Fractal Analytics Ltd Q4 FY2026 Concall Summary

Key Takeaways

• 19% FY revenue growth, operating leverage visible: Full-year revenue grew 19% (17% from existing clients), while Q4 adjusted EBITDA rose 28% on 17% revenue growth, expanding margins 189 bps YoY to 22.1%.
• Enterprise AI thesis is the core story: Management frames the opportunity as a 100x net expansion in software demand, with Fractal positioned to capture the AI-led transformation phase through domain expertise, the Cogentiq platform, and a shift toward outcome- and license-based revenue.
• TMT vertical dragged Q4, but the drag is client-specific and expected to ease: TMT declined 19% in Q4 on two discrete client issues (a JV that moved business away, and a restructuring), not a structural demand problem; management expects improvement in coming quarters.
• Qure.ai losses are lifting, removing a persistent profitability overhang: USAID funding dried up instantly, creating a major headwind, but Qure now sits on a strong order book and is expected to stop contributing losses to Fractal.
• What to watch next quarter: Whether TMT stabilizes as promised, the pace of the shift toward outcome/output/license revenue (targeting 60% in 2–3 years, from ~40% today), and whether Qure.ai’s order book converts into the forecast recovery.

Enterprise AI Opportunity & Strategic Positioning

• 100x net opportunity thesis: Management estimates the world needs 1,000x more software than exists today; AI compresses build effort by ~90% (10x), leaving a 100x net expansion. The pie grows by orders of magnitude, not shrinks.
• Value metric shift: Cost per token is the wrong metric; the right one is value generated per token. The prize is work previously impossible: real-time SKU pricing, novel drug discovery, meeting every customer need with speed and precision.
• AI-led transformation phase: The last two years were about AI infrastructure, won by frontier labs and infrastructure providers. The next several years belong to companies that can translate AI excitement into outcomes inside complex organizations, requiring deep domain expertise, frontier AI capability, and contextual understanding of design and behavioral sciences.
• Three go-to-market vectors: AI-led transformation (reimagining business processes on Cogentiq), AI foundations (modernizing tech spends with AI, cutting costs 50–75% versus other providers), and the existing analytics business.
• Platform compounding thesis: Building on one platform (Cogentiq) every time turns a service business into something closer to a software business; work compounds and so do the economics.

Srikanth Velamakanni: "The moment in front of us, frontier intelligence, agentic capability, and the transformation phase of enterprise AI is exactly the moment we have been spending decades preparing for."

Revenue Performance & Vertical Dynamics

• Client metrics healthy: Net revenue retention 117% for the year (112% Q4); NPS hit 81 in Q4, the highest recorded; $1M+ clients grew from 53 to 59.
• Revenue per employee rose 5% in rupee terms to INR 75 lakh (~USD 85,000).
• Must Win Clients strategy: Targets companies with USD 10 bn revenue, USD 20 bn market cap, or 30 mn consumers; aims for USD 2 mn to USD 100 mn per client per year. Smaller companies may be addressed later via self-serve Cogentiq versions.

Srikanth Velamakanni: "Literally every aspect of a business is up for re-imagination and rebuilding. That is what we are seeing."

Margins & Profitability

• Adjusted EBITDA reached 22.1% in Q4, up 189 bps YoY, driven by gross margin expansion plus 140 bps reduction in SG&A. A UK grant benefit and variable pay adjustments added 147 bps; some of these may not repeat.
• Full-year adjusted EBITDA was 17.6% versus 17.4% prior year. Gross margin expanded 93 bps but growth investments in relationship management and new offices added 96 bps to SG&A as a percentage of revenue.
• Fractal Alpha losses narrowing: Losses fell from INR 26 cr (FY25) to INR 15 cr (FY26), a 43% reduction, while investments in sales and R&D continued. Losses have trended down since FY23: INR 54 cr → INR 44 cr → INR 26 cr → INR 15 cr.
• Qure.ai drag lifting: USAID, a major funder of tuberculosis programs, was shuttered by DOGE, drying up a significant Qure revenue source instantly. Qure has been a persistent drag on Fractal’s profitability through the IPO and subsequent quarters. As of April 2026, Qure sits on a strong order book and is expected to stop contributing losses.
• Growth-profitability philosophy: Management wants to increase revenue growth rate while expanding gross margins, not by sacrificing them. The approach is to maximize revenue growth rate while hitting a certain profitability threshold.
• Cash flow: CFO increased despite INR 86 cr increase in working capital; DSO improved two days to 72. FY26 CFO-to-adjusted-EBITDA conversion of 70% is representative of steady state (FY25’s 82% benefited from a 12-day DSO improvement).

Srikanth Velamakanni: "That drag is getting lifted as we speak, and Qure will have a much better year and also not contribute to additional losses to Fractal."

Revenue Model Shift: Outcome, Output & License

• Targeting 60% of revenue from output-, outcome-, or license-based models within two to three years; today the share is roughly 20 points lower (~40%).
• License revenue ambition: Grow license-driven revenue from 3% today to 20% by 2030. License-driven revenues carry the highest gross margin, though ticket sizes are smaller. The best structure combines outcome- or output-based deals with a license component.
• New global commercial structure: Fractal unified commercial discipline under a Global Chief Commercial Officer, Matt Gennone (also CEO of Cogentiq), with a remit to move engagements toward outcome-, output-, and license-driven structures and pricing.
• TMT margin expansion path: TMT work is more input-driven than output/outcome-driven; as those models take off, margins are expected to expand. TMT is already profitable, only a couple of points below other verticals.
• Gross margin expansion mechanism: As the share of license revenue and output/outcome-based revenues increases as a percentage of the overall pie, each of those moves expands gross margin.

Platform, Products & Partnerships

• Cogentiq platform: An agentic AI platform allowing AI agents to plan tasks, coordinate, access enterprise systems and data, and execute work reliably. A MAG-7 technology company selected Cogentiq after a three-month evaluation of all agentic platforms for their marketing transformation, serving hundreds of millions of monthly website customers from the CMO’s office.
• Flyfish.ai launched: An agentic sales platform with 35 coordinated agents that research accounts, draft outreach, and manage pipeline.
• Vaidya 2.0: Healthcare foundation model now available on all major mobile platforms with developer API access; remains the world’s first model to cross 50 on OpenAI’s HealthBench Hard.
• PiEvolve: Agentic engine for autonomous machine learning, now available to all Fractal people and clients; ranks among top-performing agents on OpenAI’s MLE-bench.
• Fractal Alpha segment: Includes Asper.ai (supply chain AI) and Analytics Vidhya (AI learning platform). FY26 revenues grew 41% YoY (Asper +31%, Vidhya +49%); gross margin expanded 143 bps YoY. Growing faster than Fractal overall; revenue contribution will increase.
• Partnership muscle strengthening: Fractal wants to be much closer to data platform partners (Databricks, Snowflake, etc.), bringing them opportunities and working on opportunities they bring. Certified professionals number in the hundreds across each platform.
• ESG practice emerging: Working with largest clients on ESG through the supply chain practice, using data and AI for sustainability advantages and reporting. Fractal holds a CDP rating of B and expects to improve.

Guidance & Commitments

Key Metrics