Airbyte vs. FiveTran - ETL & ELT Comparison

Independed comparison of FiveTran vs. Stitch Data, the two leading SaaS solutions for ETL & ETL. Compare source, destinations, connectors, features and pricing.

[@ZhengCheng]

Thanks for the intro about these 2 tools

[@jimmyf.x.9526]

Sounds like we have a definite winner!! Thanks

[@datasleek7950]

Thank you for the comparison. I think I’ll try Airbyte. It’s not always about connectors. It’s also about how the connector handles schema changes for example or can the connector track changes. Fivetran support both of these features especially for databases. Being able to track changes across time by adding extra columns to the destination. 5tran has also been supporting DBT for a while, before AirByte was born. They have lots of DBT packages already built.
I agree being able to host it yourself is great and Cloud solution where you only pay for the compute could be a game changer.

[@Restackio]

Thanks for the feedback, super valuable. FiveTran started to support dbt core in a very basic way from 2020 on. Only in January 2022 did they expand the functionality to make it more useful. You had to trigger transformations time-based and you had to guess when the data was extracted in order to run transformations. These were already functionalities Airbyte was designed with from the start and any dbt core project for your custom normalization is triggered at the end of the sync job (www.techtarget.com/searchdatamanagement/news/252512138/Fivetran-updates-data-integration-platform-with-dbt-core)
I really like your feedback regarding the handling of schema changes. Support to track changes of data in rows is very specific per connector for Airbyte. So it depends if the connector is in alpha, beta or GA if insert + dedup history for example is available where you can set a cursor field like 'updated_at' in order to update records to the latest changes only. Extra column is also a really nice point and will include it in future comparisons for ETL tools.

[@SamStrong]

It seems like both Airbyte or FiveTran need a separate transformation tool (dbt or similar) to actually be useful. I’d love your perspective on where a tool like Rivery fits into this market.

[@datasqlai]

Does airbyte has sybase connector ?

[@Restackio]

Not to my knowledge. You can find a list of connectors here : airbyte.com/connectors?connector-type=Sources%7CDestinations

[@datasqlai]

How much is airbyte scalable ? Can we create thousands of pipelines with sources like SQL server and sybase ? How thousand of pipelines will be scheduled and managed . Any central management console available ?

[@Restackio]

Here's a video that can answer some of your questions.
th-cam.com/video/btjhlOLkUlc/w-d-xo.html
You can try Airbyte on restack for a central console experience. restack.io/products/airbyte

[@quangainguyen6209]

I think you presented the comparison in a heavily biased manner, showing a clear preference for Airbyte.

[@Restackio]

Thanks for your feedback. As I mention in the video as a platform for open source products, I am biased towards open source by default. Still, I invested a lot of time 18 months ago to research and really help data engineers to understand when to use one or the other. This video is a small part of a data engineer's research besides creating their own comparison matrix, looking into docs, and checking communities and I am convinced that this video is helpful and viewers can deal with my slight bias. Soon I will create an updated video and will take your feedback even more into account, both platforms have evolved a lot in the meantime. Best Martin

[@quangainguyen6209]

@@Restackio Thank your for your clarification. I am looking forward for new content on our channel. I hope you have a good day!