Shift Left with FinOps

In this episode of the FinOps on Azure podcast, Michael Stephenson is joined by Michiel van Oudheusden, a consultant from Xebia in the Netherlands, to discuss the concept of “Shift left with FinOps” and how to successfully implement it across an organization.

What is Shift Left with FinOps?

The idea of shift left is to move FinOps processes and cost optimization earlier into the development cycle, rather than waiting until applications are running in production. As Michiel explains, “If you find things in production, you’re probably already too late. If you can fix it earlier, the better it is, the faster it is, the cheaper it is.”

The Current State at Many Companies

Michiel notes that at most companies he works with, teams have little insight into what they are actually spending on the cloud. There is a lack of awareness and processes around cloud cost optimization. Costs are often just treated as an uncontrollable fixed expense that gets paid without much scrutiny.

One key issue is that developers can easily spin up expensive cloud resources, unlike having to justify mundane expenses like a new mouse or keyboard purchase. This points to a need for better education and changing the organizational culture around FinOps.

Getting Started with Shift Left

To begin a shift left FinOps initiative, Michiel recommends picking a champion team to start with as a proof of concept. Provide that team with:

Training on FinOps concepts and best practices
Tools for monitoring costs and setting budgets
Anomaly detection to be alerted on cost spikes
Techniques for cost optimization like reserved instances
The team should treat cost as a metric to continuously monitor and optimize, just like application performance.

Once there is an initial success story, evangelism and organizational change management become critical for scaling it across more teams. Executive buy-in, involving finance stakeholders, and making FinOps part of the culture is essential.

Potential Blockers

Common blockers to shifting left FinOps include lack of knowledge on FinOps practices and skepticism from teams on the value proposition. Teams may push back saying “I’m just a developer, I don’t care about running costs.”

Michiel stresses the importance of connecting cloud costs to business value – if teams don’t see the benefit of optimizing costs, they won’t have incentive to change. There needs to be transparency into which team/application owns which cloud costs.

Measuring Success

While reduced cloud billing is one obvious success metric, Michiel cautions against optimizing solely for the lowest bill. You need a balanced approach looking at:

Productivity impacts on developers
Risk mitigation through better monitoring
Moving up a defined FinOps maturity model
Aligning FinOps with DevOps culture of “you build it, you run it”

Some of Michiel’s most memorable FinOps wins came from areas like optimizing log storage policies, evaluating build vs buy decisions around shared services, and taking advantage of new pricing models like Azure’s consumption plans.

The Future of FinOps

Looking ahead, Michiel is interested in seeing more content around rate optimization strategies with reserve instances and saving plans, as well as emerging topics in sustainable “GreenOps” for reducing the environmental impact of cloud workloads.