No-code ELT tools are either the best innovation in data engineering or the worst nightmare for professionals. On one side, there are those who believe data engineers should be writing code from 9 AM to 7 PM every single day. On the other, there are those who prioritize quick iterations and rapid insights. The battle lines are drawn.
If you’re wondering what I mean by “no-code ELT tools,” think about platforms like Airbyte, Fivetran, Stitch, and Meltano.
These tools promise to simplify the complex process of extracting, loading, and sometimes transforming data without writing a single line of code.
For some, they represent a revolution in data accessibility and efficiency; for others, they undermine the very essence of what it means to be a data engineer.
In this post, we’ll dive deep into why these tools are loved by some and despised by others, and why they evoke such strong emotions in the data community.
The Love for No-Code ELT Tools
No-code ELT tools democratize data management by enabling non-technical users to easily set up and manage complex ingestion pipelines through intuitive interfaces.
This accessibility empowers teams to quickly deploy ETL processes without relying on engineering resources, fostering agility and reducing bottlenecks.
The result is faster, more cost-effective data integration that scales with the business’s needs. Additionally, these tools enable rapid prototyping, allowing businesses to iterate on data processes swiftly without lengthy development cycles.
By removing technical barriers, no-code ELT tools free up resources, allowing teams to focus more on strategic data initiatives and less on implementation details, ultimately driving innovation and efficiency.
It seems like a paradise on earth. But why do some people hate them so much?
The Hate for No-Code ELT Tools
I won’t focus on issues like pricing (hello Fivetran) or limited flexibility (Airbyte’s connector builder allows integration from unsupported APIs, so….) because I don’t believe these are the main reasons people dislike no-code ELT tools.
The debate often centers on whether data engineers are truly software engineers, a discussion I find puzzling. Whether or not they consider themselves as such shouldn’t be a big deal. Why does this debate even matter?
Those who dislike no-code ELT tools are often purists who believe that data engineers should be coding constantly. They want engineers to spend every moment writing and maintaining code.
But why should I write and maintain code to extract data from Stripe or TikTok Ads when it’s already been handled by others?
My take
I believe people often misunderstand the role of data engineers or any data professional in general.
Our primary goal is to provide value to the business through data. If there’s a way to deliver that value more quickly, should we ignore it?
At the end of the day, business leaders prioritize ROI and quick deployment, making no-code ELT tools attractive for their fast implementation and immediate benefits.
However, they must also consider the long-term implications, as these tools might lack the robustness needed for complex data needs as the business grows.
The challenge lies in balancing short-term gains with potential future limitations, ensuring that the data strategy supports both immediate objectives and long-term sustainability.
Conclusion
No-code ELT tools offer a revolutionary approach to data management by simplifying processes and enabling faster deployment.
They democratize data access, allowing teams to quickly set up and manage ETL pipelines without deep technical expertise.
However, the debate persists between those who see these tools as a game-changer and those who believe they compromise the depth and control provided by traditional coding.
Ultimately, businesses must weigh the immediate benefits against potential long-term limitations to ensure a data strategy that balances efficiency with robustness.
What are your thoughts on this? Are you in the “code at all costs” camp, or do you prefer “not reinventing the wheel”?
— Dorian Teffo
