97 Things Every Data Engineer Should Know

Chapter 85. Total Opportunity Cost of Ownership

Joe Reis

Choosing the right technology often comes down to cost—the time and money required. Total cost of ownership (TCO) is as ancient as enterprise software itself. While TCO has been a mainstay for decades, another cost is perhaps more onerous and far less discussed: total opportunity cost of ownership. As new technologies and paradigms come and go at an exponentially faster rate, companies need to keep their options open and take advantage of new and better ways of doing things.

TCO consists of two big areas: the asset’s purchase price and the continued operating cost over the lifetime of that asset. Companies may choose to purchase managed solutions, implement and manage an open source solution, or roll their own. Each of these decisions has its own trade-offs for both the long and short term.

Apart from TCO, there are two aspects to choosing a technology: the technology itself and the paradigm it represents. You can invest in technology X, which is part of paradigm Y. For example, on-premises Hadoop (technology X) in the early 2010s was part of the new big data movement (paradigm Y). If you heavily invested in on-premises Hadoop in 2010, you probably felt like this was a smart move.

Fast-forward to 2020, and the terms “on-premises” and “Hadoop” both seem antiquated. Since 2010, the world has moved into the cloud, and there ...