital.
• An early Microsoft/Azure partnership substituted for paid customer acquisition – supplying the first 5 to 10 reference customers and growing to 35 to 40 by $1M ARR with a team of just 22 – showing how non-dilutive channels can replace capital in the early growth phase.
• Funding came only after proof: Series A (4M, Lightspeed India, 2018), Series B (~20M, Lightspeed US, for US expansion), and Series C (~$79M) to accelerate growth post-COVID – capital deployed for expansion, not survival.
• Once raised, capital compounded efficiently rather than substituting for product-market fit: the company grew 25x to 30x cumulatively, driven substantially by 150% net revenue retention from its consumption-based pricing – growth capital amplifying an already-proven model.