In the world of data, there’s a phrase I hear so often that it annoys me. It is “Garbage In, Garbage Out.”
For a long time, it’s been used as a simple, effective shorthand to explain why a flawed dataset will inevitably lead to flawed analysis and inaccurate results.
However, today, in our more team-oriented data ecosystems, this phrase has become as much of a liability as a lesson. It’s a lazy excuse, a verbal shrug of the shoulders that fosters a culture of helplessness.
Clinging to “Garbage In, Garbage Out” is not just outdated, it’s actively damaging our ability to build robust, trustworthy data systems, and it is time we bin it.
The biggest problem with “Garbage In, Garbage Out” is that it implicitly frames data quality as an unchangeable law of nature. It suggests that once “garbage” data is created at the source, its journey through our systems to produce “garbage” insights is an immutable conclusion.
Think about the message this sends across an organisation:
- A data engineer sees a poorly formatted data stream. They think, “Garbage in, Garbage out,” and pass the problem downstream.
- An analyst pulls data into a report and finds inconsistencies. They present the flawed findings with a caveat: “Well the source data isn’t great, and so garbage in, garbage out.”
- A business leader makes a poor decision based on that report. When the results don’t materialise, the blame is placed on the “bad data,” as if it were a natural disaster beyond anyone’s control.
The phrase “Garbage In, Garbage Out” allows us to point a finger at an abstract concept and wash our hands of the problem. It creates a chain of blame instead of a chain of custody. A modern data professional’s role isn’t to be a passive courier of data, regardless of its condition. Our job is to be stewards, custodians, and curators and continuously improve a data ecosystem.
Beyond deflecting responsibility, the GIGO mindset breeds fatalism. It makes the task of improving data quality seem monolithic and impossible. If the input is fundamentally broken, why bother trying to fix it? This thinking stifles innovation and discourages investment in the very tools and processes designed to solve the problem.
The entire field of modern data management is a direct refutation of the Garbage In, Garbage Out philosophy. We have a huge range of best practices and technologies built specifically to intervene in that supposedly linear path from input to output. It isn’t an overnight fix, but addressing a culture that washes its hands of bad quality data instead of taking collective ownership is the first step.
The move should be from passive acceptance to active ownership. Data is not just something that happens to us that we have no control over, but it is an active resource that we have a responsibility to build and maintain.
The quality of our data is a direct reflection of our standards, processes, and culture. It is something that can be developed with good governance. So, the next time you hear “Garbage In, Garbage Out,” stop and ask a different set of questions:
Where did this garbage come from?
How can we stop it at the source?
What can we do to clean up what’s already here?
Matthew Wilson
Junior Data Consultant
CDMP Associate
