The 2026 Observability Forecast: India Edition

New Relic’s 2026 Observability Forecast: India Edition, produced in partnership with Enterprise Technology Research (ETR), isolates 500 IT and engineering leaders from India - the single largest country sample in this year’s global survey of 2,575 respondents. The report reveals a paradox: India amplifies the key patterns of the global cohort - on two fronts. While it is a world leader in AI adoption, OpenTelemetry standardisation, and observability adoption ROI, it also battles the heaviest operational strain and steepest outage costs anywhere surveyed.

Closing the agentic AI monitoring gap

As autonomous AI agents and AI application monitoring adoption rise, Indian organisations have moved faster than any other market to eliminate blind spots. On average, 56% of Indian enterprises actively monitor AI agents in production compared to 35% globally, leaving just 17% of deployments unmonitored. This commitment to early visibility extends across all AI features, as Indian organisations lead globally in AIOps capabilities at 54% and machine learning model monitoring at 57%. By embedding telemetry directly into AI workflows, Indian enterprises can establish greater control over autonomous agentic systems and create checks and balances.

AI-generated code and the software complexity split

The growth in AI-assisted development tools has fundamentally reshaped how software is authored across the subcontinent. 80% of Indian leaders agree or strongly agree that the rise of AI-generated code makes robust observability more critical, with 47% saying they strongly agree - the highest figure recorded in the global study. AI coding tool adoption runs hotter in India than anywhere else, led by GitHub Copilot at 54%, Google Gemini Code Assist at 51%, and OpenAI Codex at 45%, with only 1% reporting no AI tool usage. However, this velocity has split respondents: while 56% say AI integrations have simplified stack management, 39% report that AI has added complexity. This demonstrates the value in monitoring code generation.

The steepest downtime bill in the world

Scaling for rapid digital growth without the matching operational discipline has created severe financial penalties for Indian enterprises. High-impact outages take place multiple times per day for 31% for Indian organisations - a tenfold jump from 3% in 2025. The average revenue cost per hour of downtime has escalated to $2.43 million, bringing the estimated cost of a single average high-impact outage to $3 million. When it comes to the causes of unplanned outages, third-party or cloud providers, network failure and deploying software changes were the top three.

OpenTelemetry leadership versus tool sprawl

India dominates the rest of the world when it comes to standardising on OpenTelemetry, with 42% of organisations fully standardised as their primary form of telemetry collection - nearly double the 22% global baseline. Yet, this data standardisation advantage is being undercut by tool fragmentation. The average Indian enterprise now juggles six observability tools, up from five last year, with 12% running ten or more tools. Standardising telemetry collection without consolidating can prevent AI models from correlating insights and understanding, and only 20% of Indian leaders report that AI-assisted observability has significantly improved event correlation, trailing the 27% global average.

What this means in 2026 and beyond

When taken together, the survey findings reveal a clear pattern across India: organisations here report pairing aggressive AI adoption and OpenTelemetry standardisation with unified platform visibility and are significantly more likely to capture 3x to 5x financial returns. Conversely, those organisations deploying AI agents across fragmented, multi-tool environments report an escalating operational burden; spending half their engineering capacity on firefighting and experiencing the region's steepest outage costs.

Download the full report to see how your organisation compares.