There are two basic models for evaluating the cost of any technology. Organizations often gravitate to return on investment (ROI) because it models a clear financial return. Total cost of ownership, on the other hand, factors in all the costs involved in deploying and operating a solution. While one model is focused on revenue, the other is focused on cost.
But choosing one versus the other won’t give you the full picture. While TCO might demonstrate reduced setup and operating costs—particularly with a fully managed service like Confluent Cloud—cost alone is not the only reason to choose one solution over another. This model leaves out factors like:
Some of the biggest benefits of a service aren’t factors in a pure TCO model or a strict ROI assessment. That’s why Confluent combines both when we run a business assessment to help organizations understand the costs of managing Apache Kafka®. But we also add a third element: speed to market. Let’s start there.
Deploying Apache Kafka presents challenges that aren’t the core problem most companies are trying to solve in the first place. While self-managing Kafka may seem like a good idea at first to developers or organizations who want full autonomy over their Kafka deployment, it ends up creating significant operational burdens—which is not where most companies want their best people focused.
With a fully managed service for Apache Kafka, your organization can shift key resources to higher-value tasks. This refocus results in opportunity optimization that can’t be captured in TCO alone. So before we evaluate TCO, we look at a critical factor: the speed to market that a new technology will enable.
For instance, with self-managed Kafka, you need to perform the manual sizing, provisioning, expansion, and maintenance of a Kafka cluster. With Confluent, we do all the work for you. Clusters are provisioned instantly and maintenance is seamlessly managed. You can start streaming data on the day you sign up.
With Confluent, you can also skip the part where you spend six to nine months hiring and training the people who really get Kafka—cut right to the chase of product development. These are important factors to consider before you even look at TCO, which comes next.
Now, we can look at the total cost of ownership. There are a bunch of factors at play here, including:
We typically break down TCO into three areas of cost:
We also look at risk costs, including the risks of downtime, performance degradation, and security infringement—all three of which are less likely with a fully managed service. These are risks that may be hard to quantify, but they’re still important to consider.
And finally, there’s ROI.
Any credible ROI calculation should include a range of outcomes, including lower/upper bounds (sometimes referred to as a Monte Carlo simulation). We always try to model a range with a conservative estimate in the middle. This is the best approach for forecasting potential events despite assumptions and factors outside of real-world control.
Most people assume that the reason for completing these exercises is to walk away with a perfunctory number—and we can certainly do that. For Confluent Cloud, the TCO number as a general rule is up to 60%.
But the value that comes from this process is greater than just arriving at figures. The real value comes in the process of analysis and collaboration during the exercise itself to get more insight into how you run Kafka, how much it costs, and how it impacts your teams and business.
When you take speed to market, TCO, and ROI and combine them all, you get a complete view into how much time and money you save with a fully managed service and how much those savings unlock for your business.
Sign up for your TCO assessment to learn how you can save up to 60% on your cost of running Kafka.
If you’re curious to learn more about our business case evaluation process and how we arrive at both hard numbers and valuable knowledge about your use case, download the free white paper Measuring the Cost Effectiveness of Confluent Cloud.
Lyndon Hedderly leads the Business Value Consulting Team within Confluent. He has over 20 years of experience in the enterprise software industry, starting as a technologist and moving into IT strategy and enterprise transformations. Lyndon holds a Master of Science in neuroscience including neural networks & AI from Oxford University.