How to Reduce Cloud Waste at Scale: Takeaways from Our FinOps Webinar

‍

Roughly two decades ago, software engineering pioneer, Watts Humphrey, said that every company is a software company. He was speaking to the dawn of a new era where technological innovation would mark the rise and fall of a successful business.

Just as technology has infused its way into everything we do, from how we book hotels to how we watch movies, so has the adoption of cloud computing. According to an IDG study, 92% of organizations’ IT environment is partly in the cloud, and this number will continue to rise. Why? The multitude of benefits in the cloud is hard to ignore. From faster and more efficient innovation, better agility to unmatched ease of deployment and agility. 

Once businesses are in the cloud, they are then tasked with managing their cloud spend. Cloud spend can fluctuate depending on use, and without proper management and optimization, cloud spend can reach staggering heights. According to the FinOps Foundation’s recent survey of cloud adoption and use, the amount of money spent on cloud services is around $30 billion and promises to continue snowballing.

So, now what? The question left on the table is what do companies do about all this spend? How do they manage it and prevent these costs from interfering with business? Virtasant CEO Michael Kearns and Program Manager Melissa Lorten sat down with The FinOps Foundation to address this pressing issue: how to reduce cloud waste at scale.

Here are the most important takeaways from our webinar. Watch the full video for more.

Why Taking Action is Hard

The FinOps Foundation's State of FinOps Report highlighted a problem that most of us in the cloud solutions business is keenly aware of. The most difficult challenge when adopting FinOps practices is getting engineers to take action.

This lag in action is primarily because of the conflicting priorities between finance and engineering teams. The finance team remains focused on budgetary needs and the overall cost of things. On the other hand, engineer teams, by default, are focused on innovation, problem-solving and creating functional and fast infrastructure. Because of this disconnect, efficiency tends to get lost, and solving the problem feels pretty daunting. With valid fears around losing capabilities or slowing innovation, many companies have found it challenging to adopt FinOps best practices. The respondents of the FinOps survey reported various stages of adoption - crawling, walking and running.

‍

Crawl: Reactive

• Frequent check-ins and workshops
• Cloud evangelism 
• Quarterly cost reviews
  

Walk: Proactive

• Teams begin to gamify
• Wins are highlighted 
• Monitoring metrics to get ahead of issues

Run: Integrated

• High-quality training 
• Pre-allocated budgets
• Automation is used for discovery and resolution

Strategies that Get Engineers to Take Action

Highlighting the biggest kink in the cloud optimization cycle means addressing the root issue. Not just that engineers are not taking action, but why they are not and how to empower them to do so. Here is a breakdown of the strategy we recommend managers consider when faced with a lack of action. 

Be a Partner, Not an Auditor

Managers, in this case, need to adjust their thinking around how they approach engineers. The goal is not to tell them what to do or force them into a position that limits their ability to build and innovate. The goal is to create visibility and accountability. This shift is successful by partnering with engineers, sharing goals, hearing challenges and providing them with the data they need to make informed decisions. “This is less about telling people what they’re doing wrong, and more about education and training,” said Kearns, “Empowering engineers to take action.”

Make it Easy

Beyond just raising awareness and expectations is the need to create systems that support this new priority—establishing and implementing workflow and deployment standards that create a foundation of fundamental principles that mitigate overspend and mismanagement.

Give Engineering Teams Control

Engineers are working hard on exceptionally complex problems. The decisions they make matter and shouldn’t be revised blindly or purely from a financial perspective. Giving them control is less problematic when they have all the information. Ideally, the FinOps function should belong to the engineering or multi-disciplinary team and not solely the c-suite. 

Make Engineers Aware of the Costs of Inaction

Along with control, engineers should also be made aware of the consequences of the choices they make. Establishing guardrails and boundaries means telling engineers that driving outcomes within the confines of a budget requires them to weigh their decisions and prioritize what’s important. Give them a choice, make them aware of their options, provide them with data and then trust them.

Motivate and Celebrate Wins

In all levels of management, highlighting and celebrating wins has a positive impact. It’s essential to encourage the continuous implementation of these tough choices. Sharing the results of how changes impact the team helps motivate engineers. Make celebrating individuals or entire teams a regular part of the implementation process.

Make it an Organizational Priority

This load cannot rest solely on engineers or a few members of executive management. Cloud optimization and cost management should become a top priority across the board. Preventing essential new best practices from atrophying or losing steam means continuing to echo the need for action and making FinOps a collective priority. 

Establish Measurement and Controls to Minimize Risk and Impact

Often managers find that some FinOps recommendations may not fit the complexity of their infrastructure. There are fears around downtime and how it will impact the flow of business. These concerns are valid. So, it’s essential to question the impact of these decisions and measure risk. Then managers can establish guardrails to protect what can’t be touched and create awareness as to why.

Automate discovery and resolution where possible

“A thousand dollar opportunity may or may not be worth pursuing. But a thousand, thousand dollar opportunities - that’s a million dollars.” The task of identifying and responding to cost-saving opportunities in cloud spend is nearly impossible to do efficiently and effectively without automating how discovery and resolution are automated also matters. Kearns suggests, “Automate the steps but keep it within the process that engineers work in everyday.”