ZohoDay25 – The AD Era, Build Vs. Buy, and Playing the Long GameZohoDay25 – The AD Era, Build Vs. Buy, and Playing the Long Game

Last year was my first time attending Zoho’s analyst event, and one of the key ideas that stuck with me then was their focus on the “long game.” Being a private company, Zoho can chart their own course without shareholder input, and one of the key ideas that stuck with me at this year’s analyst event is how they were able to pivot toward AI without pivoting away from their long game. Everything moves faster with AI, and being such a transformative technology, you have to be agile to stay on the right side of change.

Part of charting your own course is taking a proactive approach to managing change, which Zoho does very well. Public companies tend to be more reactive, following market shifts based on quarterly reporting cycles, but Zoho does not face these constraints. This brings us to the “After Deepseek erai” i.e. the AD Era, and what’s happening right now in the AI world, where all players must respond. The biggest story in AI is how DeepSeek, with its low-cost base, threatens the AI status quo led by capital-intensive Big Tech.

Disruptive as this may be to OpenAI, Amazon, Google et al, Zoho sees an opportunity. Cofounder and long-time CEO Sridhar Vembu framed his opening keynote within the AD Era and he sees this as a pivotal inflection moment away from the “Big Capex,” or BC Era that has defined AI till this point.

In the BC Era, the Big Tech players have established AI dominance with unsustainable levels of spending - over $250 billion to date. Not only does that keep the barriers to entry impossibly high – and inhibits innovation – but the market’s appetite for AI will wane if they don’t see an ROI that merits today’s levels of investment.

How Zoho Competes

Zoho isn’t Microsoft or Google – and never will be – but it doesn’t have to be. To compete in the SaaS world, you have to be agile as well as profitable. For Zoho, being agile is coming to market with the right offering at the right time, and one of the big news items from the event was the launch of Zia Agents, their Agentic AI platform.

Chief Evangelist Ragu Vegesna provided a detailed overview of how Zia uses contextual intelligence with Large Language Models (LLMs) to empower AI agents to handle a wide range of tasks and queries. Ragu showed how Zia can process both structured and unstructured data, all in a trust-based framework defined by privacy, permissions, and policies. These details would require a separate article to unpack, with the main takeaway being that the current iteration of Zia is on par with what Big Tech has in the market today.

In terms of pivoting on a personal level, Vembu is now stepping down from his CEO post, and taking a new role as Chief Scientist. Aside from being the heart and soul of Zoho, Vembu recognizes the importance of innovation in the AD era, and will now focus his energies on R&D.

Profitability is another reason why Vembu feels Zoho is so well-positioned to compete against Big Tech. Zoho’s financial operations continue performing at a high level, and while the numbers are NDA, CSO Vijay Sundaram provided plenty of proof points. In terms of growth, Zoho maintains healthy 5-year CAGRs for revenues, number of customers, and paid registrations.

Churn is another important success metric, and with steadily declining numbers, they are clearly doing a good job retaining their customers. Sundaram also reviewed how their customer base is increasingly global, and this diversification protects them from downturns in regional markets.

Perhaps most important is the Rule of 40 as a measure of financial performance. This metric is a combination of growth rate and profit margin, and only the best-run companies get above 40. Without sharing specific numbers here, what I can say is that from a list of many household tech names, only two companies were over 40 – Microsoft and not far behind, Zoho.

While it’s too early to tell if DeepSeek will truly “demolish” - in Vembu’s words - the BC model Big Tech is using to create the world’s AI infrastructure, it’s clear to me that Zoho is playing the long game very well.

Build Versus Buy

Aside from going up against Big Tech, the AI infrastructure space in which Zoho competes is volatile, both in terms of being nascent and highly uncertain in the AD era.

More than ever, this is not the time to be reactive and at the whim of AI players whose business model is on the verge of being disrupted. Again, this is where Zoho’s long-game mentality will serve them well. Build versus buy is the cornerstone of any business strategy, and with AI evolving so quickly, the reactive response would be to buy. The risks and costs for building in-house are too high, plus it takes too long, so buying is the best way to keep pace.

That’s the path most companies would take, but Zoho isn’t just any other company. Their choice to build reflects the realities outlined above, especially around AD. DeepSeek has shown that it’s possible to develop AI at a manageable cost, and this actually aligns with the pro/low/no-code approach Zoho has been following for some time.

The big idea here is that Zoho recognizes the potential of AI to develop applications quickly and at a much lower cost than what the market has been conditioned to expect from Big Tech. Strategy-wise, the rationale here for building instead of buying becomes very clear. Aside from being another example for how Zoho is playing the long game, by building its own AI infrastructure, the company will earn more revenues, and realize higher margins, further strengthening its financial position.

Another unstated benefit of building is how this approach supports Zoho’s overall architecture, where everything runs on top of a common platform. As the visual below shows, there is an underlying orchestration layer that supports deep integrations across all their vertical platforms – which Zoho calls “domain specific”.

This means that for each domain – collaboration, CX, etc. – Zia is trained with inputs and LLMs specific to that domain, providing the “contextual intelligence” referred to earlier that enables Zia Agents to manage tasks and queries autonomously. By building their own AI infrastructure, Zoho has far more control over making everything work just right, than by relying on third-party partners who themselves are trying to navigate today’s fluid AI market.

To cap off this build versus buy analysis, during his presentation, Raju Vegesna explained what building Zoho’s AI infrastructure looks like, and here are a few examples. First of all, they are not a reseller of AI solutions or applications. They build their own, and they support their own offerings – not white-labeling other vendor’s offerings.

To support this, they operate 16 data centers globally – and are building more. This is fundamental to ensuring high availability as well as maintaining a strong posture around security and privacy of their customer’s data.

Even more fundamental is the technology that provides the horsepower that makes AI possible. For semiconductors, they don’t just buy from the likes of Nvidia and AMD; Zoho partners with them to build to their requirements.

Along the same lines, Zoho is investing in GPU capacity, not just to support AI workloads (which everybody does), but for accelerating databases, so their customers can access their data more quickly. Another area would be investing in ASICs to support specific use cases, namely to process video in their datacenters. All of these examples reflect long-game thinking, where the build strategy reflects the importance of owning your AI infrastructure for competing against Big Tech.

Conclusion

By now it should be clear that Zoho isn’t like most tech companies -- in a good way. From a distance, Zoho may look like a wannabe Google, with a sprawling collection of workplace applications. Looking more closely – as any of the 100+ analysts in attendance would have seen – Zoho has evolved from an applications vendor to a leading-edge AI platform provider.

They’re certainly not alone in doing that, but the Zoho Way is definitely working, perhaps more effectively than most would realize. I’ve only shared here a few of the many highlights from ZohoDay25 to support that conclusion, and if Vembu’s reference to the AD Era resonates for you, then so too should their long-game approach.

This post is written on behalf of BCStrategies, an industry resource for enterprises, vendors, system integrators, and anyone interested in the growing business communications arena. A supplier of objective information on business communications, BCStrategies is supported by an alliance of leading communication industry advisors, analysts, and consultants who have worked in the various segments of the dynamic business communications market.