Skip to main content
Welcome. In this lesson we cover two essential FinOps concepts for understanding cloud spend: fully loaded cost and blended rates. These help you see the full economic impact of running services, improve budgeting, and enable fair chargebacks across teams. To make this tangible, imagine you run a cloud-based pizza delivery app. The app uses compute, databases, storage, and many supporting services. Your cloud bill includes more than just CPU minutes — to understand the true cost you must consider the fully loaded cost: every layer that contributes to the service. In the pizza-shop analogy, monthly costs don’t stop at ingredients. You also pay rent, salaries, software subscriptions, and training.
The image shows an iceberg diagram labeled as "Fully Loaded Cost and Blended Rates" with "Pizza Ingredients" above the water and "Rent" and "Staff" below the water, indicating visible and hidden costs.
Cloud costs follow the same layered structure. Below is a concise comparison you can reference when categorizing your expenses.
Cost CategoryWhat it includesExample (pizza analogy / cloud)
Direct costsResources directly billed for running the workloadPizza ingredients / Compute, Storage, Network, Databases
Indirect costsSupporting infrastructure and platform servicesOvens, maintenance / Monitoring, Backups, Security services
OverheadOrganizational and shared expenses allocated to teamsRent, staff, training / Support plans, Governance, Tooling, Certifications
The image is a diagram titled "Fully-loaded Cost Components," categorizing costs into Direct, Indirect, and Overhead Costs with examples like Compute, Support Plans, and Training.
Why fully loaded cost matters
  • Teams that only measure raw compute will often underestimate the true cost of running a service.
  • Including indirect and overhead costs gives product owners and finance a realistic unit cost to evaluate ROI and pricing decisions.
  • Fully loaded cost enables defensible internal chargebacks or showbacks and reduces cross-team disputes.

​
Blended rates: definition and example

A blended rate is the weighted average unit cost (for example, cost per hour or cost per vCPU-hour) across different purchase types and instance families. If your pizza app runs On‑Demand, Reserved, and Spot instances, each has a different hourly price. The blended rate answers: what is the average cost per hour across all of that usage? Example scenario (rounded for clarity):
Purchase typeTotal costHours usedImplied unit cost
On‑Demand$4,0004,000 hrs$1.00 / hr
Reserved$2,4004,000 hrs$0.60 / hr
Spot$4002,000 hrs$0.20 / hr
Total$6,80010,000 hrsBlended = $0.68 / hr
Blended rate calculation (Python example): 
# Example blended rate calculation
costs = [4000, 2400, 400]      # costs for On-Demand, Reserved, Spot
hours = [4000, 4000, 2000]     # corresponding hours
total_cost = sum(costs)
total_hours = sum(hours)
blended_rate = total_cost / total_hours
print(total_cost, total_hours, blended_rate)  # 6800 10000 0.68
What the blended rate tells you:
  • The blended rate ($0.68/hr in this example) is the true average hourly cost across the mixed purchase strategy.
  • Use it as a single planning metric for forecasting, budget sizing, and making trade-off decisions between purchase types.

​
How fully loaded cost and blended rates help (key benefits)

  1. Optimization and planning
    • Use blended rates to identify when On‑Demand usage is inflating your average cost; plan more Reserved capacity or safe Spot usage where appropriate.
    • Fully loaded cost helps determine whether a business outcome (e.g., revenue per order) justifies the infrastructure expense.
  2. Accurate budgeting and forecasting
    • Blended rates smooth pricing and usage variability, making forecasts more stable.
    • Rising blended rates can be an early signal of shifting consumption or pricing changes that require attention.
  3. Fair cost allocation and accountability
    • Apply consistent unit rates for chargebacks or showbacks so teams are accountable for resource use.
    • Distinguish storage-heavy vs compute-heavy products and allocate indirect/overhead costs fairly.
Not all cloud providers expose a ready-made blended rate in every billing view. You may need to collect usage and pricing data, centralize it, and compute blended rates using a cost platform, analytics pipeline, or spreadsheet.

​
Practical tips for implementation

  • Aggregate usage and cost by consistent units (hours, vCPU-hours, GB-months) before averaging.
  • Include indirect and overhead allocations (e.g., a percentage of support plans, tooling, and training) in fully loaded calculations.
  • Recompute blended rates regularly (monthly or weekly) to capture changes in instance mix or pricing.
  • Use blended rates for cross-team showbacks, but retain detailed breakdowns for teams that need optimization signals.
Summary: Treat cloud spend like a business line — calculate fully loaded costs and blended rates to remain profitable, forecast reliably, and optimize your resource mix without surprises. References and further reading That’s it for this lesson. Thank you for reading.

Watch Video

Practice Lab