Turning observability into business value in the age of AI

For the modern enterprise, whether you succeed or fail directly depends on the reliability of your digital applications. But, as organisations rapidly integrate artificial intelligence, cloud-native architectures and automated agentic workflows, the digital estates we manage have grown significantly more complex.

 

Recent research across the EMEA region reveals the challenges organisations are facing. High-business-impact outages now carry an annual median cost of £79.9 million per organisation. The average cost of these outages is £1.5 million an hour. That’s an equivalent of £24,835 for every single minute systems are offline.

 

When a single minute carries this enormous financial impact and irreparably damages customer trust, the traditional, reactive approach to handling IT systems no longer works. Many businesses are trapped in a firefighting culture where they address system failures only after they have degraded the user experience. In the era of AI and hyper-distributed architectures, this reactive mindset is a drain on both business resources and engineering talent, diverting minds away from innovation and burying them in endless crisis management.

 

Technology leaders must adopt a new engineering mindset. We must shift our focus from merely reacting to system anomalies to proactively preventing them entirely. This requires a cultural and structural evolution where observability becomes a core part of the software development lifecycle from the very beginning.

 

However, a technical shift is only half the battle. True operational excellence occurs when technical metrics translate into business outcomes that executive stakeholders can instantly understand. Historically, a disconnect has existed between platform engineers reviewing complex trace logs and business leaders tracking customer conversion rates. To unlock the full potential of observability, this silo must be broken down.

 

Look at Hiscox UK. Tasked with consolidating its monitoring tooling to drive group-wide efficiencies, the team recognised that observability couldn’t just live within engineering teams; it had to deliver real-time, business-facing insights. 

 

By leveraging a flexible, data-first observability platform, Hiscox broke down traditional silos between technical and business units. It applied custom querying to design tailored dashboards that mapped complex technical architectures straight to core business outcomes.

 

The value of this approach was proven almost immediately. When the business required early warning capabilities against malicious bot price-scraping, the team deployed a specialised dashboard within an hour, exposing abnormal traffic patterns and protecting the business’s data assets in real time. Furthermore, by tracking the execution times of end-to-end workflows, Hiscox has been able to establish performance benchmarks rooted in actual system behaviour, allowing executives to instantly validate the commercial success of digital enhancements and system upgrades.

 

This is where modern observability shines. It transforms raw telemetry data – the millions of logs, metrics and traces generated by applications and AI models – into an open, composable tool for business intelligence. It empowers engineering teams to establish automated alerts and anomaly detection that flag potential failures before they ever cross the threshold to the end-user.

 

As enterprises push further into the intelligence era, the winners will not be the organisations that collect the largest volumes of uncompressed data points. The winners will be the leaders who apply intelligence to their data. By committing to a proactive, business-aligned observability strategy, enterprises can safeguard their systems, shield their reputations, and maintain operational resilience. 

---

Read New Relic’s 2026 AI Impact Report to discover how enterprise teams are leveraging AI-driven observability to reduce alert fatigue, optimise workflows and reclaim their engineering productivity

---