Do you really know where your cloud dollars are going?

Questions and guidance on managing your cloud spending

Cloud is an invaluable IT resource, but do you know if your organization is spending more than what it’s worth? Are your cloud expenses soaring? Typically, an estimated 30% of an organization’s cloud spend is not utilized, adding up to more than $17 billion in wasted cost. Do you have the ability to monitor and manage hybrid and multicloud acquisition and usage? Without true cross-functional responsibility involving ongoing collaboration between finance, IT, operations, and business users, it’s virtually impossible to optimize your use of cloud services.

Why are your cloud dollars going down the drain?

Many organizations that are experienced in planning for data center migrations are unprepared to optimize their cloud migrations. You may have an entire ecosystem built around data center migrations, with checks and balances built in to assess provisioning and utilization; However, utilization patterns are vastly different for the cloud, where any part of an organization can spin up services on demand, and in some cases never turn them off even when usage ceases permanently. In our survey of clients, two-thirds say they are not fully realizing the benefits of their cloud spend. Virtual machines and instances are being paid for by the hour, the minute, even by the second, but are only used during a 40-hour workweek. So, you may be still paying for those resources during the other 128 hours in the week as they sit idle.

Overprovisioning is one way in which you may be seeing precious budget dollars being wasted. Some 40% of instances are sized at least one size larger than needed.

Just by reducing an instance by one size reduces the cost by 50%. Downsizing an instance by two sizes saves 75%, so enormous changes can be made to your overall spend. You may be able to cut costs by identifying other sources of waste and inefficiency, such as orphaned volumes, inefficient containerization, underutilized databases, instances running on legacy resource types, and unused reserved instances. You may be paying for 50 Software-as-a-Service subscriptions, but in actually only need a handful to meet your needs. It’s even possible that different parts of a business have each separately contracted for an enterprise license of a vital application, so instead of one enterprise license you could be paying six times for the same functionality!

These management issues are further complicated when you run hybrid and multi-cloud environments where it is difficult to determine best-fit workloads. That starts with determining which cloud provider is right for a particular workload. Just because your current cloud provider seems well able to keep your financial applications up and running, that doesn’t mean it can provide the best environment for a new workload that is heavily dependent on Artificial Intelligence (AI) and Machine Learning (ML). There are many difficulties in planning and allocating budget for your cloud consumption, ranging from business unit independence to lack of visibility into used services at your cost center, workload, and application level. Shadow IT continues to be a challenge for IT planners—when Software-as-a-Service (SaaS) subscriptions may be tucked away on individuals’ expense reports, there’s little transparency into overall spending, let alone utilization.