DevOps and Business

Chapter 1. Introduction

Every organization has objectives that define its own success, but rarely do technical objectives such as deployment frequency, technology choices, or performance metrics make it to the top of an organization’s Key Performance Indicators (KPIs). And although those IT metrics are vital to a company’s success, they do not mean much to nontechnical people when presented in isolation.

Here’s an example. Imagine that a developer in a large enterprise organization that provides multiple web-based platforms to its users has found a way to improve load time across all properties by 200 milliseconds. She sees an opportunity to deliver value, so she writes up a plan for the development, QA, and deployment effort required, and conservatively estimates that the effort will take two weeks. She presents that plan to her manager, hoping it will make it into the next sprint, only to see her plan rejected. The reason? The manager sees it as a disproportionately large investment for minimal gain. Even though the task is added to backlog, it is given the lowest priority, which means it has little chance of ever actually being implemented.

In this example, the core issue is that load time is a technical metric that means very little to nontechnical people without context. The manager did not understand—and the developer did not communicate—the business impact of performance improvement; without additional context, deprioritizing the effort based on return on investment seems ...