Portable.io, a game-changing ELT tool ?
While checking out Linkedin, I came across a post saying : “Imagine an ETL tool with 125+ connectors. That’s some IMPRESSIVE engineering. Now imagine this instead : Portable just added 125+ ETL connectors to hit our goal of 500+ connectors by the end of April”
As a Data Engineer, I go through an almost daily ritual of being shown advertisements for upcoming ETL/ELT tools, and immediately dismissing them, but this post caught my attention.
If you’ve been using classic data extraction tools like Talend, you know how painful it can be to connect an uncommon business application to your data warehouse, and the complexity that adds to your workflow. Portable seems to get rid of that issue, which sounds too good to be true, so I decided to give it a try.
The first thing I noticed is that it’s only accessible from the US. Even though I had access to the landing page, I couldn’t visit the “Try it Free” page, so I had to use a VPN to open it. With all the Data regulations in Europe, I don’t say I blame them, but hopefully it will be available worldwide soon.
The tool is pretty straightforward to use, as it only has 3 tabs (Sources, Environments, Flows) :
- First you choose a connector out of the endless options : Notion, Github, Confluence, Discord…
What I like about Portable is that they provide clear instructions to establish the connection.
2) Create a destination in the Environments Tab. For my example I chose BigQuery, for which you have to provide the project name, the dataset name as well as the Service Account key. They also have all the common environments : Snowflake, MySQL, PostgreSQL, Redshift and Google Sheets.
3) Create a new Data flow in the Flows tab, simply by selecting one source and one destination :
And that’s it, once your flow is created, all you have to do is click Save and Run, and you Google Sheet data will magically turn into a BigQuery table in no time !
While Portable promises schematising the data for you, the results I got from the Google Sheet were not perfectly modelled, so another external Transform layer must be applied to the data. Hopefully Portable will add the option to edit the schema, alongside some other options like choosing a name for the tables. But even at this early stage, Portable delivers massive value.
For the free version, you’re only allowed manual execution of the flows, which is pretty great. If you need to automatise your flows, you can opt for one of the paid plans:
The Portable team also offers to build connectors on-demand for data teams, so if your connector is not on the list, you can request it here : https://portable.io/request
Final verdict : Turns out Portable is as good as “advertised”, all it takes is a couple of minutes to connect different applications to your data warehosue and extract the data. So if you’re looking to spend less time setting up extracting/loading pipelines and more time deriving value from your data, Portable is definitely worth considering.
Their plan is to build up to 10,000 connectors, I can’t wait to see that happen !
