E2E Networks Ltd Q4 FY2024 Concall Summary

Key Takeaways

• Revenue surged 67% YoY, but the quarter is small: Q4 revenue hit ₹29 crore, EBITDA ₹15 crore, and net profit ₹4 crore, all growing sharply off a modest base.
• A ₹185 crore CAPEX bet is now in the ground: The entire FY24 deployment is live on the cloud platform, and the selling cycle is just beginning, with POCs and optimization work underway.
• The ambition is to deploy another ~₹800 crore: Management wants to add roughly \$100 million in CAPEX over the next 12 months, funded by a mix of debt, vendor financing, and an equity raise.
• The business model is software-defined, not box-shifting: Management explicitly rejects asset-turn thinking, framing E2E as a technology company where value creation, not being cheapest, drives returns.
• What to watch next quarter: Whether the ₹14-16 crore MRR target materializes from the deployed capacity, progress on the equity fundraise, and any early signs of traction with larger enterprise workloads.

The CAPEX Cycle: Deployment, Selling, and the Next Leg

• ₹185 crore deployed, now entering the selling cycle: The entire FY24 CAPEX is live on the cloud platform; management notes it typically takes a quarter or two for customers to move through POCs and optimization guidance before converting.
• MRR target of ₹14-16 crore from current capacity: Management expects this level to be achievable over the next couple of months, implying a near-term revenue run-rate step-up if conversion proceeds as planned.
• The next ambition is ~₹800 crore in 12 months: In an ideal world, management would deploy another \$100 million, funded through bank financing, leasing partners, vendor financing, and some equity.
• CAPEX is fluid, not a lump-sum commitment: Deployment is spread over time and paced against customer demand, giving the company flexibility to adjust if uptake slows.
• No formal guidance on selling speed: Management declined to predict how quickly the deployed capacity converts, calling quarter-on-quarter forecasting very hard and preferring to look back after the fact.

Tarun Dua: "In an ideal world, like we would like to end another \$100 million or about Rs.800 crores over the next 12 months."

Business Model & Competitive Moat

• Asset turn is the wrong lens, per management: The company sees itself as a technology and software-driven business, not a capital-deployment story; asset turn varies by SKU from 0.5 to 2.0, with a lower bound of 0.55-0.6 but no upper bound in a software-defined model.
• The moat is the integrated ecosystem, not the box: Management argues the combination of infrastructure, software platform, and orchestration is not easy to replicate by simply deploying software onto hardware, creating a barrier versus newcomers.
• Multi-cloud reality means no forced migration: Workloads stay where they run most efficiently; E2E does not see a consolidation imperative and does not need to pry customers off hyperscalers to grow.
• Customer base spans four segments: SMEs form the bulk, alongside higher education and research, startups (which management says are back with a bang), and an auto-flex partner ecosystem in India and abroad.
• Revenue remains India-heavy, with international ambitions: The bulk of revenues come from India; the company is focused on EMEA and the US and will start targeting Southeast Asia, though no formal geographic revenue split is disclosed.

Tarun Dua: "That was not how we look at our business, we look at our business as a technology business, very focused on figuring out the various industry solutions. And over there eventually like, the asset turn doesn't matter, because it's like, very, very capability and technology and software driven business and it's not about being the cheapest on the market. It's about creating enough value for the customers."

AI/ML Platform & Technology Build-Out

• Tir, the AI/ML platform, is advancing rapidly: New features cover training, inference, model inference deployment, and scalable data pipelines, with better integration into the infrastructure services platform.
• Containers and raw OS environments now seamless: Customers can launch raw OS environments using containers on the paid platform, integrated with AI/ML software features, broadening the addressable use case set.
• Compliance, security, and monitoring in progress: The company is building out these features across the entire platform, a necessary step to target larger enterprises with stricter governance requirements.
• Storage innovation continues alongside compute: Work on storage technology for the infrastructure platform is ongoing, reflecting the data-intensity shift AI workloads demand.
• InfiniBand deployed at 1.6 Tb and 3.2 Tb: High-speed interconnects are now live, supporting the GPU clusters that underpin the AI/ML offering.
• All near-term GPU investment goes to Nvidia: No exclusivity exists, and management remains open to evaluating other vendors for specialized use cases based on customer demand.

Tarun Dua: "Now, which means that like earlier you were designing your storage and network and your entire internal bandwidth infrastructure for being able to process like very small amounts of data. Now you actually are building for like being able to handle like 20x, 100x or maybe even 1000 times the amount of the data in motion to be able to process it on like very fast, high speed, GPU processes."

Facility Transition & the Outage

• A black swan outage hit an older facility: A third-party vendor brought power up before the UPS was charged, damaging equipment; the company spent 24-48 hours verifying all customer services were restored.
• Most customers were understanding: Some raised other issues during the engagement, which management frames as an opportunity for deeper customer conversations.
• A two-year migration plan to the new facility: The company will slowly migrate all customers from the older facility to the new one in Greater Noida without disruption, prioritizing customers willing to put in effort today.
• A 200 gig link will connect the two facilities: This long-haul link is prioritized to facilitate the migration and ensure continuity during the transition period.

Tarun Dua: "I would like to point out that, in case that the third party vendor had trained their processes better, we could have avoided a world of pain for us and for our customers, by putting up the power only after the UPS was charged, instead of trying to bring it up couple of times, which damaged some of our equipment."

Chip Supply & Industry Demand

• No repeat of the 52-60 week chip queues expected: Management sees normal queues persisting due to integration requirements, but not the extreme shortages of the past.
• AI/ML workloads are 4x-5x the size of typical CPU workloads: This structural shift underpins the exponential growth expected in data center capacity for AI over the next five years.
• Nvidia relationship is functional, not formally prioritized: E2E is listed as a cloud service provider partner and has received all support demanded; management calls it a relationship without additional formal requirements.

Tarun Dua: "There is always going to be some queue. But it's not going to be the longest queue that we have seen in the past of that type okay, like we are looking at 52 weeks, 56 weeks, 60 weeks and so on. That is not going to be the future."

Financial Structure & Cost Dynamics

• Major expenses are directly tied to revenue: Data center services (power, base, bandwidth) and software licenses have a one-to-one relationship with services delivered, making them largely variable.
• Services as a percentage of revenue will decline slowly: A broad percentage exists, and management expects it to go down gradually as volume increases.
• Depreciation shifted from WDV to straight line from April 2023: No further change in accounting policy; right-of-use assets relate to equipment leases (lease-to-own), while data centers are rented.
• Employee costs may rise as hiring continues: Management operates in startup mode with doors never closed for hiring, filling gaps in technology, marketing, and operations teams organically without a formalized plan.
• No formal FY25 revenue or margin guidance given: When an analyst projected ~₹140 crore revenue and ~₹40 crore bottom line based on prior growth and margin commentary, management declined to confirm, saying they cannot promise that.

Guidance & Commitments

Key Metrics