Database Explorer
1. Database Explorer
Hello, I am Francisco Kattan, head of our Builder Education Program in our Developer Relations and Community Team, and I'll be your instructor for some of the lessons. When I started my first data job, I inherited a Snowflake account that had been around for a few years. There were dozens of databases, hundreds of tables, and approximately zero documentation. My story is not unique in the world of data. My approach for the first two weeks was just to write `SELECT * FROM ...` queries, aim at various tables, and hope something looked familiar. It worked, but it was extremely frustrating. What I wish I had then was a tool that I'm going to show you now. Snowflake has a whole tool built specifically for the "I just want to look around" problem, and it's called the **Database Explorer** — and it's infinitely better than `SELECT *` for hours a day. Let me show you what it does. You can find the Database Explorer by going to **Catalog** in the left navigation menu, and then selecting **Database Explorer**. What you see here is a hierarchical view of everything in your Snowflake account that your current role has access to: databases at the top level, schemas inside each database, and then all of the objects inside each schema organized by type — tables, views, dynamic tables, stages, functions, and more. Think of it as a file browser for your data. You're not writing any code here — you're just clicking around, looking at things to help you plan for your queries later. Let's click into our `tasty_bytes` database, and then the `raw_pos` schema. You can see all the object types listed here. Let's click on Tables to expand that section. There are all our tables listed. Now let's click on the `order_header` table. This is the table detail view. At the top, you can see basic metadata: when it was created, how large it is, and which role owns it. Below that, under the **Columns** tab, you can see every column in the table, its data type, and any comments that have been added to describe what that column contains. This is genuinely useful when you're working with an unfamiliar table and you want to understand what you're dealing with before writing a single line of SQL. Now let's click on **Data Preview**. Snowflake loads up to 100 rows of the table right here in the UI, no query needed. You can sort by any column, adjust decimal precision on numeric columns, and get a real sense of what the data looks like. One thing to know is that previewing data does use a warehouse, so it will briefly spin up your default warehouse if it isn't already running, and this will incur at least some cost. Views work the same way. Click on a view and you get the same detail page, including a **Definition** tab that shows the SQL behind the view. Very useful if you're inheriting someone else's work and need to understand what a view is actually doing under the hood. Now I want to show you one more thing that's been added more recently and is extremely powerful: the **Lineage** tab. When you select a table or view and click Lineage, Snowflake draws a graph showing where that object gets its data from and what downstream objects depend on it. Objects on the left are upstream, meaning they feed data into this table. Objects on the right are downstream, meaning they consume data from this table. You can click on the plus icon on either side to expand further and see connections multiple hops away. If you click on the arrow connecting two objects, Snowflake will show you the SQL statement that created that dependency. If a downstream table was built from this one using a `CTAS` statement or a dynamic table definition, you can see exactly what the transformation was. This is a genuinely big deal for data engineering work. If something goes wrong upstream, you can immediately see what downstream objects are affected. If someone asks where a column in a report came from, you can trace it back through every transformation to the source. One last thing worth pointing out: from a schema in the Database Explorer, you can also create new objects directly using the **Create** button. Snowflake opens a worksheet with a template SQL statement for whichever object type you choose — tables, views, dynamic tables, stages, and more. It doesn't do the work for you, but it gives you a solid starting point. To recap, the Database Explorer lives under **Catalog** in the left navigation menu. It gives you a hierarchical view of all your databases, schemas, and objects. You can inspect column definitions, preview data without writing a query, view the SQL behind a view, trace data lineage forwards and backwards through your pipelines, and kick off object creation from a schema. It's one of the most practical tools in Snowsight for day-to-day work, and something I genuinely wish had existed from my first day.2. Let's practice!
Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.