Blog

5 Cloud Economics Mistakes to Avoid

~5 minutes read

Companies adopting cloud architecture are still thinking that they’re “owning” IT instead of “consuming” it. As a result, businesses develop strategies, negotiate contracts, and make economic decisions without taking into consideration that the financial and management approaches are different for on-premises and cloud architectures. Not only does it make organizations believe that the cloud migration doesn’t worth it, but even makes some companies reverse their course and go back to traditional IT.

In this article, we discuss five of the most persistent, relevant, and crucial cloud economics mistakes that companies usually make. Additionally, we explore the possible solutions to the financial issues in the cloud.

Mistake 1: Keeping in mind CAPEX instead of OPEX

Issue

On-premises software operates according to a capital-expenditure model, where data-center capacity is built with a view to the distant future. In this model, the cost of additional resources (e.g. capacity) is minimal, and companies measure their cost efficiency and allocate future budgets by looking at the average cost and resource utilization rates.

Cloud computing has changed the capital-expenditure model (CAPEX) to the operating-expenditure model (OPEX) due to cloud infinite scalability. OPEX implies that the companies pay for what they actually consume or, to be more clear, for what they subscribed to.

All in all, nowadays the most efficient cloud economics is grounded on the ability to accurately evaluate the demand, as well as the corresponding marginal costs at any given moment. In fact, the organization should pay for resources only when it needs them, rather than paying for unused assets, letting the budget go down the drain.

Solution

Companies should develop a dynamic operating-expenditure approach to cloud economics that instantly optimizes supplementary costs. This strategy means that the business owners and IT departments choose cloud services that match the current workload requirements.

For example, a media company can scale up its compute capacity ahead of major customer promotions or special occasions like holidays to meet the demand and be able to handle increased user traffic. After the promotion ends or the holiday season comes to the end, the company scales the compute capacity down and easily avoids extra cloud spending.

Mistake 2: Forecasting cloud spending based on the past utilization only

Issue

In the context of the capital-expenditure model used in traditional IT, it was normal to plan the future budget, as well as allocate resources keeping in mind the past consumption level and costs. This approach hardly fits the operating-expenditure world of the cloud, as the cloud nature is pretty much fluctuant.

Additionally, the traditional IT method to forecast future expenses is a big issue in the cloud architecture perspective, as it’s more difficult to estimate the cloud spending and make allocations to support new cloud-based products. Forecasting according to the capital-expenditure model often results in a greater than 20% mismatch between the planned and actual spending and leads to rapid and stressful rebudgeting.

Solution

In order to improve the forecasting and budget planning experience companies should consider the business goals and priorities. For example, if an organization is planning a huge promotion tied to Cyber Monday, it will probably see a rise in customer interest. So, the company should think ahead of the future and think of the additional resources, and the possible cost spikes as well.

Since the cloud costs depend on the usage, the supplementary resources required during the busy seasons can cost a company an arm and a leg. Establishing the unit economics (such as compute cost per customer) for the major applications can help in this situation. This approach requires a mindset shift toward a consumption model. This way, the company owners better understand the business drivers of their cloud spend and its corresponding impact on unit economics.

Mistake 3: Considering only the benefits of cloud elasticity

Issue

The elasticity and scalability of the cloud are economically perfect for workloads with flexible cloud utilization patterns. Unfortunately, the majority of companies suffer from the constantly increasing costs caused by the inability to analyze and predict the application demand that would match the real one. So, most companies end up having overprovisioned resources, which leads to wasting money.

Here is another problem: if we take a video-streaming company as an example, we’ll see that the storage consumption is steadily increasing as the number of subscribers grows. It’s great, as the storage is used to the fullest extent. However, the continuous growth in subscriber data means continuous increases in storage costs as well.

Solution

Considering the case of underutilization, companies need to pay close attention to the examination of their workloads which can help to assess whether their elasticity patterns would lead to savings on the cloud or not. If the cloud elasticity drains the company’s budget, the cloud cost monitoring tools are a must. Only with a detailed view of the asset usage and corresponding expenses the business can evaluate the excessive resources and downsize them.

If we’re talking about exponentially rising costs, greater visibility can help as well. You can’t control what you can’t see. So, constant monitoring of the expensive resources should be the companies’ highest priority, too.

For instance, with Binadox (cloud asset management and cost optimization platform) you can see the whole picture of your cloud resources usage and the costs. Also, you can create the rightsizing recommendations based on the current workload, as well as adjust the system setting up the planned workload to receive more accurate recommendations.

Mistake 4: Migrating all the workloads to the cloud

Issue

One of the points why companies move to the cloud is the cost-saving opportunities that cannot be achieved within the on-premises architecture. That’s why some organizations shift the entire existing infrastructure and services to the cloud. However, it’s vital to remember that not all business applications are suitable for migration to the cloud.

Sometimes running apps in the cloud is not physically possible. Other times, it’s not financially wise to run in the cloud, as the cost of rearchitecting may be really high.

Solution

As InformationWeek wrote:

“Not all business applications should migrate to the cloud, and enterprises must determine which apps are best suited to a cloud environment”.

Organizations should compare the cost of the app deployment within the traditional infrastructure, as well as the cost of migrating this app and all corresponding data to the cloud.

The applications that shouldn’t be moved to the cloud, may be run on the company’s custom-designed on-premises infrastructure. The scale and homogeneity of these workloads may create on-premises economics that is equivalent to or better than those offered by cloud providers.

Mistake 5: Ignoring security threats and compliance requirements

Issue

The more data and services are stored in the cloud, the more chances of security issues occurring. Although awareness about cloud security aspects is growing, more often it is challenging to identify and remediate these issues.

The consequences of the security and compliance violations are tangible – they include huge fines, reputational damage, and even lawsuits. So, organizations don’t want any trouble with cloud security aspects at all.

Solution

Companies should design a cloud security strategy, as it leads not only to the trust issues from the client side but also may cost a company a lot of money. It’s better to start with the identification of the potential security threats, such as data breaches, account hijacking, unauthorized access, data loss, and vulnerabilities of applications and systems. When the weak spots are discovered, the organization should solve the issues as soon as possible. Read about the most widespread security concerns and their solutions in our article “Shared Responsibility Model for Cloud Security”.

Besides security questions also should be taken into consideration. You can read about the compliance regulations and the best practices in our previous article “Cloud Compliance Regulations and Best Practices”.

To sum up

These cloud economics mistakes are familiar to businesses of all sizes. When migrating to the cloud, and in the post-migration state, it’s essential to have a clear picture of the costs and hurdles you will have to overcome to enjoy all the benefits of the cloud.

When only planning the migration, ensure you find the right cloud vendor — the one that follows all the industry regulations and offers the services that suit your budget and business needs. Once this step is completed, think about the practices you will use to manage your cloud assets.

The sooner you’ll start optimizing your cloud costs, the better for your company’s well-being and development. Binadox offers real-time cost data, as well as instance resizing recommendations, immediate security notifications, tagging, and automation rules features.

Go Up