Skip to main content

Part 2: Using FinOps

Cloud is not just a hosting platform, but instead it is a business accelerator and FinOps is not just cost optimization, but about value maximization.

Cloud value gives better products which gives improved business outcomes which gives an increased use of the cloud which gives....better products :)

Cloud computing is OSSM: On-Demand, Self-Service, Scalable, and Measurable. This means that cloud resources can be provisioned and deprovisioned quickly and easily, allowing for greater agility and flexibility in how organizations use the cloud. However, this also means that it can be easy to lose track of cloud spending and usage, which is where FinOps comes in.

This allows everyone to deploy cloud resources without going through procurement approval processes. Implementing a micro-approval process is a nogo too.

Traditional data centers:

  • CapEx: Capital Expenditure - upfront costs for hardware, software, and infrastructure.
  • OpEx: Operational Expenditure - ongoing costs for maintenance, power, cooling, and staffing.
  • capacity management: requires forecasting and planning for future needs, which can lead to overprovisioning or underprovisioning.
  • cost paid upfront and then amortized over time.
  • reviewed monthly or quarterly, because the costs were basically the up front divided by the term.

Cloud computing:

  • OpEx: Operational Expenditure - pay-as-you-go model based on actual usage.
  • capacity management: allows for dynamic scaling based on demand, reducing the risk of overprovisioning or underprovisioning.
  • costs are variable and can fluctuate based on usage, making it important to monitor and optimize cloud spending in real-time.
  • busy periods can lead to higher costs, while idle periods can lead to lower costs, so regular monitoring and optimization are essential to manage cloud costs effectively.
  • billing is not simple to understand, and can be complex due to the variety of services and pricing models offered by cloud providers, making it important to have a clear understanding of how costs are incurred and how to optimize them.
  • Reviewing costs in anything other than real time leads to missed opportunities for optimization and cost savings, as well as the risk of unexpected cost overruns. (ANOMALIES :)

Not adopting FinOps

In the cloud, your engineers can spin up millions of dollars worth of resources in minutes, and if you wait until the end of the month to review your bill, you may find that you have incurred significant costs without realizing it. This can lead to budget overruns and financial strain on the organization.

3 ways to transform the way your organization works with the cloud:

  1. focusing investments on business where the cloud can enable increased revenues
  2. Selecting a technology model that aligns with the business needs and goals
  3. Developing and implementing an operating model that is oriented around the cloud.

PoC Stall - MVP's are built but stall because of a lack of business case Cloud Gridlock - initiatives become jammed up in ques because IT cannot build out the architecture or automation to use services quickly Lack of Value from Lift and Shift - Moving existing applications to the cloud without optimizing them for cloud-native features and capabilities can lead to suboptimal performance and increased costs, resulting in a lack of value from the migration. Cloud chaos - Uncontrolled cloud spending and resource sprawl due to lack of visibility and governance.

Quick wins are necessary until the organization develops confidence in FinOps and the business value of the cloud. Every time a FinOps initiative results in improved business value, it gives both the cloud value and the FinOps value, providing that it is communicated to the stakeholders that are investing in the cloud and FinOps.

in addition to the cycle of better products leading to better business outcomes, which leads to more cloud value, which leads to better products, there is also a cycle of more IT innovation which leads to more investment in FinOps which leads to a stronger FinOps capability as well as more value improvement which leads to more IT innovation.

Ignoring​

Ignoring FinOps is where an organization doesn't do any measurement of spend or usage, and therefore has no visibility into their cloud costs. This can lead to significant financial risks and missed opportunities for optimization. There may be off the cuff spikes and reviews, but nothing set in any sort of regular cadence, and no real data to make informed decisions.

Scenario: We need this app, cost is not a concern, we will just pay the bill at the end of the month and revisit later. This is a common scenario where an organization may prioritize the need for a particular application or service over the cost implications, but it can lead to unexpected financial strain if the costs are not monitored and managed effectively.

The result of this is that costs are looked at once, showing $5000 in spend, deemed acceptable, and then ignored until the next review, which may be couple years later, at which point the costs may have increased significantly ($40,000) without any opportunity for optimization.

This is where the "Spend Panic" comes in, where an organization realizes that they have incurred significant costs without realizing it, and they may panic and make hasty decisions to try to reduce costs without fully understanding the implications. This can lead to suboptimal decisions and further financial strain on the organization.

Informed Ignoring

Informed ignoring is where an organization has some visibility into their cloud costs, but they choose to ignore it because they believe that the costs are justified by the value that the cloud is providing. This can be a valid approach in some cases, but it can also lead to missed opportunities for optimization and cost savings. The company watches their bill increase, forcast growth based on it and find some low hanging fruit to optimize, but ultimately they are comfortable with the costs because they are getting some value from the cloud.

This is the phase where the organization is starting to understand the costs and value of the cloud, but they may not have fully embraced the principles of FinOps yet. They may set up the potential to do some active FinOps, reporting, visibility and cross teams. But they may not have the culture or processes in place to fully optimize their cloud spending yet.

This is also where the engineering team is aware of costs, the finance team is aware of specific levels of costs and executives may make decisions based on the raw data available. Engineers will start to deploy cost efficiently, and the finance team will start to understand where the money is spent and the business will start to ask for more optimization, the start of a feedback loop, if you will.

Tagging​

Tagging is more difficult to implement after the fact - You have millions in spend and you're constantly trying to find ownership and accountability, but you don't have the data to do so. Implementing tagging from the beginning allows for better visibility and management of cloud costs, as well as easier identification of optimization opportunities.

Cost allocation without tagging is simply not possible, and one of the most important aspects of FinOps is being able to allocate costs to the appropriate teams and projects. Without proper tagging, it can be difficult to determine who is responsible for which costs, and this can lead to confusion and inefficiencies in managing cloud spending.

Lift and Shift​

Lift and Shift is where an organization migrates their existing applications and workloads to the cloud without optimizing them for cloud-native features and capabilities. This can lead to suboptimal performance and increased costs, resulting in a lack of value from the migration. This is a common scenario where organizations may prioritize the speed of migration over the optimization of their applications for the cloud, but it can lead to significant financial strain if the costs are not monitored and managed effectively

The wins of Lift and Shift fade over time, as the lower upfront cost, initial migration provide great value, but 2 years down the line, those heavy VMs are still running tasks where optimized and efficient serverless or containerized solutions could be used, and the costs could have decreased significantly with the ability for opportunity for optimization.

The core concept of informed ignoring is that it is fine for your organization to choose not to spend time on reducing your cloud spending. But there needs to be awareness within the business to know the implications of ignoring cost and FinOps altogether.

Conclusion

To summarize:

  • Cloud spend has—or soon will have—a major effect on organization balance sheets.
  • A lack of investment in FinOps early on can cause cloud adoption to hit headwinds or stall.
  • The procurement team no longer has control of the spending. In the cloud, this power has been pushed to engineers.
  • FinOps allows you to operate at the per-second speed of the cloud rather than relying on traditional monthly or quarterly spend reviews, which allows you to avoid unexpected costs.
  • Start with an informed ignoring approach before you are ready to perform all the domains of FinOps for an active FinOps practice.