Earnings call transcript: Fractal tops $100 million in Q2 2027 as profit jumps

Fractal’s Q2 FY2027 revenue rose 20% year over year to INR 957 crore (more than $100 million for the first time); adjusted EBITDA grew 30%, with margin expanding 123 bps to 16.2%, while net income increased 139% to INR 79 crore. Gross margin fell 225 bps to 44% as the company invested in AI engineering talent, but management expects improvement in Q3 and Q4 and sees enterprise AI adoption accelerating in FY2028. Products generated 4% of revenue, and shares were down 1.25% at $732.8 in recent trading; no forecast figures were provided for comparison.

Analysis

The key signal is the gap between reported and constant-currency growth: the rupee provided a material tailwind to the reported result, so the underlying demand trajectory is less exceptional than the headline growth implies. Gross-margin recovery is the near-term underwriting test, not the earnings growth rate. The margin bridge depends on newly hired forward-deployed engineers becoming billable and on the currency benefit not reversing; if utilization lags, Fractal could face both under-absorbed delivery costs and renewed FX pressure. EBITDA growth despite lower gross margin also reflects SG&A leverage, which may be harder to sustain if sales investment rises to pursue larger deals.

The product story remains option value rather than a proven re-rating catalyst: licensing traction is company-reported, while product revenue remains small relative to services. Partners such as Microsoft and Amazon Web Services may accelerate distribution and implementation demand, but also control customer access and could internalize more of the AI stack over time. Broader AI-services peers face the same tension: demand can expand, while automation and client insourcing pressure labor-based economics.

Reported net-income growth is not a clean run-rate measure: lower ESOP-related charges and the Qure.ai associate loss affect comparability. The next 1–3 months hinge on Q3/Q4 gross-margin recovery and collections; the 6–18 month thesis hinges on utilization, product mix, and enterprise AI budgets. Management claims about future adoption are not yet independently validated. No direct Fractal equity trade is actionable from the supplied identity data; the stock’s valuation and liquidity are also unprovided.

Key Decisions for Investors

• Keep Fractal on a results-monitoring list rather than initiating a directional position: verify constant-currency growth, billable utilization/revenue per employee, and gross margin in Q3 before underwriting the promised recovery.
• Treat product-led upside as a catalyst watch, not base-case earnings: track product revenue as a share of sales, renewals/expansion within customers, and contracted product backlog; verify management’s product run-rate targets because the materials use potentially inconsistent target descriptions.
• Reassess the thesis if gross margin fails to improve sequentially in Q3/Q4, utilization remains weak, or DSO does not return toward its historical range by year-end; these would weaken the case that current hiring is a temporary investment ahead of demand.
• Separate core operating performance from reported EPS: monitor ESOP charges and Qure.ai’s contribution, and avoid extrapolating the year-over-year net-income increase without a normalized earnings bridge.
• For Microsoft and Amazon, treat Fractal’s partner activity as a modest ecosystem positive rather than a material earnings catalyst; watch for evidence that partner-sourced enterprise deployments convert into recurring platform revenue rather than remaining implementation-led services.