Google Cloud made more money last quarter than Zoho has made in its entire 29-year history combined.
Not ever. Not cumulative. One quarter.
And yet Zoho, a company that has never taken a single dollar of outside funding, just told the world it built its own cloud from scratch specifically to avoid depending on AWS or Google. Not rented servers dressed up as independence. Actual, owned infrastructure: its own databases, its own data centers, its own identity systems, built in-house over nearly three decades.
So either Zoho’s founder is making a wildly overconfident claim, or there’s a version of “competing with the cloud giants” that doesn’t require matching them dollar for dollar. Turns out it’s the second one, and the details of how the Zoho cloud actually works, and holds its own, are more interesting than the headline suggests.
What the Zoho Cloud Is Actually Built On
This isn’t a recent pivot. According to co-founder Sridhar Vembu, the decision to build in-house infrastructure dates back to Zoho’s early years, when the company needed somewhere to run its own applications. Over time, that internal system grew to cover the core building blocks of a cloud platform: databases, compute, storage, search, and identity management, all developed internally rather than licensed from a third party.
The company now operates more than 18 data centers worldwide, with data hosted according to the country or region it comes from. Indian customer data, for instance, stays in India, split across facilities in Mumbai, Delhi, and Chennai, with another planned in Odisha. Zoho has also been expanding fast outside India. It opened two new data center regions in Dubai and Abu Dhabi in January 2026, and according to a Bloomberg interview with the company’s Middle East and Africa head, Zoho now plans to build additional facilities in Saudi Arabia and South Africa, with the UAE alone posting customer and revenue growth rates of 77% and 45% a year since 2021.
That global buildout matters because the pitch isn’t “trust us with your data.” It’s closer to full control. Vembu has said Zoho doesn’t host its products on AWS or Azure at all. The one exception is regional switching nodes, which some large cloud providers help route traffic through for speed, though Vembu insists customer data never sits there.
The Numbers Nobody Can Ignore
Here’s where the comparison gets uncomfortable for Zoho.
AWS controls roughly 28% of the global cloud infrastructure market, with Microsoft Azure at around 20 to 21% and Google Cloud close behind at 14 to 15%, based on Synergy Research Group’s figures for the first half of 2026. Google Cloud alone pulled in $24.8 billion in the second quarter of 2026, up 82% year over year, an acceleration from the previous quarter’s growth rate.
Zoho, by comparison, reported operating revenue of 123.1 billion rupees, about $1.27 billion, for the fiscal year ending March 2025, up 17.7% from the year before. Net profit actually slipped slightly, to 31.9 billion rupees ($330 million) from 33 billion rupees the year prior. The company remains entirely self-funded. It has never raised outside capital, which is unusual for a company operating at this scale and almost unheard of among cloud infrastructure providers.
Put simply, Google Cloud generates more revenue in roughly seven weeks than Zoho does in an entire year. That gap in scale shapes everything else about how this competition actually plays out, and it’s worth keeping in mind before buying into any “underdog beats hyperscaler” framing.
Why Zoho Chose the Harder Path
So why build your own cloud when renting one is cheaper and faster, at least at the start?
Part of the answer is control over data residency, something that matters increasingly to regulated industries and governments wary of storing sensitive data with US-based hyperscalers. Part of it is cost predictability over the long run, since owning infrastructure means Zoho isn’t exposed to the pricing decisions of AWS or Google. And part of it, based on how Vembu talks about the decision, seems philosophical: a belief that owning the full stack, down to the hardware layer, is what let Zoho stay profitable and independent while competitors burned through venture funding chasing growth.
That tracks with a broader pattern in how companies evaluate build-versus-buy decisions for core infrastructure, the same tension that shows up when organizations design their technology operating models around long-term control rather than short-term speed. The build-your-own route is slower and riskier early on, but it removes a layer of dependency that can become expensive later.
The Real Cost of Staying Independent
Owning your own infrastructure sounds appealing until the hardware market turns against you, and that’s exactly what’s happening right now.
In October 2026, Vembu announced Zoho was freezing pay raises, directly citing the cost pressure created by the AI boom. He pointed to “a massive 4-fold increase in server prices which we need for data centres,” alongside rapidly increasing AI usage costs the company is trying to control. Because Zoho runs its own data centers instead of renting capacity from a hyperscaler, it’s more directly exposed to swings in hardware pricing than competitors who simply pay whatever a cloud provider charges per hour.
It’s a reminder that the server and chip supply chain doesn’t move independently of who’s buying from it. The same dynamic played out during the 2021 chip shortage, when demand for components outran what manufacturers could supply. AI compute demand is now doing something similar to server and data center hardware pricing, and companies that own physical infrastructure, like Zoho, feel that pressure far more directly than companies that merely rent.
Can Catalyst Actually Compete?
Catalyst is Zoho’s answer to the obvious next question: if you’ve already built a cloud, why not let other people use it?
Originally built as an internal tool for developers extending Zoho’s own applications, Catalyst has grown into a full-stack serverless platform, covering frontend hosting, backend functions, authentication, data storage, and AI and machine learning tools, all in one console. Pricing follows a pay-as-you-go model with no subscription requirement, developers get a free sandbox tier to experiment, and new users can claim $250 in free credits over six months. In September 2026, Zoho went further and made Catalyst completely free for students building non-commercial projects, with no credit card required.
That’s a meaningfully different pitch from AWS or Google Cloud, which can feel overwhelming for a solo developer or small team navigating dozens of overlapping services and pricing tiers. Catalyst’s simplicity is the point, not a limitation. Whether that’s enough to pull serious developer mindshare away from established platforms is a separate question entirely.
Where Zoho Has an Edge, and Where It Doesn’t
Independent reviews suggest the honest answer is: it depends what you’re optimizing for.
A 2026 comparison from Cloudwards, evaluating Zoho Workplace against Google Workspace, found that Google’s larger global data center footprint translates directly into lower latency, wider availability for distributed teams, and greater scalability for handling large user volumes. That’s a direct byproduct of infrastructure scale. Google simply has more computing resources spread across more regions, and no amount of efficient engineering fully closes that gap for latency-sensitive, globally distributed workloads.
Where Zoho tends to win is elsewhere: predictable usage-based pricing, tighter data residency guarantees for regulated industries, and a product philosophy that doesn’t monetize user data through advertising, something Zoho has pointed to directly in its own marketing. For companies prioritizing cost control and data sovereignty over raw global scale, particularly in markets like the Middle East and South Asia where Zoho is actively expanding, that trade-off can be the deciding factor.
What This Means Going Forward
Zoho isn’t trying to dethrone AWS or Google Cloud in the traditional sense, and it would be a mistake to read its infrastructure buildout that way. What it’s actually building is an alternative for a specific kind of customer: one that values data control, predictable pricing, and independence from hyperscaler pricing cycles more than having access to hundreds of integrated cloud services.
That’s a narrower bet than “beat the giants,” but it’s also a more defensible one. The same forces pushing hyperscaler debt and capital spending to record levels, a dynamic examined in how Wall Street has grown more cautious about financing the AI infrastructure boom, are the same forces now pushing up Zoho’s own hardware costs. Everyone building physical infrastructure right now is exposed to the same pressure, just at wildly different scales and with very different balance sheets behind them.
Whether Zoho’s self-funded, infrastructure-first approach ages well will depend less on matching AWS feature-for-feature and more on whether enough customers keep valuing control and predictability over sheer scale. Right now, that bet looks sustainable. It just isn’t the same competition the headline suggests.
The takeaway
Zoho’s cloud strategy isn’t about outscaling AWS or Google Cloud. It’s about offering something the hyperscalers don’t prioritize: full control over infrastructure, predictable pricing, and independence from the pricing cycles that come with renting someone else’s cloud. That’s a narrower bet than challenging the giants head-on, but a more defensible one. Zoho isn’t shielded from the pressures reshaping the industry either. The same forces driving up hyperscaler debt and capital spending are also pushing up Zoho’s own hardware costs, just at a fraction of the scale and without the balance sheet to absorb it as easily. Whether this approach holds up long-term will come down to one thing: whether enough customers keep choosing control and predictability over sheer scale. Right now, that bet looks sustainable, but it’s a different competition than the headline suggests.
