For week #5 of the push into FinOps best practices for the upcoming O’Reilly book, we sat down to talk goals & OKRs for FinOps teams with Joe Daly, Director of Cloud Optimization at Nationwide. This area that has lacked FinOps standards for some time and Joe shared his approach focused on three main areas: credibility, sustainability and control.
If you’re not familiar with OKRs is a framework for defining and tracking objectives and their outcomes. For each OKR, there is an Objective to be achieved, along with a set of metrics that will measure the achievement of that Objective, called Key Results. They are typically have a shelf life of a quarter.
The FinOps Foundation is a big proponent of metrics driven cost optimization. In fact, you’ll see a chapter on it in the upcoming O’Reilly FinOps book. Having a metric by itself (say, reserved instance coverage, etc) isn’t enough. That metric needs a target so you know when to take additional optimization actions and when to leave well enough alone. OKR’s, with their measurable system of achievement, proved to be the perfect vehicle for FinOps goal setting at Nationwide because of their measurability and flexibility.
As Joe said, “We call our shots with OKRs and focus on results to provide clarity, accountability and measurable outcomes”. His tip on properly setting key result targets? Aim for the impossible then take two steps back from it.
Joe also made clear that the most important thing with OKRs is focusing on results. When an enterprise is moving to cloud there’s a massive amount of disruption happening in teams who have a mountain of new knowledge that they need to pick up. It can be very intimating for teams who have done things another way for years. As Andy Grove, the CEO of Intel, said years ago: “It almost doesn’t matter what you know. It’s what you can do with whatever you do know.” Don’t worry about what you don’t know today but focus on achieving results. The more you achieve, the more expertise you’ll build.
OKR Focus area #1: Credibility
Joe’s FinOps team at Nationwide has only been around for around a year. “Credibility is probably the most important area to focus on when you’re starting up a FinOps practice. Credibility equates trust and if you don’t have that trust you’re constantly trying to prop up the services you provide.”
So a big focus area of his OKRs is to build credibility internally. He does this by providing transparency from the end user all the way down to the code. A key part of that is regular spend updates (daily, weekly, monthly) at whatever granularity each stakeholder needs: for some it is high level, while for others it is granular. They must be simple and easy to understand. The numbers must also tie out to what their accountants will report to them at the end of the month.
In order to achieve the objective of credibility for his cloud optimization team, here’s a sample of the key result he put in place:
- Deliver daily updates on cloud environment spend to key stakeholders
OKR Focus area #2: Sustainability
All too often, new FinOps teams approach their work in unsustainable ways such as not enforcing tagging in an automated fashion. As Joe said: “Meaningful data needs to be managed by the people to whom it’s meaningful. So what we’ve done is create a tag repository that’s maintained by the application portfolio managers and business facing folks so that we can tie all the application data and business data to the resources without depending on engineers to whom the data is not as meaningful.”
Two examples of key results his team set in this area:
- Be able to tie Application Name, Application Owner and Business Group to each resource
- Automate routine, time consuming tasks like chargeback
OKR Focus area #3: Control
The goal here is to focus on establishing control while also enabling speed. In Joe’s words, “We push accountability for usage control to the application/product teams”. They’ve done this by establishing a direct chargeback model, producing education and encouraging user adoption, and putting in policies to protect against autoscale nightmares.
Some examples of key results his team set in this area:
- Double tag compliance rate
- Establish chargeback for cloud and container usage
- Automate compliance policies
Being flexible and celebrating the wins
At the end, Joe gave a review of how effective they were against the key results in each of these areas. They realized that one of their key results was leading them down the wrong path. They started with the idea of publishing 5 best practice articles internally but quickly realized that wasn’t adding enough value to their users who were just getting comfortable migrating to the cloud. They pivoted to a key result that they would sit down with every application team starting their migration in the quarter and give them personalized training. Conversely, they set—what they thought was—an ambitious dollar figure goal for cost avoidance and ended up achieving nearly double the key result. His team was very excited to see this happen.
What’s else did they accomplish? Joe’s team put container chargeback in place, added application business mappings for making for meaningful reporting tied to their organizational structure, and improved his user views to their cloud environment giving everyone a granular view of just their spend. This gave more transparency and enabled them to chargeback more of the overall cloud bill to responsible parties.
A FinOps maturity model
Up next for Joe’s team is developing a FinOps maturity model that can be used to measure progress of user adoption and highlight areas of development specific to them.
Joe referenced that the talk Jason Fuller gave during one of the first FinOps Foundation first member call’s changed his thinking in this area. Jason shared how important it is to take a crawl, walk, run approach in FinOps.
On the back of this advice, Joe decided to work with each team to meet them where they are instead of pushing out standardized best practices that are too advanced for teams just learning to crawl in cloud.
Focus on collaboration
Joe closed with the importance of ensuring both the engineering and finance teams are collaborating on your OKRs:
“When you have engineering and finance talking together you usually get a solution that works for both departments as opposed to engineers just getting in the room and coming up with the solution that only works for the engineers. Likewise, finance coming in developing policies that make life harder for the engineers. Actually getting people to work together to build better focused OKRs will drive better results for everyone.”
To join the conversation about FinOps, we invite you to join the FinOps Foundation. Each week, we host conversations like this with the world’s thought leaders in this space. Up next week, we’ll be diving into Allocation and Chargeback with Ally Anderson, Business Operations Manager from Neustar.
Keep breaking down the silos…