Cloud spend optimization is a challenging endeavor for most companies. The major reason for this is that most don’t have sufficient visibility into their live cloud costs, so they can’t tie day-to-day decision-making to those staggering AWS bills that arrive at the end of each month (way too late to make meaningful changes).
That said, there are a few key questions that teams can pose internally to understand current capabilities around cloud spend optimization and determine the best path forward. In today’s blog post, we’ll share the five major questions we believe every company should ask themselves when it comes to cloud spend optimization.
1. Where Are We Wasting Cloud Spend?
In a sense, the best place to start with cloud spend optimization is the simplest. It’s very similar to how you might approach tightening up a personal budget. Review statements and look for where your money was spent in a way that was unnecessary or didn’t really bring enough value to be worth it. In the cloud, this means looking for wasted spend or cloud cost “mistakes.”
Recently, we rounded up the five most common AWS cloud cost mistakes. The areas where we see the most wasted spend include:
Misconfigured NAT Gateways
Overuse of AWS Snapshots
Too Much Overhead on AWS Management Services
You can get more details on each of these in the blog post linked above.
If you’d like to start your cloud cost optimization journey by picking off some low-hanging fruit (which is a great way to generate momentum and demonstrate value to higher-ups), these areas are a perfect place to start.
2. Do Teams Have Sufficient Data for Cloud Spend Optimization?
The worst way to go about optimization is to do it without visibility into current state. In fact, this lack of natural visibility into cloud costs is the reason why most organizations aren’t operating in a cost-efficient way in the cloud. So it’s worthwhile to ask the team what exactly they do have visibility into today and where their visibility ends.
Typically, DevOps teams, who are building the applications and infrastructure that ultimately drive cloud costs, have almost no visibility into how their day-to-day decisions affect the bottom line.
That said, finance also likely does not have sufficient visibility into how cloud costs are generated. To move forward, there needs to be an open dialogue between developers about the value they derive for the organization from their cloud spend, so that finance can understand its role in revenue generation. This will reduce friction between the teams and ensure that both have the data they need to make good decisions about cloud spend optimization.
3. Do We Need to Hire?
Sometimes the best thing your team can do to move the needle on cloud spend optimization is to hire. If your team is currently unable to predict, report on, and attribute cloud costs, there can be a lot of value in hiring a financial cloud operations manager.
The difference between this role and a more traditional cloud operations manager is that, while a cloud ops manager may be part of DevOps and/or report to engineering, a financial cloud ops manager would have a direct line into finance.
This role also typically works to actively shape the cloud budget, and proactively drive cloud spend optimization for the entire team. If you’d like to explore whether this role is a good fit for your team, check out our post about hiring a cloud financial operations manager.
4. Is it Time to Go Serverless?
Sometimes cloud cost optimization can be best achieved by starting out with a more modern architecture. At CloudZero, we run completely serverless and find that it allows us to be more nimble and optimized across the board.
The time is coming when companies will no longer adopt cloud-native and serverless approaches simply because they can get to market faster. More and more organizations are moving to serverless to get rid of surprise cloud bills at the end of the month.
Soon cloud cost will be a first-class metric that unites developers and finance by providing both with a clear view into live cloud costs. This will happen sooner for organizations who embrace serverless technologies.
5. Do We Know Our Unit Cost?
Unit cost is an important metric all teams should be aware of. Different businesses will have different unit costs. For example, if you are a SaaS business, your unit cost may be one of the following:
cost per widget
cost per API call
cost per report
Regardless of how you define it, this is an important metric to pay attention to, because clarifies how to spend resources in the cloud and how to charge customers. In other words, it’s fine to have high cloud spend if the business is seeing strong returns on that investment.
Additionally, with a better understanding of unit cost, cloud spend optimization decisions can be made that will positively impact the bottom line without significant negative impacts for customers.
What Comes Next
Once you have answered the questions above, you should have a clearer sense of where to focus your cloud spend optimization efforts. Whether that means picking off some easy areas for optimization to get the ball rolling or making a significant hire, it’s better to take action now because it will almost certainly produce significant savings over the long haul.
STAY IN THE LOOP
Join thousands of engineers who already receive the best AWS and cloud cost intelligence content.