The growing adoption of FinOps practices for companies around the world can be a double-edged sword. While it’s obviously a big benefit to have cross-functional teams that can plan, execute, and monitor cloud spend in alignment with larger business goals, the actual steps required to establish a FinOps discipline and gain traction with it can be overwhelming.
Part of the reason for this is because FinOps requires buy-in from multiple teams within the organization – we explored the best way to get your engineers on board in a previous blog post – but also because it can be challenging to get cost optimization procedures up and running. Even something as simple as understanding who should be involved in these strategies can lead to confusion within an organization.
Who needs to care about FinOps?
To better understand the challenges, let’s first take a look at which personas within the company should have a stake in establishing these practices. Given that FinOps functions as a bridge between business, IT, and Finance teams, it’s critical that leadership for each of those departments has a sense of ownership in setting up the strategy, and helps to establish clear lines of communication around the company’s cloud spend.
On the business side of the house, the executive leadership must champion the idea throughout the organization and ensure that everyone is working towards the same goal. The finance department is likely already responsible for budgeting and forecasting overall costs for the coming months and years, so their ultimate goal is to understand and accurately track the cost of cloud expenditures to ensure that the money is being spent wisely. And of course, the IT/engineering/operations side of the house is likely responsible for actually forecasting, provisioning, and utilizing the cloud environment(s), so they must be able to communicate the company’s needs to the rest of the teams and justify the spending decisions that are being made.
Source: FinOps Foundation
Additional roles like product owners and procurement should also have a stake, but at the center of it all is the FinOps practitioners. This person or team – regardless of whether they have “FinOps” in their job title – is ultimately responsible for bridging that gap between the departments listed above, and establishing a culture wherein every team has a playbook of best practices that help optimize cloud spend and maximize the return on investment.
However, getting the right people on board is only the first small step in the FinOps journey. To properly establish the right culture, you must have a clear set of goals in mind. The FinOps Foundation breaks these out into specific domains, each of which is composed of separate capabilities (these capabilities often overlap across domains).
The six FinOps domains are:
- Understanding Cloud Usage and Cost
- Performance Tracking & Benchmarking
- Real-Time Decision Making
- Cloud Rate Optimization
- Cloud Usage Optimization
- Organizational Alignment
Within these domains, there are several capabilities that are likely already being done to some extent within your organization. For example, managing commitment-based discounts is something that most companies already do long before they have an established FinOps practice; it’s also a critical component of Performance Tracking & Benchmarking, Cloud Rate Optimization, and Organizational Alignment. The same can be said for other capabilities like managing anomalies, measuring unit costs, and data analysis & showback.
Are you really ready to take the plunge?
Taken together, these domains and capabilities can be viewed through the lens of the FinOps Maturity Model, which is based on a crawl-walk-run framework and sample goals that can give companies a better idea of the overall scope of the problem, and where they stand at the outset of this journey.
Yet for those who are in the crawl level of the maturity model, even this framework can leave you scratching your head and wondering where to begin to get your arms around everything. This is especially true for smaller or younger companies that may not have the requisite tools or expertise in place to properly execute these strategies – and if they do, they still might not have enough time or human resources to dedicate to tackling the problem.
Simply put, you may not be ready to fully embrace a top-down FinOps practice across your different teams and stakeholders. There is still a chance that the time and money required to set up this practice might not have sufficient ROI in the long run, which will ultimately lead to not only wasting time and money, but potentially damaging your long-term business success by diverting resources away from where they’re most needed. The last thing you want to do is establish a company-wide project that ends up costing more money than it saves.
How, then, to proceed?
Cloud cost optimization can offer a quick win
It goes without saying that the easiest way to start building advocacy and momentum for a new initiative is to show tangible results; if you’re able to generate some quick wins in the early phases, it builds credibility among the wider organization and lays the foundation for additional steps to be taken down the road.
In the case of establishing a FinOps practice whose whole point is to optimize the processes around purchasing and consuming cloud resources, the best way to do that is to focus on cost optimization. Lowering the company’s monthly cloud bill by even a small amount will not only help prove that this endeavor is worth undertaking, but it can also serve a foothold into the larger domains and capabilities that you’ll be focusing on as you continue.
Since compute costs often comprise well over half of a company’s cloud bill, this area serves as the biggest opportunity to lower your overall cloud expenditures. And while cloud providers like AWS and Google Cloud offer discounts in exchange for long-term commitments, the actual process of taking advantage of those offers is much more challenging than it may seem.
For one, the nature of a compute commitment evokes the old-school approach of reserving dedicated space in a data center and losing the flexibility to spin up new instances that may be in a different region or machine family than your normal usage. Both AWS and GCP have multiple types of compute commitment plans that offer varying levels of flexibility with the general rule that the more specific you can be, the greater your discount percentage; you’re usually trading flexibility for cost savings. Of course, you also may not know how much flexibility you’ll need over the course of the commitment.
The truth is that managing a cloud commitment portfolio can be enough work for a full-time job or even a team of people. Not only are there many different factors that one must consider when forecasting compute usage – e.g. machine types, regions, cloud services, etc. – but the ongoing tracking of usage and expiration dates across multiple commitments and multiple teams is a constant, year-round headache. And even if you’re able to get your hands around it all, there is still the underlying risk of committing to specific compute usage over a long period of time without any guarantee that your compute needs will be what you think they’ll be this time next year. If you overprovision, you risk paying for unused instances; if you underprovision, you risk leaving cost savings on the table.
Automating the solution
The answer to these management challenges – and the secret for your FinOps quick win – lies in automation. A tool like DoiT Flexsave™ was built with the understanding that cloud architects, product teams, and DevOps leaders need to focus on building their products and growing their businesses rather than poring over spreadsheets for hours every month to track and monitor their commitment spend. Meanwhile, the finance and executive teams need to understand how cloud resources are being consumed (e.g. What teams and applications are your biggest drivers? What cloud services are the most critical to the business?) so that those expenditures can be tied back to the wider business goals.
Flexsave solves the management headache by providing the flexibility of on-demand cloud compute with the benefits of commitment-based discounts. It does this through machine learning by analyzing your ongoing AWS compute spend and identifying workloads that are not already covered by existing commitments. Flexsave then automatically applies the equivalent of a 1-year Savings Plan to those on-demand workloads, thus removing the need for ongoing commitment tracking, accurate long-term forecasting, and the inherent risk that comes with purchasing resource commitments.
Of course, cloud technology covers a whole lot more than just compute, and FinOps is about a lot more than just cost optimization. To fully understand the scope of their cloud environments and the associated usage and costs, FinOps teams need to be able to allocate spend and implement showback to specific teams, services, or applications within their organization. This is why all DoiT users have access to attributions and cost analytics reporting for their whole cloud environment, allowing them to go beyond the quick wins and start establishing a wider-reaching FinOps practice.