Sramana Mitra: I have one last question, Yash, before we adjourn. The SaaS industry has been around for 15-20 years. Around 2008 is when it started really hitting its stride. It’s 2024, so it’s been a while. Then AI came in around 2015-16. Generative AI came in last year. There are cloud companies, SaaS companies, and vertical SaaS companies that are much further along, that have some heft. But now there are AI companies coming into the same markets with AI-enabled solutions, competing with those somewhat hefty vertical SaaS companies. And the answer from the incumbents is, we do co-pilots to introduce AI into the process.
What are you seeing in the market? It’s not a sufficient solution to produce an AI-enabled SaaS application. What is your perspective? What are you seeing in your portfolio?
Yash Hemaraj: This ties back to the data challenge I mentioned earlier. You have external data, customer contacts, and internal communication, all of which need to be integrated with the right role-based access controls and identity governance. There are still a lot of unanswered questions in this space. For instance, in an enterprise setting, if an autonomous decision is made, who is responsible? You can’t simply say, ‘the machine made the decision, so I’m not accountable.’ There has to be a clear responsibility matrix for these situations.
This creates a fertile ground for innovation in areas like role-based access control, data identity, and identity governance. Not all companies have the right data architecture or critical feature sets in place, so they’ll need to upgrade their stacks. It’s important to separate the fear, uncertainty, and doubt (FUD) that large players might introduce into the market from the real innovation needed here.
I’ll give an example in the call intelligence space. There’s a company which has kind of invested hundreds of millions of dollars to be the best call intelligence software for sales reps. They worked on it for 7-8 years and have a lot of patents around that. But today, out-of-the-box generative AI models like Llama or OpenAI delivers better call intelligence than than what they’ve spent years and millions developing. That’s the reality that they have to grapple with.
On top of that, they’re sitting on valuations of $5-$10 billion, and now they need to show growth to justify those high valuations—often at 100x or 200x multiples. While they might create fear, uncertainty, and doubt in the minds of potential buyers, they also have to confront these challenges themselves.
What some of our younger companies are doing is leveraging how easy call recording has become. They can take recorded calls and integrate the context from external data, internal data, and customer information—essentially offering the same functionality that took other companies seven years and millions of dollars to build. This gives them a significant advantage, including the ability to undercut costs.
It’s not as simple as established companies introducing co-pilots to their existing customer base. In reality, the market isn’t responding that way because their ability to price these solutions has dropped significantly.
Sramana Mitra: The older companies are stuck on outdated stacks, and their architecture isn’t optimized—they don’t have things in the right places. I come out on the side of the young companies who are starting fresh, who may not have all these customers, but who are starting fresh with a fresh contemporary stack and building specific to that stack and have new, fresh, nimble solutions that can go in and disrupt these existing companies.
Yash Hemaraj: Indeed, 100%.
Sramana Mitra: All right, well, it’s a pleasure. We will continue the discussion in a few months. This industry is moving so fast. There’s so much exciting stuff happening. So thank you for coming. Great to have you and see you soon.
This segment is part 6 in the series : 1Mby1M Virtual Accelerator AI Investor Forum: With Yash Hemaraj, General Partner at BGV
1 2 3 4 5 6