SQL Basics in Snowflake — SELECT, FILTER, GROUP BY

A story-based introduction for modern Snowflake users

Imagine Snowflake as a giant data library.
Inside are millions of books (tables), shelves (schemas), and rooms (databases).
SQL is your magic spell that helps you quickly find, filter, and summarize the exact information you need.

Let’s explore the essentials using simple, real-world examples.

---

📄 1. SELECT — Choosing Your Columns

SELECT is like saying:
"I want to read these specific columns from my table."

Syntax

SELECT column1, column2
FROM schema.table;

Example

SELECT first_name, last_name, email
FROM SALES_DB.CUSTOMERS;

Story Example: You’re a store manager looking at just customer names and emails instead of the whole customer file. That’s SELECT.

---

🔍 2. WHERE — Filtering Rows

WHERE lets you filter only the rows you want.

Syntax

SELECT column1, column2
FROM schema.table
WHERE condition;

Example

SELECT first_name, last_name, total_purchase
FROM SALES_DB.CUSTOMERS
WHERE total_purchase > 500;

Story Example: You want VIP customers only. WHERE is like a filter in your workflow — showing only the important people.

---

🎯 3. FILTER — Using Conditional Logic

FILTER works mostly with aggregates to refine calculations.

Example

SELECT country, COUNT(*) AS total_customers
FROM SALES_DB.CUSTOMERS
GROUP BY country
HAVING COUNT(*) > 50;

• HAVING is like a filter after grouping.
• Only countries with more than 50 customers are displayed.

Story Example: You’re summarizing sales per country but only want to show active markets, ignoring tiny countries.

---

🧮 4. GROUP BY — Summarizing Data

GROUP BY groups rows that have the same value in one or more columns.

Syntax

SELECT column1, AGG_FUNCTION(column2)
FROM schema.table
GROUP BY column1;

Example

SELECT country, SUM(total_purchase) AS total_sales
FROM SALES_DB.CUSTOMERS
GROUP BY country;

Story Example: Think of it like counting total sales per country. GROUP BY helps you summarize big data into meaningful insights.

---

⚡ 5. Combining SELECT, WHERE, GROUP BY

You can combine all three for powerful insights.

Example

SELECT country, COUNT(*) AS vip_customers, SUM(total_purchase) AS total_sales
FROM SALES_DB.CUSTOMERS
WHERE total_purchase > 500
GROUP BY country
ORDER BY total_sales DESC;

Explanation:

• WHERE → filters VIP customers
• GROUP BY → groups them by country
• SELECT → chooses the columns to show
• ORDER BY → ranks countries by total sales

Story Example: You’re a regional manager checking where your VIP customers are spending the most. Done in one simple query.

---

📝 6. Tips for SQL in Snowflake

1. Always start small: Use LIMIT 10 while testing queries.

2. Use aliases: Makes results easy to read.

   SELECT country AS region, SUM(total_purchase) AS revenue
   FROM SALES_DB.CUSTOMERS
   GROUP BY country;

3. Use CTEs for clarity: WITH clauses make complex queries readable.

4. Use consistent naming: Snowflake is case-insensitive by default, but best to be consistent.

5. Preview before aggregating: SELECT * FROM table LIMIT 5 can save compute and time.

---

🎯 7. One-Sentence Summary

SELECT chooses your columns, WHERE filters rows, GROUP BY summarizes, and FILTER/HAVING refines aggregates — these are the building blocks of SQL in Snowflake.

---

🚀 Next Step

Internal Stages, Table Stages, User Stages — Deep Explanation