top of page

How to forecast multi-cloud costs

Business woman writing on a transparent board

Analyzing multi-cloud costs is essential to understand the necessary resources for operations and forecast cloud cost to establish the budget to allocate.


Within the business context, managing multi-cloud costs is an essential practice used to avoid excessive consumption and to estimate future management expenses.

In the previous article, we looked at three aspects to consider when managing multi-cloud costs. Now as a next step, it is necessary to analyse the associated data in order to understand both how much budget to allocate to subsequent years and to know and monitor the state of the infrastructure.

Generally, it's good practice to combine an analysis of past expenses with future plans, considering potential changes in planned cloud infrastructure, the lifecycle of current applications, and cloud investment plans to establish current and future budgets.

This process requires close collaboration among various Finance, IT, and Management teams. The goal is to develop shared prediction models and Key Performance Indicators (KPIs) as a basis for establishing budgets in line with business objectives.

Unfortunately, there is no universal prediction method that fits all organizations. Cloud spending is inherently volatile and can be challenging to forecast due to its variation based on processes, applications, and the company's structure itself.

For instance, engineers might initiate workloads at any time, often without following a standard provisioning process. This need for dynamism adds a level of complexity to cost forecasting, making it a constant challenge in the cloud computing environment.

Maturity Level

Primarily, to turn this practice into a strength, the first step is to analyze your current situation to understand your organization's level of maturity.

For a long time, we've managed our clients' cloud costs and over time, we've developed guiding questions to help, divided into four main areas:

1) Forecasting processes and tools used:

  • How diverse are the data sources and tools used for cloud cost forecasting?

  • Are forecasts primarily created manually or according to a pattern?

  • Is there an analysis methodology based on past trends for predictions?

2) Stakeholder involvement and forecast review:

  • Do team managers have access and involvement in cloud cost predictions?

  • Is there regular review of forecasts with stakeholders?

3) Accuracy and alignment with business objectives:

  • Do the forecasts consider cost adjustment based on actual usage and are they used to define budgets?

  • Are plans reviewed and updated periodically?

4) Data optimization and integration:

  • Are there automatic data flows between cloud cost predictions and backend accounting systems for broader organizational reporting?

Answering these questions will help you understand your company's level in cloud cost prediction. It's generally categorized as basic, intermediate, or advanced:

  • Basic: Use a variety of tools and data sources for cloud cost forecasts, though they are largely manual and based on past trends. Variance analysis is manual with limited visibility, focused only on specific business units or cost centers. Lack of involvement from various teams limits monitoring of deviations from spending forecasts.

  • Intermediate: Cloud cost forecasts are optimized and include trend-based models. Forecasts are regularly updated and used to set budgets. Stakeholders have access to predictive data, integrated with backend accounting systems, and FinOps teams regularly review forecast thresholds and trends with stakeholders.

  • Advanced: The organization comprehends global policies for metadata allocation and monitors cloud cost forecasts in real-time, adjusting them based on cloud usage. Forecast models are highly sophisticated and aligned with organizational allocation constructs, providing detailed visibility for business units, teams, and products. Data integration and flow are completely automated, providing a single source of truth for monitoring and influencing cloud cost forecast trends and budgets.

According to research conducted by the FinOps Foundation, currently, most companies are at the basic level, with 48.2%, 32.8% at the intermediate level, and only 19% have implemented automation to reach an advanced level.

Success Measures in Cloud Costs

To evaluate the success of predictions, there are Key Performance Indicators (KPIs), Objectives and Key Results (OKRs), and thresholds defining acceptable deviations. FinOps practice defines a set of success measures to classify and allocate cloud costs.

These indicators vary greatly based on your level of operational maturity. Below are the main factors to consider when creating cost predictions for your company:

  • Cost Classification and Allocation: According to FinOps practice, complete allocation is achieved when at least 80% of cloud spending is allocated to a specific maturity level FinOps practice, such as scanning level (80%) or Run level (90%).

  • Optimized Forecast Models: Forecast models use adapted cloud usage data for discounts, facilitating more accurate cost prediction.

  • Prediction Accuracy: Acceptable variances are established compared to actual spending, with specific accuracy levels for different maturity levels.

  • Forecast Notifications and Updates: Stakeholders are notified if the forecast deviation thresholds are exceeded, reducing the risk of excessive budget consumption.

  • Forecast Frequency and Updates: Forecast frequency includes intermediate updates for any budget adjustments.

  • Business Unit Responsibility: Business units should have the freedom to manage their budgets based on predictive data, thus improving responsibility and alignment with spending expectations.

Data Optimization for Efficient Management: Your Ally in Cloud Costs

As you may have inferred, data optimization is crucial for effective cost management in the cloud.

However, this process often requires an investment of time and energy, involving a long series of manual activities.

By choosing the Xautomata platform, you'll have a powerful tool that frees you from tedious manual tasks, allowing you to focus on your company's core business.

The solution integrates data from various providers and aligns allocations, providing a unified and real-time view of costs directly on a single dashboard.

In addition to simplifying prediction, you'll obtain a detailed overview of anomalies, trends, and advanced forecasting, empowering you to make data-driven decisions.

Choosing Xautomata means opting for a more efficient and targeted method in managing your cloud costs.


bottom of page