Hexaware reduces false positives by over 75% with Elastic

Detecting and addressing anomalies across environments at scale

Hexaware’s IT operations generate immense volumes of data. However, surfacing actionable insights across infrastructures was slow, manual, and fragmented. As the business scaled, inefficiencies began to compound:

• Alert fatigue: With 1,000+ false positives per week in one use case, and another logging 500+ alerts weekly, false positives were overwhelming engineers, consuming hours of review time, and diverting focus from real issues.
• Inefficient onboarding: With onboarding lasting up to 18 months, new engineers were dependent on senior staff to navigate systems and dashboards, taking time from lead engineers.
• Underused historical data: Valuable context often went untapped during incident investigations, slowing mean time to repair (MTTR).
• Inefficient workflows: Dozens of dashboards and hundreds of KPIs needed to be checked to assess service health and incident impact, while complex queries were slow to run and prone to inaccuracies.

With high operational drag and limited visibility across mission-critical environments, Rao needed a scalable way to detect and resolve anomalies across infrastructure, applications, and user experiences.

Workflows transformed with AI to improve issue resolution

To overcome operational inefficiency, Hexaware created new issue-resolution workflows by implementing Elastic Observability and Elastic’s AI Assistant, both built on the Elasticsearch Platform, to:

• Accelerate problem-solving: Shift from manual, expert-dependent operations to a knowledge-amplified model where engineers of all levels can resolve issues faster with contextual guidance.
• Streamline onboarding: Replace memorization of dashboards and codes with task-centric, in-workflow assistance.
• Get proactive alerts: Upgrade from reactive alert chasing to proactive, insight-led operations that surface anomalies, correlate signals, and recommend next actions.
• Make decisions fast: Use conversational analytics to replace dashboard chasing with natural-language access to KPIs and real-time visualizations, enabling faster decision-making in live operations.
• Increase reliability: Reduce noise by filtering out false positives and streamlining alerting. Teams can focus on high-value engineering work that strengthens service reliability.

With Elastic Observability, Hexaware used machine learning to detect anomalies and reduce false positives across 25+ applications. By moving from reactive monitoring to proactive insight, Rao’s teams can correlate signals across infrastructure, applications, and user experience. They also leveraged autoscaling in Elastic Cloud for self-healing patterns and real-time resilience.

With Elastic AI Assistant, Hexaware accelerated workflows for managed services teams and new hires, enabling them to make meaningful contributions faster. Elastic enabled natural language incident investigation and automated charting to streamline workflows and surface insights instantly.

When Hexaware wants to expand AI use cases in the future, the teams can easily automate IT operations with Elasticsearch Platform, especially in call centers and other support activities.

Improved operational efficiency and client satisfaction

With the Elasticsearch Platform, Hexaware saw dramatic improvements in performance and efficiency while gaining complete visibility into its environments. New engineers now become project-ready in just three months instead of a year or more, reducing reliance on lead engineers. Now, Rao’s teams can have engineering talent focus on higher-value work and strengthen client satisfaction.

With Elastic AI Assistant, Hexaware achieved:

• 50% improvement in operational efficiency: Engineers can access KPI data and generate reports in minutes instead of hours.
• 75% decrease in onboarding time: Hexaware’s “Mavericks” program now brings new hires up to speed in a quarter of the time, with 90% less reliance on senior engineers. New-hire ramp time went from over a year to under a quarter.
• Boost in client satisfaction: Elastic’s “wow factor” in observability led to new value proof points across mission-critical applications.

“Elastic has made significant strides in integrating generative AI into its core technology. By analyzing vast amounts of data and identifying patterns, generative AI empowers us to streamline processes, enhance efficiency, and drive innovation for both Hexaware and our clients.”

– Sanjesh Rao, SVP, Enterprise AI and transformation, Hexaware

With Elastic Observability, Hexaware saw:

• False positives reduced by more than 75%: Elastic Observability has transformed monitoring accuracy.
• Achieved accuracy of 92%: Hexaware reduced a client’s MTTR by 40% and achieved a 92% accuracy in incident matching.
• Increased productivity: New engineers to the team can contribute sooner, and senior engineers can concentrate on value-driving tasks.

Foundation for scalable AI adoption

For Hexaware, the Elasticsearch Platform — which includes Elastic Observability and the AI Assistant, among other out-of-the-box tools — provides a foundation for scalable AI adoption.

Unlike other application performance monitoring (APM) tools, Elastic offered:

• A fully configurable platform that could scale and evolve with the company’s needs
• Built-in unsupervised and supervised learning, third-party model onboarding, and integrated generative AI capabilities
• The power to move faster, allow for flexibility, and deliver more reliable services in complex enterprise environments

“Elastic features such as unsupervised learning, supervised learning, and third-party model onboarding and training are invaluable. You don’t get that with other products in the market.”

– Sanjesh Rao, SVP, Enterprise AI and transformation, Hexaware

Elastic’s integrated generative AI capabilities enabled conversational operations, automated charting, and faster incident response across the stack. Combined with Elastic Cloud’s autoscaling features that create self-healing patterns, Hexaware can deliver that “wow factor” for clients.