Getting MTTR to zero: the failed promise of observability

There’s an old cliche about sales and jobs to be done - no one wants to buy a drill, they need a hole… actually, they want a home with pictures on the wall. To get to that beautifully designed home, they will buy a drill, make holes for brackets that can support their various artwork and family photos, and progress toward their dream home experience.

Similarly, no one wants to buy observability software. They want their mean time to resolve (MTTR) issues to be zero. They don’t want to have software or security incidents, and if they do, they want to find and fix them ASAP. Observability promised to be the drill, the tool to make that dream come true. But so far, most observability tooling has not lived up to that ideal.

What people are really looking for in observability

Most teams that discover they need an observability solution are looking for four key elements: proactive issue detection, reduced MTTR, improved performance, all with automation and workflows to make remediation easier.

• Proactive issue detection: the process to achieve this is monitoring. You’ll need real-time telemetry from your infrastructure and applications to tell you when something goes down or stops operating at an optimal level.

• Reducing MTTR: This inevitably comes back to faster root cause analysis. The faster you can figure out what happened, the faster you can fix it.

• Improved performance and user experience: Teams don’t just want to release new features and move on. They need to know those features are working well and users are having positive experiences. With those insights, organizations can improve and continue to delight their users.

• Better remediation of issues, including proactive methods: Remediation with workflows and automation, so when you find something that’s not working at its best, you can consistently fix it with minimal expert intervention.

But all of these things point back to the true goal of MTTR of zero. Proactive root cause identification helps accelerate the investigation part of the process. Rapid root cause analysis is vital for incident response, plus it informs the last stage of remediation, which should be fast and reliable. And improving the customer experience shouldn’t bring new bugs or issues into the system.

See how Automation Anywhere is reducing MTTR with Sumo Logic.

How we build observability better this time

I’ve already called out some of the pitfalls of using a legacy approach to “Observability 2.0”. The market and experts agree that trying to use multiple telemetry types, while useful in certain scenarios, is not going to be the “golden signal” driving the new era of observability. For that, it comes back to logs. Specifically, it comes back to unstructured logs from security, infrastructure, and critical applications.

Sumo Logic has always been fueled by the atomic level of logs. Our emphasis on this most powerful and robust form of data helps our customers get to the root cause faster, consistently driving faster MTTR.

See how OpenPayd reduced their MTTR by 80%

By leveraging machine learning and generative AI over unstructured logs, we can both simplify our approach and finally deliver on the goals of observability:

• Resolving issues faster

• Maximizing your best talent

• Simplifying your tech stack

• Reducing overhead and instrumentation

The threshold for innovation now lies in how we can do that faster, more easily, and for even more customers. Of course, in the year 2025, that means we have to mention AI.

You can dive into my explanation of how layering AI on top of a system of record built on logs gets us to Dynamic Observability. But aside from the technical details, ultimately what this means for our customers is that more of your team can investigate and resolve incidents or threats faster, without interrupting your power users, and keep your teams focused on the work they enjoy most–innovating on interesting ideas instead of finding and fixing issues.

As we continue to build toward our vision of an AI-driven dynamic future, it’s vital that people realize how much of this is already possible. With powerful log analytics, you can work towards an MTTR of zero. You don’t need to hope for a better future for observability. It’s already here, and only going to get better.

See how reducing MTTR has a direct impact on ROI.