As organizations scale, it becomes hard to have a single DevOps or Platform team managing all the infrastructure requirements of all product teams. Instead, they opt for using Infrastructure as Code (IaC) to create a set of easy to use modules that are managed by the central platform team, and used by the rest of the organization as and when infrastructure is needed.
The problem they face is that there is a lack of cost visibility and understanding of what the infrastructure will cost. This results in costs going out of control and budget being broken.
The central DevOps and platform teams have built up a lot of knowledge and best practices as they have created and scaled infrastructure. Self-service has enabled quick launch of infrastructure for all product and engineering teams; central guardrails and policies help them stay within these guidelines.
In many cases, the changes that are being made to infrastructure are due to additional requirements, features, or changes to existing features of your product. These originate from product managers and product owners, who are ultimately responsible for the profit and loss of the products they run.
Product should also be aware of the cost implications of additional features as they are best placed to make cost-benefit analysis, and request higher budgets.