What is FinOps?
In the simplest terms, FinOps is the joining together of finance and operations. Just like DevOps is the joining of development and operations. Both are partly tool-based and partly cultural. And both aim to fix problems that face technology or cloud-driven businesses.
The goal of FinOps is to align cloud cost optimisation with the broader goals of the business by introducing transparency, accountability and clear reporting.
Given the dynamic nature of cloud service costs, and the many services leveraged by tech-driven businesses at any one time, gaining visibility (and making better decisions) can be a challenge.
This is the problem FinOps was designed to solve.
What is FinOps automation?
FinOps automation refers to the tools and solutions aimed at reducing the manual effort involved in implementing FinOps for savings, visibility and agility. These break down into four rough categories.
Monitoring & visibility automation
Some tools focus on providing real-time insights. These solutions empower stakeholders across the business to identify inefficiencies, forecast accurately and save time on data collections and analysis.
Examples include Apptio Cloudability and CloudHealth by VMware.
Resource optimisation automation
Other tools are built to identify and rectify inefficiencies in your cloud environment and resource usage. These tools often act directly, right-sizing or spinning down idle resources.
Examples include Spot.io and our own product Turn it Off, which focuses on saving costs and carbon by ‘pausing’ any idle non-production resources.
Rate and purchase automation
These tools automate the strategic buying and swapping of discount instruments. Discount instruments can refer to things like AWS Reserved Instances, Spot Instances or Savings Plans: any discount offered for a model of purchase of a cloud service.
Examples include Flexera One and ProperOps.
Governance automation
This last category offers solutions for enforcement and governance. This might cover resource tagging, budget and cost controls, resource provisioning, or, moving somewhat outside FinOps, access and security.
Examples include ServiceNow Cloud Management and OpenOps.
Do you need a separate tool or platform to deliver each area of FinOps automation?
No. Although many tools have an area of focus, there is a great deal of overlap, too. In practice, a mid-size FinOps team might use a handful of core tools to deliver all the functionality they require.
Which FinOps processes any particular team may prioritise will vary. Let’s look at some examples to make this point clearer.
Example 1: a B2C SaaS product with cloud cost optimisation as their no.1 priority
Let’s imagine we have a streaming or otherwise high-resource-usage SaaS product serving a large number of customers.
Tech-driven businesses like these will use a range of FinOps tools, but cloud cost optimisation might lie at the heart of their strategy, (as this directly affects their margins on each tenant or user).
A business like this might automate rightsizing, scaling, and anomaly detection to handle dynamic workloads efficiently.
This would allow them to deliver at scale while minimising costs.
- Tools used: Apptio Cloudability, Spot.io and Datadog
- Leaning into features: while these tools provide a range of FinOps capabilities, this company may prioritise the cost optimisation features of Spot.io, such as auto-scaling and identifying idle resources, as these directly impact their margins
Example 2: a managed services provider who needs a clear allocation of cloud resources
Now let’s say we have an MSP. This company manages multiple complex workloads for a variety of businesses.
In this case, governance automation and monitoring and reporting may be the FinOps team’s key priority.
Here, proper tagging, visibility and cost reporting allows the MSP to scale and innovate their clients’ workloads without compromising their ability to report costs (and their own charges to their clients), which is crucial to any MSP.
- Tools used: Apptio Cloudability, AWS Config, and ServiceNow Cloud Management
- Leaning into features: these tools are also capable of cost optimisation, but the MSP’s FinOps team may focus on tagging enforcement, client-specific reporting, and multi-cloud visibility to meet their billing and governance needs
Example 3: a game developer with cost and ESG targets
Finally, let’s imagine a games developer who’s looking to reduce cost while reaching the ESG targets they’ve just committed to.
Live-service multiplayer games are infamous for their cost-hungry dev and staging environments. After all, they need to continually test and release features in near-production conditions.
Add to this that game environments are on the more complex side of things, and you have a clear case for FinOps automation in the areas of cost optimisation and reporting.
- Tools used: Turn it Off, Apptio Cloudability, Spot.io
Leaning into features: Our own solution, Turn it Off, is a specialised tool for pausing non-production resources when they’re not in use. We focus on a single outcome and make a difference within minutes of sign up, unlike generalist FinOps platforms. On top, we provide sustainability reporting for companies with net zero targets. In this example, Apptio Cloudability allows for detailed reporting and forecasts while Spot.io handles broader optimisation across production environments
Key benefits of FinOps automation, from cost savings to greener outcomes
In the previous section, we looked at how FinOps automation can help solve key pain points. In this section, we’ll consider why companies are leaning into FinOps automation more broadly.
Scalability
While some degree of automation will likely be present from the start, scaling a business up exacerbates this need.
Managing cloud costs effectively at scale is far easier when reducing manual errors, time spent and unnecessary costs.
For businesses wanting to deliver their product to 10 times the number of users they served in the previous quarter, only heavily automated processes will suffice.
Agility
Clear information means clear choices. Demystifying cloud billing data and cloud financial management allows teams to see where money is being spent, parts of the business that are more/less efficient and potential solutions.
Again, this is more true of tech-driven businesses, but if this is you, automated FinOps can mean the difference between an obvious path forward and an endless struggle to gain visibility on your environment.
Collaboration and organisational health
Just like DevOps, FinOps was invented to tackle a major organisational pain point: the disconnect between finance and operations in cloud-driven organisations.
Like the other entries in this section, this connection requires a high degree of automation at scale to operate at peak efficiency.
With a robust FinOps automation approach, an operation’s cloud use is clear and transparent, allowing for a meaningful and intelligent dialogue not hampered by the need for manual/partial data collection – which may leave out crucial parts of the puzzle.
Sustainability
Though not every daring scale-up will consider sustainability and reducing their carbon footprint as mission statement essentials, many do. And as businesses scale up, sustainability becomes ever more important.
Not only does good ESG reporting unlock a range of investor and financing opportunities – what saves on carbon usually saves on costs, too.
Many FinOps tools, our own included, help businesses track their reduced costs in carbon terms, providing clear and accurate data that can be used to make smarter decisions internally and gain access to investment elsewhere.
Best practice when adopting FinOps automation
Now we’ve looked at the main areas of FinOps and why automation is an attractive choice for controlling cloud usage and cost reduction, let’s examine the how-to’s of FinOps adoption.
Each organisation will have different requirements, so we’ll cover 4 popular approaches. And you can pick what works for you.
The crawl-walk-run approach to FinOps automation
Crawl-walk-run is a generalist methodology often applied to types of organisational and technological adoptions. As you could probably infer, the idea is to begin with the most simple aspects, before going on to hit a comfortable mid-level of adoption and finally embracing the change wholeheartedly.
DevOps adoption is often guided via the crawl-walk-run method, also. In a FinOps automation context, it might look like:
- Crawl: This stage often means laying the groundwork for FinOps automation. In many cases, although not all, this entails focussing on visibility and reporting. The more interventionist elements of the walking stage cover things like automated tagging policies and infrastructure monitoring tools. These lay the groundwork for easily accessed data and information.
- Walk: The walk phase is more interventionist. With your data collection firmly in place, you can now begin to automate more areas of cloud cost management. The walking phase often involves the handing over of rightsizing and the spinning down of unused resources.
- Run: At the run stage, more strategic decisions are given over to automation. Provided the previous two stages have gone smoothly, organisations now bring automation to budgets, forecasting and the selection of discount instruments (like Savings Plans and Reserved Instances). This is also typically where governance is automated.
Is this approach for me?
As you can imagine, the crawl-walk-run approach works well for organisations with less specific goals and who have no pressing time concerns. If there is a business-critical need for cost savings, this likely isn’t the way to go.
If you want to test the waters, and head in no particular direction, crawl on in.
The pilot and scale approach to FinOps automation
The pilot and scale model is also a generalist framework. Unlike crawl-walk-run, this method sets out to solve a specific problem which might incorporate FinOps processes from the crawl, walk and run stage in a single, defined pilot.
Depending on the success of that pilot, broader changes may be rolled out.
The below is an illustrative example, but it can be applied to different FinOps problems:
- Define a use case: Begin by defining the key pain point you’ll be addressing. Let’s imagine our company has recently scaled and we’re now adopting a multi-cloud environment to serve our customers.
The problem is, this expansion has seen a huge rise in untagged resources as well as budget overruns. So, we could look at hiring more people and reviewing our internal processes; or we could pilot an automated governance policy.
- Deploy a pilot: The pilot will typically be implemented in a limited capacity. Hence the term pilot. In this example, we could use AWS Config or Terraform with Sentinel to enforce tagging standards and remediate non-compliance in a particular project, region or other area.
Although we’re only looking at this one area right now, the pilot will incorporate broader FinOps principles like visibility and cross-team collaboration.
- Evaluate and scale: Assuming the pilot succeeds in its goal, it’ll be rolled out to other areas of the business. In so doing it will demonstrate the specific value of the work undertaken as well as the broader FinOps principles needed to pull off the advanced use case. This sets you up to embrace FinOps more broadly.
Is this approach for me?
The pilot and scale approach works well if you have a specific problem. If a problem needs to be solved fast, and you think FinOps automation is the solution, this is a great way to test that hypothesis and lay the groundwork for further development.
The CoE approach to FinOps automation
The last approach is a little different. Also a multidisciplinary idea, a CoE (centre of excellence) is a cross-departmental team of experts who work together to deliver an outcome. In FinOps automation adoption, this team might deliver:
- Cross-functional expertise: The team will be made of decision-makers in finance, operations, engineering and any other relevant departments. Together, they’ll be able to deliver the project in a way that meets each stakeholder’s needs.
- Unified governance and standards: As well as delivering automated governance, this team will have the authority to say what those standards should be: they act as FinOps lawmakers for the organisation or business.
- Enablement and training: The CoE team is no only a planning committee. They implement policies directly where they can and via training and enablement when a change requires a company-wide shift.
- Results and reporting: The CoE team sets clear FinOps automation KPIs and reviews them according to a set road map.
Is this approach for me?
This approach works well for large businesses with defined power structures. Where a change is complex, and must be delivered in a certain way, having a small team of leaders meticulously plan and execute is the best policy.
This would make less sense in a small organisation just testing the waters, where a crawl-walk-run approach would work better.
Do you have a microwave? Our 60-second pitch
Lots of businesses opt for a few generalist FinOps platforms and call it a day. That’s great, but there are greater opportunities, costs (and carbon) savings to be had with the addition of complementary micro-SaaS tools.
You wouldn’t not have a microwave because you already own an oven. One does the bulk of your cooking, the other saves you time and does those little jobs an oven’s just not right for.
We like to call it the microwave/microSaaS analogy.
In our own case, we edge out generalist FinOps platforms on intelligent idle resource detection and immediate impact. With Turn it Off, companies can achieve immediate savings within minutes of sign up – offering a laser focus that can’t be beat.
You might stick with a general platform for real-time rate management. But for a few clicks (and less than no money, because we’re always cost-neutral) you could be saving costs and carbon before you can say – ding.
