3 min read time

The Do’s and Don’ts of Observability

by   in IT Operations Cloud

Observability

With the seismic lift and shift to the cloud, a mighty gust of complexity has blown into ITOps. On top of our existing complexity, we now have more systems, higher costs, and growing numbers of breaches and outages to deal with. Luckily, observability can calm things down.

If you’re not sure what “observability” is, it’s a new approach to monitoring complex, dynamic environments both off cloud and on SaaS. To start gaining its benefits, keep it simple. Observe these two do’s and two don’ts.

Two Do’s

1.    Do tie observability to business context.

According to Digital Enterprise Journal, 68% of IT’s time is spent on tasks that don’t contribute to key business outcomes. Teams lack the bandwidth to do much else. But connecting the dots between technology projects and business outcomes is critical to elevating IT’s value across the enterprise. Given the market confusion surrounding observability, the best way to show its value is to link it to business context.

Sharing visual dashboards and reports is an easy way to do that. You know the old saying, “A picture is worth a thousand words.” So highlight important metrics in tables, charts, bars, and pie charts. Democratize data with internet news feeds, live programs, streaming videos, availability status, and customer experience metrics. These relatable formats help the business understand IT’s strategic value.

Take the early days of the pandemic, for example. At that point, most organizations pivoted to a work-from-home model, which created an urgent need for ITOps teams to communicate the live status of VPN health, online business services, and collaboration tools. With the Business Value Dashboard in Operations Bridge, they were able to deliver high-level views of key remote application services in real time.

“Today, we use a sophisticated solution with more than 20 dashboards for different roles to support dynamic application change and real time data analytics.” – IT Director, Turk Telekom

Using the Flex Designer in Operations Bridge, you can easily design dashboards and reports that tie observability to business context. Learn about Flex Designer in the 2022.11 release.

2.    Do lead with customers in mind.

If you’re not sure where to begin your observability journey, start with customer needs. What services are most important to them? What level of service speed or uptime do they expect? For example, a common SLO is availability—meaning that services are up and running when customers need them. If all your customers are in the same time zone, then service availability outside their active hours won’t matter as much to them.

When you lead with customers in mind, your SLOs will serve as meaningful guideposts, and you won’t spend time on unimportant tasks.

 

Two Don’ts

1.    Don’t keep adding more monitoring tools.

A study conducted by Automic found that 53% of large IT organizations use 20 or more tools. An unnecessarily high number of IT tools—each one for a different issue—creates tool sprawl. Nothing good comes from that—only poor visibility, high inefficiency, collaboration breakdowns, and lots of frustration.

No shiny new tool can make observability magic happen. David Linthicum discusses our propensity to keep adding tools in Observability Trends and Pragmatic Techniques to Optimize Multicloud Operations. He suggests that before adding a new tool to your collection, evaluate and integrate your existing tools. Then, if you’re still convinced the new tool is needed, make sure your ITOps team can use it effectively.

2.    Don’t do observability half-heartedly.

True observability is not possible when there are data or team silos. Even when all your data is in one place, in one console, if teams aren’t working together to solve issues, your observability efforts will be subpar.

To break silos down, consider turning to DevOps. DevOps is a means for ITOps teams and developers to work together, iteratively. Observability fits in well with the DevOps model. It enables Dev and ITOps teams to collect, correlate, and analyze massive amounts of cross-systems data. And as a result, they are empowered to monitor effectively and deliver great customer experiences. Half-hearted efforts can’t achieve that.

 

Observability—the New Path Forward

Traditional monitoring tools are no match for today’s IT complexity—and the breaches and outages that come with it. But by following these simple do’s and don’ts, you can gain the observability needed to predict and remediate problems proactively, even before they have a chance to occur.

Labels:

Operations Bridge