Introduction
Hi, I am Akira, the editor-in-chief of Data Without Code. In our previous tutorial, we explored how to extract data from a public website using web scraping techniques. You are now officially gathering data from the outside world.
But let’s look back inside your own company. How do you currently get your official sales data or customer records? Do you have to email the IT department and say: “Hi, can you please run a SQL query and send me the CSV export for last month?”
If you rely on another person to generate your raw data, your automation pipeline is broken from the start. As a DX manager, my goal is to give business users direct access to the data they need.
In this Automation Hack, I will show you how to connect KNIME directly to internal SQL databases (like MySQL or PostgreSQL) so you can pull the data yourself—without writing a single line of SQL code.
Step 1: The DB Connector Node
To connect to a database, you need three pieces of information from your IT team: the Server Address (Host), the Database Name, and your personal Username/Password. (Tell them you need “Read-Only” access!).
Once you have those, open your KNIME workspace. Search your Node Repository for the MySQL Connector or the PostgreSQL Connector node (depending on what your company uses) and drag it to your canvas.
Double-click the node to open the configuration.
- Enter the Hostname (e.g.,
database.yourcompany.com). - Enter the Database Name.
- Under the “Authentication” section, enter your Username and Password.
Click OK and execute (press F8). If the traffic light turns green, congratulations! You have successfully established a secure connection to your company’s core database. No SQL coding required.
Step 2: Select Your Table (DB Table Selector)
Notice that the output port of the Connector node is a solid red square. This means it is a database connection, not a standard KNIME data table.
To tell the database which specific table you want to look at, we use the DB Table Selector node. Connect the red square output from your Connector node to the red square input of the DB Table Selector.
Double-click the DB Table Selector. You will see a button that says “Select Table…”. Click it, and KNIME will display a visual list of every table inside your company’s database! Simply click the one you want (e.g., sales_transactions_2026) and click OK.
You have just written a SQL SELECT * FROM query using only your mouse.
Step 3: Pull the Data into KNIME (DB Reader)
Right now, your data is still sitting inside the database server. To actually bring it into your local computer so you can clean it, you need to “read” it.
Add the DB Reader node to your canvas. Connect the red output of the DB Table Selector to the DB Reader.
Double-click the DB Reader, click OK, and execute it. KNIME will now pull all the rows from that SQL table directly into your workspace. Notice that the output port on the right side of the DB Reader is a black triangle. Your database records have officially been converted into a standard KNIME data table!
The True Power of DB Automation
Why is this so powerful? Because you can now connect this DB Reader node directly into your standard cleaning pipeline. You can use a Row Filter node to drop bad records, or a Joiner node to blend this internal database data with an external Excel file.
Next week, when you need the updated report, you don’t email IT. You just open your KNIME workflow and click “Execute All”. The DB Reader will instantly fetch the live, up-to-the-second data from the server.
Conclusion: Your Next Steps
Connecting directly to SQL databases is a massive milestone in any data professional’s career. You have bypassed the manual CSV export process and tapped directly into the source of truth.
Now that you have direct access to massive CRM databases, you will quickly notice something: Human beings make terrible data entry mistakes. You will see misspelled names, weird phone numbers, and duplicate accounts.
If you are about to upload clean data back into Salesforce or Hubspot, you need to scrub it first. Join me in our next tutorial where we cover the ultimate real-world scenario: Using KNIME to clean up messy CRM data before uploading.
