29 June 2025
Announcing IBM Instana’s latest AI feature, Intelligent Incident Investigation powered by agentic AI, now available in Preview. Instana already uses causal AI to identify the probable root cause of an incident. But now, with a single click, users can launch a full agentic AI investigation into the incident and Instana will immediately build a hypothesis to identify details of the problem, where it occurred and its impact, delivering complete knowledge of the incident in seconds.
With downtime costs exceeding USD 300,000 per hour for mid-size and large enterprises according to ITIC, seconds matter. Combined with the dynamic, distributed environments that many companies operate, including the use of microservices, containerization, cloud-native patterns, and hybrid and multi-cloud architectures, IT systems are now incredibly large and complex. In addition, these environments generate a huge amount of operational data making it hard to identify signals from the noise.
Full-stack end-to-end observability tools, like IBM Instana, include automation and AI capabilities that accelerate the identification, tracing, and resolution of issues and help reduce MTTR. But it still takes time and expertise to follow the thread of an issue to research and troubleshoot across your applications and infrastructure.
It’s 2am and your observability tool just detected a business-impacting incident. Your on-call SRE is paged out and opens their laptop. Their mission: figure out what the specific problem is, what caused it and then resolve it ASAP. Instead of jumping from dashboard to dashboard following the thread of the incident across applications and infrastructure, what if our SRE could quickly and easily use agentic AI to do the investigative work for them?
In Instana, the reasoning and actions of the AI agent doing the incident investigation are displayed in real-time, allowing the SRE to supervise its train of thought and validate the findings. Early users of Instana’s Intelligent Incident Investigation reported an increase in the time it takes to investigate incidents by up to 80% faster.
“We are excited to use Instana's new Intelligent Incident Investigation,” said Mahesh Billa at client zero IBM CIO organization. “This has the potential for our Operations team to go from incident detection to investigation to remediation in record time.”
After investigating the incident and verifying Instana’s hypothesis, it’s time to remediate. Instana uses agentic AI here too: using the incident investigation results, the SRE can evaluate a step-by-step runbook of actions created to remediate the incident. Now the SRE focuses their time on acting to fix the issue, while maintaining full control over what actions are implemented.
Not able to remember the arguments and pararmeters at 2am for a given action? You’re not alone. Instana helps here as well. With a couple clicks, a Bash script is created for each step, which can then be exported to GitHub for review, testing and deployment.
A critical but tedious task after any incident is to document exactly what happened and share with stakeholders. Instana takes care of that too: click “Generate a summary” and Instana will summarize the incident, including key information like root cause, business impact and the actions performed.
Built-in agentic AI provides faster, more comprehensive and more accurate root cause identification so you can obtain complete knowledge of the incident in seconds. With Intelligent Incident Investigation and remediation, you can:
IBM Instana provides complete visibility across applications, services, and infrastructure for rapid issue detection and resolution. Why Instana over other observability and APM tools?
IBM Instana Intelligent Incident Investigation is now in Preview. To test it out for yourself, sign up for the trial. Or for a full walk-through and to get your questions answered, book a live demo.