IT is Losing the Cloud Blame Game

Dennis Damen, Principal Product Manager at Nexthink, discusses the how and why behind the fact that IT is currently “losing the cloud blame game.” This article originally appeared in Insight Jam, an enterprise IT community that enables human conversation on AI.

IT has an accountability problem. In years gone by, when virtually all major digital assets were on-prem, IT leaders were responsible for and accountable for the smooth running of the IT environment. However, since the shift to the cloud, IT leadership has remained accountable, yet so much of the responsibility lies in the hands of a third party. Like being sideswiped by someone running a red light, CIOs today can do everything right and still find themselves in deep, deep trouble.

Some of the dangers have been mitigated by service-level agreements (SLAs), dictating that cloud providers must ensure that their platform is always available. However, while outages are a serious concern, they are far from the only issues that can go wrong. Take, for instance, contractors at a bank who use virtual desktops to connect to sensitive data and applications. While the cloud platform might technically be available, if it’s too slow to use, these workers can find themselves losing entire days of potentially productive time, causing them untold frustration and costing the businesses thousands of dollars.

To combat this, businesses are increasingly looking beyond SLAs to experience-level agreements (XLAs), which require not just that a system be available but also that users
have a minimum quality level when using the tool. But agreeing to an XLA and enforcing it are two very different things. So, how should organizations go about ensuring they’re getting what they pay for?

Did the Experience Measure Up?

There are many aspects to a good cloud experience—fast, bidirectional data migration, robust connectivity, and strong integrations. But while it’s easy to identify what is needed for a good experience, it’s far harder to break that experience down into its component parts, measure, and assess it. And without this, XLAs become functionally useless. Let’s extend our example with our banking contractor. She needs a virtual machine to securely access the system, and during that session, she needs to use, say, six different applications.

The applications she uses will greatly impact the performance of her virtual desktop and vice versa. At some point, her machine starts to lag badly. Clicks are slow, and she’s randomly disconnected from her session, dragging her productivity levels to near zero.

There is clearly an experiential issue, but where does the fault lie? It could be with the cloud provider, meaning that the XLA can kick in. But what if it’s the applications that are causing the problem? Or what if her home connection is weak because she’s working too far from the router? The cloud vendor isn’t likely to voluntarily demonstrate that their service is the problem, so the onus is on the company to find clear evidence that the XLA is being violated.

Getting Out of the Blame Game

In order to have clear evidence to trigger an XLA, organizations need a holistic view of application and desktop usage patterns (both virtual and physical). This means gathering not just technical data related to performance, availability, and stability but also experiential data regarding user productivity, usability, and adoption.

Having this level of detail across the entire IT environment allows businesses to identify root causes. Going back to our contractor, this may mean being able to show that she is not the only person having problems and that, rather than any particular application, the only common factor between her and other users dealing with bad virtual experiences is the cloud provider.

Equally, if the problem has nothing to do with the cloud vendor, the business can save time by not pointing fingers and instead can focus on addressing the problems with their internal configurations to solve the problem. Indeed, the ability to skip the blame game isn’t just important for reducing remediation time. By having proper visibility into where issues are occurring, customers and cloud vendors can reset and improve their relationship significantly. When there are clear experience quality benchmarks in place and customers can see that vendors are hitting them on a consistent basis, it builds trust and helps demonstrate the value of a specific provider far more than SLAs.

What Are We Paying For?

Ultimately, when businesses pay millions of dollars to cloud providers, they aren’t interested in a promise that the system will be always available—what they really care about is whether the provider can make their team more efficient and productive. And that comes down to the experience those employees get. Hence, the shift towards XLAs is inevitable.

But putting on in place and using it to effectively ensure accountability are two very different things. Every IT leader should keep in mind that, without proper visibility and reliable metrics across the entire environment, an XLA is really just a bit of paper.