📝 usncan Note: Screws Tighten On FinOps Fissures
Disclaimer: This content has been prepared based on currently trending topics to increase your awareness.
plumber at work in a bathroom, plumbing repair service, assemble and install concept.
getty
FinOps fixes fiscal fissures. That’s the promise that stems from the FinOps movement, a combination of financial management and DevOps (developers, plus IT systems operations and administration). It is designed to promote an operational framework and cultural practice which maximizes the business value of cloud and related technologies. According to the FinOps Foundation, this methodology “enables timely data-driven decision making and creates financial accountability” for cloud computing environments. It does this through fostering collaboration between engineering, finance and business teams.
But rather like installing an exercise bike in your bedroom or office, putting a vegetable juicer in your kitchen or taking out a new gym membership, unless a new initiative or product or service is actually adopted and put into functional use, it risks languishing comparatively useless.
If You Don’t Use It…
So it is with FinOps. Simply “adopting” a FinOps initiative, employing a FinOps specialist and formally signing up to any given FinOps vendor’s platform and toolset is not the same as “doing” FinOps. This is the suggestion coming from CloudBolt (which, no surprises, is a FinOps cloud return on investment specialist) in a recent market study, the firm has sought to highlight the disparity between companies with high confidence in their FinOps practices and the actual ability to measure, prove, or act on those capabilities.
The company’s user analysis questioned some 350 senior IT leaders on their FinOps preparedness. Findings suggest that cloud ROI still remains elusive, optimization recommendations go unimplemented, Kubernetes costs can still spiral and automation accelerators (though widely claimed to be well-embedded within IT teams) rarely deliver scalable results.
So what’s happening?
Aspiration To Action Chasm
“The issue isn’t the FinOps discipline itself; it comes down to how companies have interpreted and implemented it,” said Kyle Campos, chief technology product officer at CloudBolt. “There are systemic barriers to be aware of here, from automation illusions to Kubernetes opacity, which reveal how strategic goals are outpacing operational execution. There’s a whole field of organizations deploying FinOps, but they’re caught between aspiration and action – i.e they are eager to move forward, but they are missing the scaffolding to do so. As AI, hybrid cloud and containerized infrastructures redefine the modern enterprise, the need for companies to evolve their FinOps grows more urgent. Not just in tooling. Not just in org charts. But in the operational frameworks that focus on turning insights into action.”
While technology surveys are arguably always worth holding at arm’s length (unless you can get a friend to hold one slightly further away), there may be a couple of industry pointers in this study.
When asked what’s holding them back in terms of getting FinOps to work, technology leaders pointed to difficulty linking cloud spend to business outcomes, some degree of organizational misalignment or the existence of business unit (or IT stack) silos and the presence of inefficient resource management in the cloud. Being more specific, in the world of cloud, inefficient resource management means things like inconsistent accountability (buying some cloud, but not knowing which business unit functions it is supposed to be charged back to) and the act of “poor tagging”, where metadata-based tags are used to categorize and filter specific resources within a cloud environment.
Notably, unit economics and business-value metrics ranked dead last, despite being the most direct path to demonstrating value.
What Is Technology Tagging?
As clarified by FinOps platform company Finout, AWS tags are labels that may be applied to resources such as EC2 instances, S3 buckets, etc. “They are a simple way for teams to associate metadata with a resource. Each tag is comprised of a key-value pair, both of which are user-defined. AWS resource tagging is the first step in a cloud cost management strategy. AWS allows users to set up tag policies and to tag by resource group and provides a tag editor.” Comparable cloud tagging functionalities also exist with Google Cloud, from the platform offerings within Microsoft Azure and in cloud platforms from major tech vendors such as Oracle Cloud Infrastructure Tagging.
If the cloud cost iceberg has a sharpest edge, it may be Kubernetes cloud container orchestration. Campos says that nearly every organization agrees that Kubernetes significantly drives cloud spend. But when asked whether they can actually optimize it, most say they can’t. This is said to be because Kubernetes isn’t just complex, it’s opaque i.e. some even call it a black box. Containers accelerated deployment velocity and portability, but at the expense of cost clarity. Cost gets buried in layers of abstraction: shared nodes, ephemeral workloads, disconnected tagging and autoscaling behaviors.
Forrester Analyst Viewpoint
“There are potentially useful insights here. At the same time, it’s important to remember that FinOps surveys that emanate from FinOps vendors are inevitably designed to deliver a payload of ‘findings’ that generally fall in line and resonate with the company’s core message set and technology proposition. That being said, let’s take what we can here on merit,” said Tracy Woo, principal analyst and FinOps practitioner at Forrester. “It’s unsurprising to me that unit economics is ranked dead last. Despite it being a few years old, it’s very, very hard to do. That’s because tooling can’t help; this requires rigorous soul-searching to determine the unit costs that are unique to each organization.”
On the suggestion that there’s a growing gap between what teams say they’ve automated and what that automation actually enables, Woo completely agrees. She says that automation is a sign of FinOps maturity, which may be why teams will claim automation and use it as a benchmark to having achieved “run” or “walk” status when in reality their automation is along the lines of automated report delivery or something else that has little impact on actual optimization.
“On the question of FinOps governance often being a flaky as a readout, not a response. I love this line, there are a lot of governance efforts that are nothing more than a dog and pony show, i.e. grand ambitions with little effect. On the question of the most funded priority being hybrid multi-cloud management (a domain many assumed was ‘solved’ a decade ago), is not surprising,” agreed Woo.
“We define them by two different terms: ‘multi-cloud management’ – management of public clouds, with some on-premises. Then ‘hybrid cloud management’ i.e. management of private cloud and public cloud but with capabilities that started from on-premises private cloud with capabilities that extend into public cloud. It’s a small nuance, but crucial at Forrester. This area of multi-cloud or hybrid cloud management has never been solved. It’s an industry that has flickered in and out throughout the past decade without actually taking off. The problem? Hybrid cloud management is hard to do,” detailed analyst Woo, in a personal briefing on the FinOps market.
FinOps Key Market Vendors
A wider competitive analysis in this space will see usual suspect vendors line up including Morpheus Data; VMware Aria for its cloud automation, operations and cost management services; ServiceNow for its workflow-focused cloud operations management tools; Terraform by HashiCorp for its FinOps-centric infrastructure as code technologies; Scalr; the above-mentioned Flexara, BMC, Cloudability, CloudZero and big vendor platform plays from Red Hat Ansible Automation Platform and Cisco with its own-brand Intersight Cloud Orchestrator.
From a software tooling usability perspective, the suggestion here is that tools are either too simple (not enough visibility or management/automation capabilities), or they are too complex (too much functionality that is buried, difficult to find, difficult to use and is difficult to prove ROI from) today.
But FinOps isn’t standing still. It’s expanding across clouds, across teams, across domains, across mandates. It seems apparent that the future of FinOps won’t be owned by any one function. It will be built through orchestrated action. Organizations will need to think about what automation really means, treat tagging as infrastructure, not hygiene and consider governance to be operational, not observational.
